VGEGIS%20Training%20Tutorial

3DGIS TRAINING TUTORIAL

State Key Laboratory of Information Engineering in Surveying , Mapping and

Remote Sensing (LIESMARS)（Wuhan university）

APRIL , 2005

CATALOG CHAPTER 1 3DGIS SOFTWARE..................................................................................................3

§1.1 Main modeling software................................................................................................3 §1.2 Browsing software .........................................................................................................5 §1.3 Database builder............................................................................................................5 §1.4 Flow Process Diagram..................................................................................................5 §1.5 About the training tutorial..............................................................................................6

CHAPTER 2 THE WHOLE FLOW OF 3D CITY MODEL DATA PRODUCTION .........................................8 CHAPTER 3 DATA PRODUCTION PLAN ..........................................................................................10

§3.1 Measuring region division...........................................................................................10 §3.2 Attribute structure table design ..................................................................................10 §3.3 Userid coding ............................................................................................................... 11

CHAPTER 4 DATA PREPROCESSING ...............................................................................................12 §4.1 Digital elevation model data processing...................................................................12 §4.2 Digital orthophotoimage data processing.................................................................13 §4.3 field survey and texture collection and producing ..................................................15 §4.4 3DS model production ................................................................................................25

CHAPTER 5 THE MODELING KERNEL OF 3D VISUALIZATION MODEL DATA ................................29 §5.1 The modeling of geometry model..............................................................................29 §5.2 3D model edit tool in 3DCM .......................................................................................35

CHAPTER 6 THE GENERAL PROCESS OF 3D MODELING ..............................................................42 §6.1 Transfer Data and Build Model ..................................................................................42 §6.2 Modeling Important UserID ........................................................................................54 §6.3 Map Texture..................................................................................................................63 §6.4 Inserting 3DS Model....................................................................................................65 §6.5 Assemble Object ........................................................................................................65 §6.6 Input attribute and multimedia data...........................................................................66 §6.7 Merge sub-workspace .....................................................................................................68 §6.8 Data quality check ..........................................................................................................69

CHAPTER 7 BUILDING DATABASE ................................................................................................73 §7.1 building database with DEM data ..................................................................................73 §7.2 creating digital orthographic image database ................................................................76 §7.3 Creating database of 3D visualization model ...........................................................77

CHAPTER 8 FREQUENTLY QUESTION ANSWER .............................................................................81 §8.1 how to adjust multi neighboring surface node at the same time ? ..................................81

CHAPTER 9 APPENDIX .................................................................................................................81

2

Chapter 1 3DGIS software

§1.1 Main modeling software

3DCM (the data producing personnel should be specially skilled) It is a modular integrated with modeling and model editing (geometry editing, texture editing and attribute editing). It can accept different sources of data to build complete 3D surface models. And it can edit arbitrary 3D objects including changing geometry shape, associating different attributes and textures and so on. The modeling and editing results will ultimately be stored in database with appointed model and structure. Next this tutorial will introduce four toolbar of 3DCM.You can model conveniently by using them. Standard Toolbar

It provides the following commands in Standard Toolbar; they can accomplish general operations as follows:

New Create a new workspace. Open Open a workspace.

Select an userid as current userid.

Query Object Query or edit attribute of the current selected object. Modify UserID Modify the userid of the objects. Compose Compose many objects to one object. Decompose Decompose one object into many objects. Delete Delete an object of workspace. Copy Copy the selected objects. Paste Paste the copied objects. Translate Translate the whole scene. Rotate Rotate the whole scene. Scale Scale the whole scene. Update Update the displaying content of the window. About Display the version of 3DCM.

NodeEdit toolbar

It provides the following commands in NodeEdit toolbar, they can accomplish operations

3

including moving object, rotating, scale, moving node, linking, merging, deleting roof point, adding and changing roof triangle net and so on.

Select Select edit object (including object and node, press [CTRL] to select additionally or inversely). You can select objects by clicking them or dragging selected box.

Modify Object Position Set object's coordinate by dialog. Move Move objects or nodes. Rotate Rotate objects. Scale Scale objects. Edit Move Unit Set the move unit of moving, rotating and scaling of the

object. Change Display mode Change between global view and edit view. Set Scene Center Set the current selected object as scene center. Merge Node Move one node to the position of another node. Link Node Link the two nodes of the same object. Current X axis Set X-axis as current axis (the reference when edit) Current Y axis Set Y-axis as current axis (the reference when edit) Current Z axis Set Z-axis as current axis (the reference when edit) Delete Roof Point Delete current selected roof feature point (delete one

point once). Add Roof Point Add a roof feature point. Change Roof Triangle Net Interchange triangle connection relation of a polygon. Undo Undo (at the state of editing). Redo Redo (at the state of editing).

upper lod object Switch to the upper LOD object of the current object.

lower lod object Switch to the lower LOD object of the current object.

Volume Check toolbar

Check Volume Check the mosaic relation between model geometry

structure and DOM. On this basis edit and modify the inconsistence directly.

Zoom out Zoom out 2D image. Zoom in Zoom in by clicking or drawing a box. All Display the whole figures. Find first Find the first house of one userid. Find prev Find the previous house of the current house. Find next Find the next house of the current house.

4

Current Edit Object Select an object and set it as the current edit object. Coplanarity Check whether points are coplanarity or not. Line Parallel Check whether lines are parallel or not. Line Vertical Check whether lines are vertical or not. Orthogonal Do orthogonal process for the roof edge. Same Elev Check that points are the same elevation or not. Colline Check whether points are colline or not. Compose Surface Compose several facets of the arc surface into a

complicated surface in order to map conveniently. Auto Check Coheight Check coheight automatically for the current selected

object. Auto Range Acquire the texture coordinate range automatically. Box Range Decide the roof texture coordinate range by box. Auto Link Link roof texture automatically.

Object Modify toolbar

Update Texture Update BMP or 3DS file which is linked to point object. It also can update texture file which is linked to area object.

Modify Color Modify color of line object. Texture Format Brush Apply texture format (including texture file name) of a face

to other faces.

§1.2 Browsing software 3Dviewer (it should be mastered by data producing and checking personnel). It includes file and database operation, data management, visualized browsing, cartoon, querying and analyzing. The main character is the management of magnanimous data, large scope of rambling and visualizing, 3D model analyzing and appliance.

§1.3 Database builder 3DDB (It should be mastered by part of data producing and examining personnel). It is a modular to build database with scattered data provided by users (including DEM, DOM, 3D model data) and to distill and distribute it. It can manage file system and database system.

§1.4 Flow Process Diagram This diagram shows the relationship of three sofeware and modeling flow process.

5

§1.5 About the training tutorial The content of data production in 3D city model is 3D Model Data (3DMD) including Digital Elevation Model (DEM), Digital Orthoimage Map (DOM) and objects (artificial and natural objects). DEM and DOM should be produced according to the traditional and mature criterion. 3D city model data production based on aerial remote sensing and 2DGIS should include three aspects: aerial survey 3D geometry model building, field texture collecting and building, terrain sight building and CAD model building for parts of objects.

3D city model building based on aerial survey mainly adopts JX4A workstation and 3D model builder 3DCM. Field texture collection relates to the usage of digital camera and image processing software Photoshop and so on. Terrain sight building should adopt 3D model builder 3DCM. CAD model building for parts of objects should adopt model builder 3DMax.

3D city model is different from 3DMax cartoon scene, virtual imitation scene, and is especially different from 3D game scene. So if we want to produce regular data product, something special should be mastered well in actual work.

6

The tutorial will guide you how to realize 3D GIS in proper sequence according to data producing process. It will provide you examples to practice so that you can master the main functions and skills of 3D GIS software in short time.

We hope that readers can master the essence of the software and the thought and flow of digital 3D city production as a whole. So the entire technical group can favorably accomplish “3D city model” data producing task.

7

Chapter 2 The whole flow of 3D city model data

production 3D GIS software is developed according to 3D GIS construction. The main character of 3DGIS software is that it can provide 3D auto-modeling from certain city’s 3D coding data, GIS data and CAD data etc, which are captured based on digital photogrammetry workstation (DPW) such as JX-4A. It supports management of city’s mass data integrated for three databases (DEM+DOM+DLG) in wide scope and 3D real-time and seamless ramble. The software includes and extends regular functions in GIS such as multimedia query about information, expression, analysis and decision-making and so on. Several tools in 3D GIS software are used in 3D GIS data production. They complete different work in different 3D GIS building stages. These tools mainly includes:

◆ 3D GIS model tool 3DCM

◆ 3D GIS database builder 3DDB

◆ 3D GIS browsing tool 3Dviewer

The basic flow of 3D GIS data production can be divided to 6 parts: 1． Data preparation: it mainly produce Digital Elevation Model（DEM）data, aerial

orthophotoimage（DOM）data , measuring region division, 3DS model production, attribute structure diagram designing, userid coding and so on

2． 3D spatial points collection and 3D model building based on aerial stereopair. This

part of work is based on JX4 photogrammetry workstation, including two steps operated on observable objects (mainly is building) of stereopair, which are 3D feature point coding and collecting and 3D model creating and editing. The operated userids are mainly the userids （1）defined in 《criterion and flow》5.1.1 . The 3D surface models created in the stage are the foundation and basis of the latter work, and also the main part of measurable and analytical 3D city model.

3． Field survey and texture image collecting and producing. The 3D models collected

and created directly on stereopair only present the 3D geometry structure of partial city objects. In order to make the 3D scene realistic, we should adopt field survey to record the spatial and attribute information of every userid, in the meanwhile collect the texture image of relevant objects.

4． Building sight model producing. According to the relevant information and texture

image of buildings collected in the former stage, we can produce texture of every building profile, produce building CAD model incompletely measured in the aerial survey image, and input the building attribute information.

5． Terrain sight producing. The goal of terrain sight producing is to increase the

fidelity and the visualization effect of the scene. We should model and present the city ground object selectively. The related the objects include all other objects apart

8

from (1) defined in 《criterion and flow》5.1.1

6． Database building. We can build database for Digital Elevation Model（DEM）data , aerial orthophotoimage（DOM）data , 3D visualization model with 3DDB software. The goal is to meet the demand of 3D visualization application when using 3DGIS browsing tool 3Dviewer, so we can use the data in dynamic way without restrict of the data amount.

7． We can accomplish 3DGIS construction through the 6 steps above, and then we

can browse a live and realistic 3D city with 3Dviewer on PC.

9

Chapter 3 Data production plan

§3.1 Measuring region division 3D model production is carried through with the unit of sub-region. We should firstly ascertain the region division in the two producing stages, which are 3D model collection and terrain model production based on aerial survey. The region division of terrain model production can either use the division in aerial survey or new division, which depends on the needs of actual production. The region division depends on orthographic images and large-scale relief maps. The division should be according to the natural boundary such as roads and streets. The sub- regions should be joined seamlessly and the data amount should be controlled in a certain range. Generally the number of buildings in sub-region should not exceed 100. The region division is affirmed by producing department before the data is produced. The concrete boundary is recorded in large-scale relief maps and orthographic images (in the form of digital and paper). The sub-regions are named according to stereopair number in the aerial survey stage. The regulation is as follows: The stereopair number of partition + the surveyor number+ the serial number of surveyor measuring partition

To record and examine workspace data more conveniently, the name of workspace should be unique in the whole measuring region. According to the data producing experience in practice, the concrete name can be affirmed in the following way: ① Give different code to different surveyors. For example, identify different

surveyors with A, B. ② Divide the workspace according to the number of objects. ③ Encode the measuring region initially according to the workspace serial number

measured by surveyor. For example, the second block data of A surveyor is named A02.

④ Make workspace name unique with stereopair code. For example, the stereopair code of A02 region is 18731874, and then the complete name of workspace is 18731874A02.

After feature encoding data collection is done, DEM acts as the 3D model terrain surface stratum in the process of geometry model building. So we must acquire the checked and accepted DEM in the measuring region, or else the latter feature data collection cannot be processed. The data acquired in this part is explicit sub-regions and their numbers.

§3.2 Attribute structure table design After the 3D visualization data is produced, we need to combine the buildings and

10

record their attributes. The table in which the building attributes are recorded should be designed and constructed according to the attribute items. Data structure design -- All the attribute data must be set according to promissory data types and field length, or it can not process attribute data at the time of merging workspace and putting workspace to library. It mainly includes:

Building name, character type, 128 bytes Building layer numbers, int type, 2 bytes

Remark information, character type, 2048 bytes The data acquired in this part is completely designed attribute structure table.

§3.3 Userid coding When labors produce 3D visualization data in 3DCM, every userid should be endowed a userid code. There are many userids and they are disordered in ground scene producing, so if we don’t code the userids strictly before data producing, the ambiguous userids are hard to be ascertained their names and chosen in the window. In order to definitely give the userid code in data producing, we should code and sort all the userids strictly according to field survey before the data is produced. When we code the userids we can combine the userid coding in the manual and the region situation. The data acquired in the part is the strict userid code. It is the strict basis of coding every userid when labors produce 3D view model in 3DCM. The corresponding table of userid and userid code can be consulted in appendix 3.

11

Chapter 4 data preprocessing In part two we have planed the entire data production and acquired the definite sub-regions and their names, userid codes, attribute structure. After the entire data production has been done, we should preprocess relevant data. The key of data preprocessing will be introduced in this chapter. The data acquired in constructing 3D GIS includes Digital Elevation Model（DEM）data, aerial orthophotoimage（DOM）data and 3D visualization model data (CGS). But the provided source data may not accord with these formats. We should preprocess data with relevant software.

§4.1 Digital elevation model data processing DEM data is the basic data describing terrain surface fluctuant in 3D spatial database. In the 3D spatial data dynamic browsing, DEM is firstly loaded and displayed. Generally the original DEM is single breadth. But the DEM data to which a region is related may be single breadth or multi-breadth. If it relates to multiple breadths of DEM, we should splice and clip them according to the coordinates of region range. And if it only relates to a single breadth of DEM, we need to clip it according to the coordinates of region range. Read terrain data from DXF file of version 12.0 or 14.0 to interpolate on DEM, we can accomplish the conversion of contour line and discrete point data to irregular triangle

12

§4.2 Digital orthophotoimage data processing §4.2.1 image correct Orthophotoimage can display terrain character and terrain surface covering situation realistically, which has gradually become a new trend. But the original orthophotoimage will be geometry transformed when it is acquired and its accuracy may not meet our demand. So we need to correct the image. When we correct the selected DOM, we need control points file whose name is as same as DOM. Its extended name is “.ctl” and it is below the same directory of DOM. We need at least three control points to correct the image. Steps:

1. Execute 3DCM.exe to enter the system interface. 2. Click [File│ Image Correcting】and we will see a pop-up “open” dialog .Select the

image file that needs correcting. 3. After the correcting is done, we will see the match prompting information. 4. The system will create correct image in the same directory of the original image.

The file name is: original file name +“_new”.

File format of control points: Control points number: N point 1: 1 X1 Y1 x1 y1 z1 point 2: 2 X2 Y2 x2 y2 z2 ... point n: n Xn Yn xn yn zn Where, Xi, Yi is coordinate of point image, xi, yi, zi is geodetical coordinate of the control points. §4.2.2 locate image Digital orthophotoimage and corresponding digital elevation model should be used at the same time, and the four vertex coordinates of them should be completely identical. Or else there would be errors when we overlap DOM and DEM. The DOM data relates to a region may be single breadth or multi-breadth, which is similar to DEM. If the DOM data is multi-breadth, we should splice and clip them according to the region range coordinates with Photoshop. If the data is single breadth, we should clip it according to the region range coordinates. We should clip the DOM data according to the four vertex coordinates of the completed DEM data in this region. We should write the locating file. Locate orthophotoimage with coordinates and save the result as a DOM file. The menu is below the initial interface.

13

Steps: 1. Click [File│ locate image】menu item, we will see a pop-up “Geocoding orthoimage”

dialog. Click [Browse】button to select Bmp file, and set the locating parameter:

Image Coordinate[col, row]: the colum number and row number of locating

point in the image. Geocoordinate[x,y]: the x,y coordinate of locating point.

The relation between the coordinates of the image points and the terrain points is as follows:

14

2. After the locating information is inputted, click ，the system

can calculate terrain image resolution according to spatial locating information. 3. After the configuration is done, click [OK】, the system will create homonymic dom

locating file in the directory of Bmp file automatically.

§4.3 field survey and texture collection and producing In common situation, if we want to accomplish 3D modeling well, the cooperation of many softwares is needed. In the view of 3D model collection based on photogrammetry, the 3D model data production tasks can be divided into four aspects according to their working sort. The four aspects are spatial object feature points collecting, 3D modeling, image processing and spatial model creating, and they are related to software in different types. The spatial userid feature points collecting can adopt mature photogrammetry workstation such as JX4. If there are no special declarations in the following contents, the photogrammetry workstations which we referred to are all JX4 photogrammetry workstations, and they are written as JX4 for short. Image processing can adopt commercial image processing software such as Photoshop; creating special model can adopt the popular software in current international market such as 3DMAX. §4.3.1 taking photo of texture and field survey 3D geometry model building complete partial work of 3D view model; it is the foundation of 3D view model measurability. The aim to build 3D demo system is to present the terrain real view as completely as possible. The work of field survey and taking object image, on the one hand is getting the distribution information of terrain object in different types, on the other hand is collecting object photos which are able to present view texture details. Considering the efficiency and connection of actual work, this stage’s work can be divided into two parts: building profile texture collection and terrain sight survey. §4.3.1.1 The usage of digital camera Digital camera is the main data collecting device in the stage of work. The performance of digital camera and operation methods directly affect texture photo quality. In order to meet the demand of texture image collection, digital camera needs the following conditions:

① At least 3 million pixels; ② Can take photos in different resolutions; ③ The focus can be adjusted;

15

④ The figure change in the photo is tiny; ⑤ Real color presentation; ⑥ At least 16M memory, assistant storage device is recommended to save more

photos; ⑦ High quality power supply;

We should get familiar with camera performance and master the view-find skill when we use the camera. When we take the texture image, we should pay attention to the followings items:

① Get familiar with the camera “white balance” parameters and set them. White balance is a vital index of digital camera, which can ultimately decide the hue of all the photos taken by camera.

② Set the correct resolution. According to the basic requirements of texture management, the surfaces of key object and the main surfaces of common object which confront the street and road should be mapped with higher resolution image; common image should use lower resolution. ③ Use light condition correctly. The nature and balance of 3D hue is the important aspect of visualized effect presentation. Because the digital camera totally depends on light, light which is too strong or too weak will severely affect photo quality. ④ Select good shooting angle. Because we need to correct the image into orthograph in the latter image processing, the image quality will be severely damaged in the image correction if the original shooting angel is too great. ⑤ Diminish the impact of irrelevant objects. Eliminating the irrelevant objects will affect image quality and add workload. So we need to select the angle and position to avoid irrelevant objects. ⑥ Select good weather condition. On the one hand, fine weather can provide good light conditions; on the other hand, natural object shade can present 3D view visual effect realistically. ⑧ Bring spare accessory. Bringing spare accessory can promote efficiency, prevent

the halt of taking photo for the reason that battery runs out or storage space is not enough.

§4.3.1.2 Collect buildings texture and terrain sight objects texture ■Collect building roof texture： There are two ways of acquiring building roof texture: one way is clipping from aerial image; the other way is replacing it with colorized material. 1、Building roof texture is generally clipped from aerial image. We can clip aerial image through artificial way, and adopt external image software such as Photoshop to process orthographic image, and connect building surface with texture in general texture mapping operation. For the building roof texture collection have the character of its own, 3DCM provides a way to automatically combine manual operation to the type of texture collection, so that we can operate the roof texture in high efficiency. The concrete operation is as following:

16

① Select all the objects whose userid is BODY; ② Get through every BODY object; ③ According to the geometry shape and texture scope of building roof, we

use[BOX】command of texture collection to define the texture scope of building top.

④ Use auto association to accomplish the operations of clipping, naming, association, inputting into database, etc.

2、Replace building roof texture with colorized material. Generally we replace high building roof texture with colorized material. The main reason is that the real building roof texture can’t be acquired by manual photos. The top texture clipped from aerial image is low in resolution and can’t realistically present the effect of building top sight, meanwhile influence the effect of whole building sight. So the top textures of flat roof high buildings are usually replaced with colorized material. If the building roof is complex, we can realize it with 3DS modeling method. ■Building profile texture collection Building profile texture collection refers to the following aspects:

① The preparation work of texture collection and attention items in the screening should strictly execute the prescribed content in the criterion. ② When we take photos of buildings, we can screen in the unit of buildings and take photos of all the profiles of the building in order at one time; we can also take the view point as center and take photos of building profiles around it. ③ We can take more than one photo to a profile from different aspects. Apart from detail image shoot, we need to take photo of the building panorama to grasp its whole structure. ④ Record the corresponding relationship between building code and camera photo code according to definite criterion and flow.

§4.3.1.3 Building attribute data collection and recording. The operation of building attribute data collecting and recording should follow the steps:

① Attribute data collection and taking photo of building profile texture should be proceeded at the same time. The record of attribute data and texture data use the same table. ② The collection content and concrete requirement of concrete attribute item can be affirmed by definite criterion and flow.

§4.3.1.4 Every types of terrain userid survey Apart from buildings, every types of terrain userid survey collect terrain userid information including roads, vegetation, public device, etc. The concrete content relates to type and distribution information of different userids defined in 《criterion and flow》5.1. In field survey we should strictly record the userids in 3DS model, and hand them to 3DS model producing staffs to produce models. The field survey of every userid includes character explanation and texture image

17

recording. Concrete content should be filled in the field survey table of《criterion and flow》

appendix 3. Field survey character explanation mainly records userid types and distribution explanation; texture image on the one hand provides userid texture information, on the other hand provides relative relation among different userids. §4.3.2 produce texture data The creating of 3D city view is related to the texture image collecting and processing in large quantity. The original photo of texture image is taken in the field. For the disagreement in the form, size and angle with 3D model corresponding surface, the original texture photo must be processed by relevant image processing software. Considering the availability and convenience of software operation, Photoshop is widely chosen as image processing software. The mature Chinese edition currently is Photoshop 5.0 or 6.1. The Chinese edition currently we used is Photoshop 7.01. Because this edition can set texture channels to make texture-processing effect better, and we can conveniently create model. The image processing functions such as clipping, orthographic correction, splice, removing irrelevant objects, hue adjustment etc. are included in Photoshop. In order to promote efficiency in image processing, we can choose some practical effective image processing tools, and we can define a series of operations to realize operations in image hue, size and naming in batches. 1、 Produce texture image The processing of texture image mainly operates with image processing software. If 3DCM model or 3DS model surface need to correspond to original photos taken in the field and to be mapped with them, the processing content of texture image should include:

① Image clipping: clip the original image according to the scope of 3D model surface.

② Image correction: proceed the orthographic correction to the inclined surface in the level direction.

③ Sundries elimination: get rid of irrelevant objects in the image such as trees and passengers.

④ Image splicing: splice different image pieces into integral texture image. ⑤ Hue adjustment: If the image is gloom or image hue doesn’t harmonize with

the integral scene, we should adjust the image hue. ⑥ Size adjustment: frontal texture image size of building generally doesn’t

exceed 512*512 pixels; we can properly diminish the texture size to profile images.

⑦ Naming and preservation: name and preserve the image according to the advance regulated requirements.

The image process operations above are generally accomplished with Photoshop. Image processing requires other software, including CorelDraw software when producing zebra crossing. Describe zebra crossing with curves in CorelDraw, and then overlap it with texture image of road surface, export it as BMP texture image to produce zebra crossing. When we process image operation in batches such as hue adjustment in batches, image size adjustment, image naming and so on, we can use some professional image process

18

software. 2. Produce transparent texture 1. Open the original photo in Photoshop.

2. Select magic stick tool to choose and clip the outer range of the tree.

3． Select magic stick tool to choose the outer range of the tree.

4． Click[Select| Reverse select】 5． Click[Select| modify| shrink】, one or two pixel. 6． Add tunnel in navigation map.

19

7. Select “save to ” command, pay attention to the red mark: select Alpha tunnel. Then the transparent texture will be produced successfully.

3. The producing method of Herringbone profile texture The texture processing in special situation mainly refers to the Herringbone profile texture processing shown as the right picture. The herringbone profile is a familiar building profile. But right now 3DCM hasn’t provided any special texture mapping methods to herringbone profile texture. If we follow the common method to map the two faces separately, the .bmp texture corrected in Photoshop will be distorted on the corresponding model, regardless of which mapping methods we adopt. To avoid this situation, the Herringbone profile texture producing method should follow a certain flow:

① Select the building that you want to modify in 3DCM. Click model checking,

select two faces of a Herringbone profile texture and click to combine it into a face.

20

② Produce the texture image of the combining profile and map the texture. It may appear two exceptional situations such as picture A and picture B. In picture A the eave is eliminated, and in picture B the sky and tree are also mapped in the profile texture. The main reason is that the proportion of the triangle part in the upper side and the rectangle part from the underside in the building texture doesn’t accord with the fact. And we should adjust the scale of the triangle part and the rectangle part in the texture.

Picture A Picture B picture C We can solve the problem in this way. Divide the triangle part and rectangle part of the texture with gauge tool in Photoshop. And apply the height proportion of the triangle part and rectangle part to the texture. Use transform/distort tool in Photoshop to adjust the proportion triangle part and the rectangle part. Then map the texture again. We can see the texture mapping is correct in picture C. §4.3.3 Produce manual texture Generally image data is absolutely necessary in real reappearance of terrain 3D view. But in some cases we are unable or hard to get image data, manual texture can also get good result. The appendix provides a primary manual texture production flow. It is remarkable that we need the cooperation of multiple software to get beautiful manual texture production flow. Because different users have different understanding and demand in the texture, the content in the appendix is mainly thought and method rather than all-purpose tool.

21

Manual texture production flow is as follows: 1. Open DEM data and divide it in altitude. The two sides of 0m contour must be set in different colors. Output hypermetric tinting. Input hypermetric tinting in files in

CorelDraw10.0. Process steps: select slice tool, and divide ocean and land along

0m contour. We can also process it with Photoshop. Use to select land part and clip

it, leaving only ocean part. 2. Use 3DViewer to open data, and output orthographic romance image. The concrete operation steps can be referred in manual. 3. Input orthographic romance image in CorelDraw10.0, and then put bmp image of ocean part above the orthographic romance image. Output 0m isoline and import it into CorelDraw10.0 to overlap with 0m contour of ocean part. Use Beziers curve tool to close 0m contour along the scope of orthographic romance image in the bottom layer. Get rid of frame with tool.

4. Use Fill Flyout toolbar as follows:

Select ‘Texture Fill’, and we can see a pop-up dialog:

22

① In ‘texture library’ we can select different texture designs. ② In ‘Style Name’ we can set different parameters to attain definite result. ③ Click ‘Options’，and we can see a pop- up dialog：

We can set the resolution of fill texture bitmap. The unit is the maximum value of texture image pixel. ④ Click ‘Tiling’，and we can see the pop-up dialog：

23

‘Origin’ is the coordinate of center point. ‘Size’ is the size of meta-bitmap. Transform is the transformation of Meta bitmap, such as the parameter of

slope and rotation. Row or column offset is the overlap parameter between meta bitmaps. Transform fill with object: bitmap transforms with the change of object Mirror Fill: Mirror Fill

6. Use transparency tool(as follows) to set the filled texture as transplant color.

7. Overlap the userid such as river, house, farm, etc in the picture to complete the manual

texture production. The example is as follows:

Example of manual texture production

24

§4.4 3DS model production For the reason that the geometry structures are complex or for the reason of object dimension, some objects related to 3D city view cannot be seen in the aerial survey stereopair, and it is hard to collect feature points in common aerial survey methods. On the other hand, we cannot build model with 3DCM because the object structure is complex. So firstly we need to build 3D model with external modeling tool, then insert complete 3D model directly into 3D scene through the import function of 3DCM. 3DCM supports the 3D model in 3DS format. Generally we use 3DMAX model building software to build 3DS model, then output 3DS format data though data export function after the model is built. In order to realize the support of 3DCM to 3DS model, we choose the edition of 3DMAX 3.0, 4.0 or 5.0. 3DS model production should strictly obey relevant regulation criterion.Owing to producing methods, 3DS model has large number of triangles, so 3DS model production should pay special attention to presenting the form with few triangles. The other aspect of 3DS model production is texture mapping data management. The demand of texture naming and format in 3DS model is strict. Some 3DS point models have indicative functions, such as street nameplate, traffic sign, etc. So the character and picture must be clear. §4.4.1 3ds modeling criterion When building 3ds model, user must unify both system coordinates and Cartesian coordinates system. The origin of coordinates is not necessary required. User only need to insure that the unit length and directions of the coordinates system must be consistent with the original data provided. For example, coordinates of a building's bottom center point is（25542.1, 422345.3, 24.4）, and its length, width, height are 20,10,4. User only need to insure the unification of length, width, height and their directions of the model . Texture When creating 3ds models, the name of the textures must be limited to eight characters（including eight characters）,and put the texture file and 3ds model to the same directory, the texture file must use ‘bmp’ bitmap file. Preparing work Make sure the entire uniqueness of the point bmp file and model file. The entire uniqueness of the associated texture of the model file also should be taken into consideration. §4.4.2 3ds model inserting methods 1. Select [Create Model| point object】, then click left key in 2D view to locate the position where you want to add 3DS model. We can see a pop-up dialog as follows:

25

Parameters: UserID: UserID of point object which is added. Height, Width, Azimuth: these items are irrelevant to 3DS model. X，Y，Z: If you locate a position with left key in 2D view, it will automatically

enter the value. Link: the path of 3DS file.

2. Fill the parameters above and click OK. We can see the inserted 3DS model in 3D view. §4.4.3 The general 3DMAX modeling method. ① Flow The flow of 3D object production is generally as follows:

26

② The principle of 3D modeling a) We should emphasize the accuracy. We should get detailed data before work

and take photo of the object from each side. Then we can produce real and intact 3D model.

b) We should diminish the points and segments as much as possible in the premise of meeting the structure demand, which is a important modeling principle. If we don’t control the number of points and faces in the beginning of modeling, the latter processing will be difficult. The experienced personnel will reduce the file size and point and face number as much as possible. Putting limited points and faces in most important place will bring great convenience to latter work..

c) By the strong modeling function of 3DSMAX, the same object has several methods to create it. We should select an accurate and quick method to create it.

③ Modeling method a) We should begin with 2D modeling, then modify and loft adjust it to create

3D model. This is the traditional process of 3D modeling. b) Create model with 3D modeling tool in 3DSMAX creating panel directly. c) Use Bool operation in 3DSMAX.

2D lofting,3D modeling and Bool operation are three main modeling methods in building modeling. They are often integrated as a whole and used neatly.

§4.4.4 Attentions in 3DMAX modeling The objects presented with 3DS are generally complex, so we need large number of faces and points. Because the 3DS file format hasn’t been published, uncommon modeling methods may result that partial data can’t be completely and clearly interpreted and read. And so the model can’t be normally displayed in 3DCM and 3DGIS. So we should pay attention to the following items when creating 3DS model:

1、Please use the edition above 3DMAX3.0. The data format below 3.0 is outdated now and the exported 3DS file format can’t get support of other software.

2、The modeling personnel had better collect the field texture of 3DS model on his own, and record the height, size and position of the model.

3、If you need to use “clone“ function in the process of modeling, please use “copy” option rather that “instance”. The “Reference” and “Mirror” can’t be used either.

27

4、The current 3DCM and 3DGIS can’t support camera, sunlight, shadow, particle system, synthesized result, cartoon and environment.

5、You should set “Ambient”, “Diffuse” and “Specular” at the same time when setting texture. Try to make the three ponderances appear white.

6、The current 3DCM and 3DGIS only support the “Diffuse” texture-mapping tunnel. Other tunnels such as the texture of “Ambient” and “Bump” are null.

7、The current 3DCM and 3DGIS only support the name of Bitmap file (bmp file). The bmp filename can’t exceed 8 characters.

8、We should display the transparent effect such as glass in modeling progress. And we could change opacity, the transparent parameter in material. 100 means opaque and 0 means fully transparent.

9、When we complete a complex model, we should optimize the model with optimizing tools, and we can reduce the points and faces number at our best in the premise that the basic model shape isn’t lost.

10、If we need to modify 3DCM model or map it in the way of 3DMAX, we can export the model as 3DS model. Export the 3DS file with Deep Exploration and save it to 3DS model again. Then we can put it in 3DMAX to modify and map it. Attention, if a face in an element is mapped, the whole element should be mapped again. And the model imported to 3DCM is changed to 3DS model, so we can’t edit it in 3DCM edit window.

11、The default length, width and height in 3DMAX and the height value in 3DCM share the same unit. When the model is completed, we can create “Tape” in Helpers to measure the height. Before the model is exported, zoom the model and try your best to make it accord with the real object.

12、3DS texture and model naming: the name of bmp file and 3DS file should only be 8.3 format. In other words, the filename should be no more than 8 characters such as filename.bmp and filename.3DS. (Suggestion: the name is composed of the model name+ 2 bit of identification code+ the unique name of producer.)

13、To avoid that errors appear in 3DCM and 3DGIS, we should export model as 3DS file in 3DMAX, then use Deep Exploration to export the 3DS file and save it as another 3DS file.

14、If the model appears misplacing situation in 3DCM, we can “ATTACH” the whole model, then combine the model with “COLLAPSE ALL”.

28

Chapter 5 the Modeling Kernel of 3D Visualization

Model Data

§5.1 The modeling of geometry model The three-dimensional model data (*.asc) which is processed in 3DCM is get from photogrammetry workstation. Because of the restriction of the image, many objects can’t be distinguished on the image so they can’t be measured. In order to represent the 3D scene completely, 3DCM provides the functions of creating and editing for those objects. Take into account the desire of the landscape and character of objects now, 3DCM provides the functions of creating and editing for the point model, area model, volume model and other types of models. §5.1.1The modeling of point model in the scene The character of point model In 3D scene, point model refers to the object whose structure is simple and it is more isolated than other objects, such as trees, lamps, telephone booth. Point model is classified into two types, the first is that the objects’ form in each directions is the same, such as trees; the other is that objects’ body are small, but some parts are complicated, such as lamps. For the modeling of point model, the former is represented by the texture in one direction or two directions; the latter is represented by 3DS model. They can be imported to 3DCM by the same way.

The modeling of point model in the scene Point models are represented by isolated point model and regular point model in 3D scene. Isolated point model is point model which exists solely, regular point model is point model which is arranged by fixed routine and interval, such as trees and lamps. There are two types for the modeling of point model, one is represented by texture, such as trees, the other is represented by 3DS model, such as lamps, telephone booth, bridges, special buildings.

Trees

29

3DMAX model

Point models are represented by isolated point model and regular point model in 3D scene.

The modeling steps of isolated point model:

① Click [Display | Display 2D View] submenu; ② Click [CreateModel | Point Object] submenu; ③ Click mouse’s left key in the scene where you need to import the model; ④ You can select the userid, linked BMP image (texture) or 3DS model of point

object and related parameters in the pop-up “Point Object Setting” dialog. For the BMP image, you need to input the geometry size of texture, you needn’t consider the azimuth and coordinate; For 3DS model, its geometry size need to be set in modeling software such as MAX, you needn’t consider other parameters;

⑤ Click [OK] button.

The modeling steps of regular point model: ① Click [CreateModel | Line Object] submenu, and you can model for the line object

along the routine of point object.

This object is used in the process of modeling; it should not exist in the last data.

② Click [Edit | Convert Geometry Type | Line To Point] submenu; ③ You can set geometry attribute of point model in the “Point Object Setting” dialog,

you can select point object which you want to import by clicking “link” button.

There are two types for point object: one is BMP file, the other is 3DS model; ④ You can set the parameters of point object according to the parameters setting of

30

isolated model in the pop-up dialog; ⑤ Adjust the space position of model.

The height of the trees is confirmed according to the practice, on common circumstance, its height is about eight meters; its width is from three to five meters. For some special vegetation, such as shrub, their height and width are confirmed by the measurement.

Attention at the time of modeling for the tree singly or batchly: Avoid that the tree go into the building’s wall when you rotate the tree, the distance between the tree’s center point and buildings must be more than half of tree’s width.

§5.1.2The modeling of area model in the scene The character of area model

31

Area model is attached to terrain surface; it is represented by abstract area without thickness. Area model can be classified into two types according to the texture character, the first is the one whose texture is single and there isn’t special geometry character, such as lawn; the second is the one whose texture has special character, such as the road with zebra crossing. For the former area model, you only need to map the texture of lawn repeatedly after ensuring the boundary of area model; for the latter, you need to map different texture according to the texture character in different area model. The modeling of area model in the scene There are two methods for the modeling of area model: ⑴direct method horizontal

1. Click [Display 2D View | CreateModel | Surface Object] submenu, you will see the pop-up “Surface Object Setting” dialog; please refer to 3DCM manual about the meanings of the parameters.

2. You can draw the area by mouse's left key in the workspace after setting parameters and you can press mouse's right key to end. The system will add this area automatically in the workspace.

Vertical surface This selection will enable you to create vertical surface, such as billboard, fence and wall. Steps: 1. Click [CreateModel | Vertical Surface] submenu, you will see the pop-up “Create Vertical Surface” dialog. You can draw the area by mouse's left key in the workspace after setting parameters and you can press mouse's right key to end.

32

2. You will see a pop-up dialog to ask you to close or not, you can select “Yes” or “No” according to your desire. 3. Click [OK] button, the system will add this vertical surface automatically in the workspace.

⑵Indirect method Steps:

1. Draw a line; 2. Click [Edit | Dense Line Node] submenu; 3. Click [Edit | Convert Geometry Type] submenu to convert line to surface; 4. Click [Edit |Plugin | Interpolate DEM] submenu;

You’d better use this method for the modeling of area model. §5.1.3 The modeling of volume model in the scene The character of volume model Volume model is attached to terrain surface, it is the model that there is definite thickness in 3D model, it is often used to represent the area object with definite height on the ground, such as parterre, greenbelt, their flank texture must be mapped according to the practical texture. The modeling of volume model in the scene There are two methods for the modeling of volume model: ⑴direct method The direct method is that you can create volume model directly in 3D scene. Steps:

1.Click [CreateModel | Rectangle Volume] submenu or [CreateModel | Polygon Volume] submenu in 2D window, for the rectangle volume, you can confirm the range by drawing a box; for the polygon volume, you can confirm the range by mouse and press mouse’s right key to close the polygon.

33

2. You can set object’s height in the pop-up “Volume Height Setting” dialog. If the volume model that you create is on the ground closely, you need to click [Edit |

Plugin | Interpolate DEM] submenu to let volume model be on the ground, for other objects, such as buildings, this operation is not necessary. ⑵indirect method Steps:

1. Draw a line; 2. Click [Edit | Dense Line Node] submenu; 3. Click [Edit | Convert Geometry Type] submenu to convert line to volume; You’d better use this method for the modeling of volume model which is on the

ground.

§5.1.4 The differentiate principle between area model and volume model It needs to consider different types of factors that a certain object is represented by area model or volume model in the process of making data.

1. Parterre is represented by volume model because it is independent of ground; the road surface is represented by area model because it is on the ground closely.

2.The lawn must be represented by different object on different circumstances: If the hypsography is smart, the lawn is represented by area model because the browser has different displaying and processing mechanism for area and volume. But on other circumstances, such as the lawn besides the road, this must be represented by volume model and map texture in the flank of volume. You can't define the same name in the different types for the same object in 3DCM, so you need to consider defining the object with different userid name. Particularly, for userdef_userid.txt file, you can define meadow as "lawn" in #SURFACE field, but you can define meadow as "meadow" in #VOLUME field.

In order to avoid unusual situation at the time of browsing, when you create area and volume object, you had better avoid creating the model whose geometry size is too large, on common circumstances, the longest edge of area or volume that you create in the scene must be less than thirty meters. To reduce the data account of area model and volume model, you need to tick in “simplify” choice in the process of modeling. §5.1.5 Attention in the process of creating area model and volume model You must meet the following request in the process of creating these two models because of the resemblance between area model and volume model.

① You must confirm the range and type of area model and volume model according to the photo taken in the open air;

② Area model and volume model in 3D scene must overlay the ground completely, and the area model and volume model that they are close must adjoin seamlessly;

③ The edge of road should represent the shape of roadbed shoulder, on common circumstances; meadow is beside the roadbed shoulder. During the process

34

of creating meadow and roadbed shoulder, whether they are created separately or not is decided by the actual condition. If they are on the same area, they are created by a volume, if the roadbed shoulder protrudes the ground, it is created separately and represented by area.

④ If the area model or volume model is long and narrow, and it exceeds certain length, it should be cut to some models and created separately, or it will appear the unusual situation when displaying.

⑤ The texture of road surface must represent the pattern of zebra crossing and make use of the practical texture of road surface as the material. If the texture is at the corner of zebra crossing, it can be created by little area or imported to 3DCM after creating in 3Dmax.

§5.1.6 The modeling of other models This includes the modeling of sphere, cylinder and so on. Steps:

Click corresponding menu item of [CreateModel] menu in 2D edit window, the steps is similar as the above stated.

§5.2 3D model edit tool in 3DCM §5.2.1 Add Roof Point If there is difference between built buildings and actual buildings, the reason is that model is lack in one or more roof points, which can be added in 3D editor directly. Steps: 1.Select the building that requires to be edited; 2.Select “Add Roof Point” submenu, and then click mouse's left key in the position where you want to add a roof point. Then a roof point will be added in the appointed place. If you want to transfer a roof point, firstly select “Change Display Mode”, secondly select the roof point, then select “Move” tool and the reference axis, lastly drag left mouse's button left and right in view until the roof point is in the correct position. § 5.2.2 3D edit, for example: change roof triangle net and so on: If there is difference between roof shape and actual shape, the reason maybe that the direction of the roof triangle net's construction is wrong. It can be corrected by adjusting roof triangle net. Steps: 1. Select the building needed to edit; 2. Click “Change Display Mode”, and select two nodes which need to be adjusted (select

by dragging a box or clicking both nodes after pressing [CTRL] key); 3. Click “Change Roof Triangle Net” to adjust the link direction of the two nodes. §5.2.3 Point objects' correlated model (3ds,bmp), position inserted, size, angle rotated and so on If buildings are built by 3ds,the model will be inserted as a point object. When inserted, you need to:

35

1. Calculate coordinates of model's bottom center point; 2. Select “CreateModel | Point Object” menu in 2D window to find the position of bottom

center point of the model and click left mouse's button; 3. Click ”link” button in the pop-up dialog, and select “.3ds” (The default name is “.bmp”.)

in the file type list of the pop-up file selection dialog; 4. Then select a 3ds model that needs to be linked. The size of the 3ds model has been

fixed at the time of modeling, so user needn't to give the height and width of the point object. But he or she can set azimuth (i.e. rotate direction of model) and it can be modified in 3D window.

§5.2.4 Parts that cannot be finished by auto modeling, such as the fence of the cloverleaf junction and so on. If there is cloverleaf junction data in the imported data, users need to delete the bridge surface, fences at both entrances (Delete after selection). §5.2.5 Check Model: Check the mosaic relation between model's geometry structure and DOM. On the basis of it, edit and modify parts where are not consistent. If you want to make DOM as a background, please select corresponding menu to display DOM first. Types and steps of check Model check is classified into six types, i.e. points coplanarity, parallel line, vertical line, orthogonal, identical elevation, and collinear points. It can automatically clip and link the roof texture of a building and adjust lateral points of buildings. Preparing work Location of DOM (Print or display 2D vector data of this area). §5.2.6 Overlay DOM: Firstly select “Edit | Check Volume” menu and switch view to check volume view. Then select “Edit | Display DOM” menu to select corresponding DOM for this workspace.

§5.2.7 Assemble Object Assemble houses, which belong to a building but are disassembled to several volume models, to a group object. It can be used for attributes linked. Attention:

1. When you assemble houses, make sure that you've selected merged objects rightly. You cannot miss any of them or select redundant houses. 2. All the userid merged must be group userid, or it will appear the mistake prompt.

§5.2.8 Plug-in Board 1. Modify Line Width in Bat

36

Modify the width of selected line objects with appointed width. “userid name” refers to the userid which you want to modify. If you select “self-defined line width”, you should set the line width according to “unified width”. And if you don’t select “self-defined line width”, you should set the line width according to “the attribute item of line width”. 2. Locate by OID

Input the OID of object in the pop-up dialog.And the system can automatically locate the object. 3. Convert the ZLJ of selected point object to SUP

It is mainly used when we import 3ds model. We can promote the display quality of imported 3ds model. ZLJ and SUP are all file suffixes. The display quality can be promoted when it is converted to sup. Steps: select the object which you want to convert. We can see the pop-up dialog and input the parameters. 4.Data Statistic Stat the data amount of built object in 3DCM. Select the object and we can see the pop-up

37

dialog.

The ”Tex Total Num” refers to the whole texture number of repeated textures. ”Tex Real Num” refers to the texture number without the repeated textures. Repeated textures refers to several faces mapped to the same texture. 5.Drop down Bottom Corner Pull down all the bottom corner points of the object to a certain distance. Steps: select the object, select [edit│plug in】menu item. Select “Pull down the bottom corner points” plug in. Then input the pull down distance in the pop-up dialog. 6.Search surface whose texture is incorrect. Select the object, then choose [Edit│plug in | Search incorrect faces of texture】, and the system can automatically search the lost face of the object textures. 7.Insert many 3ds files Insert multiple 3ds objects at the same time. It’s mainly used in a building composed of several 3ds objects. Select plug in, then we can see the pop-up dialog as follows.

Select several 3ds files, input the zoom parameter of the coordinates. Then select userid(generally points) before they can be inserted. The imported 3ds model is at the origin of the scene. 8.Model Joint If there is aperture between two linked models, you can use “Model border merge” to deal with it.

38

9.Attrib to Anno Change the marker of object’s attribute, keep it the same with the attribute. 10. Convert texture to color Change the texture files to the color the user want

11.Search model gap

Input the size of the aperture to search, the system will search automatically in the workspace, it will list the results 12.Save polygon file Save the line surface feature selected as *.sel file, choose the saving path in the popup dialog 13.Select polygon file Choose object according to the polygon file, choose a polygon file in the popup dialog. Then the system will select the objects in the polygon. 14.Set feature color

39

Set the color for the surface feature selected. 15.Interpolate DEM Interpolate DEM feature points in the surface selected, it can be chosen directly when you create surface model.

16. Split surface

Divide the surface objects according to some “horizontal mesh space” and “vertical mesh space”. 17.Compare workspace If you add some objects in A workspace and change it to B workspace, you can compare the different steps between the two workspace. (1) Open workspace B (2) Select all the objects in workspace B (3) Select [Edit | Insert piece | Compare the workspaces], choose workspace A in the popup dialog (4) You will see the cue as follow:

(5)Choose “Yes”, the system will select the repeated objects in workspace A and B. 18. Import TIN data Input TIN data, the inputted data are at the origin of the scene. 19.Modify Sup material Insert SUP file, select the insert piece, you will get the popup dialog .In this dialog, you can modify the color of material environment light, scatter light and the mirror surface light.

40

20.Input borehole file and add to the workspace import bore files(*.txt) into workspace.You need to choose bore files(*.txt) and color files(*.txt) After imporing the bore datas .you can check and modify the attribute of every bore segment.And you can also compose multiple bore segments into a whole bore segment. 1.Check and modify the attribute of every bore segment. Choose multiple bore segments, then choose [Attribute | Query Object ].System will pop up a dialog. In this dialog, you can choose a segment, Then check and modify it's attribute. If this segment has not a table of attribute, you can create a table. Please refer to 6.6 section. 2.Compose multiple bore segments into a whole bore segment .Choose multiple bore segments, then choose [Edit | Compose].System will pop up a dialog, ou need to name this group object Group. Then you need to choose [Edit | Modify UserID】and modify the name of this group object. 21. Dispose all the converted work from 3DS to SUP Convert 3DS file to SUP directly. This selection compose the function of “Convert ZLJ of selected point object to SUP “and “Modify SUP Material”. 22.Search surface whose network is incorrect choose this plug-in board, system will find out the surfaces which should have been built network If a surface is a cave polygon, it must be built network in order to display correctly. 23.Adjust Location Height of Object Choose this plug-in board, system will adjust the height of object. Such as if a object is not at ground ,system can set it to the ground.

41

Chapter 6 The General Process of 3D Modeling The former three sections introduce the pretreatment of data, which make the abundant data preparation for the modeling data production of 3D visualization. After the pretreatment, you will get the data as follow: correct single digital elevation modeling data and digital orthographic image data, each measure’s intact digital elevation modeling data and digital orthographic image data, 3DS modeling data, field map attribute data, texture data. On the basis of the these data, next we will introduce the process of creating 3D viewing modeling to make workspace in 3DCM, building modeling of each surface feature and building after creating the workspace, mapping the texture of building profile, making the surface viewing modeling, inserting 3DS modeling, assembling modeling and inputting attribute, checking the modeling on the basis of each measure area’s digital elevation modeling and digital orthographic image data. The following we will introduce according to the data production procession, the beginner can be familiar with it according to the corresponding practical data.

§6.1 Transfer Data and Build Model There are multiple modeling methods in 3DCM.We often use three kinds of data source:3D coding data of JX-4A, DXF, two-dimension vector and attribute data of GeoStar. 3DCM can build whole 3D surface models by these source data and edit any 3D object, including changing geometric shapes, associating different attributes and textures with models and so on. The result of edit and modeling will be saved in the database according to the appointed model type and structure. Next this tutorial will introduce the modeling on the base of three kinds of data source separately. §6.1.1 3D Code Data(JX-4A, “*. ASC”): You can perform this process with *.asc data file Steps: ⑴ Execute the 3DCM.EXE file and enter the systematic interface of the program. (2)Click [file│new] submenu, you will see pop-up “workspace” dialog as follows. Choose “transfer data in workspace” and click [browse], you can choose working directory of new workspace. Then you need to input the name of workspace in “Name”

42

(3)Click [create],system will popup “open” dialog. you can choose DEM file for new workspace. Then system will popup “open” dialog, you can choose original model data file(*.asc). (5) After transferring modeling successfully, system will popup dialog as follow. This dialog mainly set the visible scope of DEM and modeling surface feature .If you want to display the whole workspace, you need to click [cancel] button. Then system will open new-creating workspace automaticly. If you want to set the visible scope of DEM, you need to tick in “visible scope”. The middle white district in the left is DEM range. You can set visible range by drawing a box with mouse in this range. When the mouse is in the range of DEM, you can see XY coordinate corresponding with current mouse position real-time at the bottom of the left.

If you need to set visible UserID, tick in “visible userID”, then click [select], system will pop-up dialog as follow:

43

I. Click [Add], dd a new layer which contain all visible userid.

II. After input the name of visible layer, click [ok]..The new layer will be showed in the “All layer” list box. Choose this layer, then in “optional” list box choose the userid which you want to display in this visible layer (all visible UserID of workspace list in “optional” list box ). Then click [Add>>], the UserID which you want to display is added into “After Set” list box. A visible layer may have multiple UserID.

44

Parameter declaration:

: create a visible layer

: delete a visible layer

: modify layer name

: add a UserID into a visible layer

: delete a UserID in a visible layer

: add all UserID of the workspace to a visible layer

: delete all UserID in “After Set” list box

III. After setting, click [ok], return to the upper dialog.

45

IV. Right box list the visible UserID which you setting. Click [OK],system pop-up main interface to display workspace. §6.1.2 Workspace data (*.gws) whose data source is Geostar. You can perform this process with *.gws data file. Steps: ⑴ Click 3DCM.exe file, you can enter the interface of this program. ⑵ Click [File | New] submenu, you will see the pop-up “Workspace” dialog. You can select “Transfer data in workspace” in the dialog, then click [Browse] button to select the project directory of the new workspace, or you can input the project directory of the new workspace in “Directory” edit box, then input its name in “Name” edit box.

⑶ Click [Create] button, you will see a pop-up dialog to let you select DEM file corresponding with the new workspace, then you will see a pop-up “open ”dialog to let you select original model data file (*.GWS) which you need to import. ⑷ You will see the pop-up dialog as follows:

46

1. Select the userid that you want to import in “Feature Name” dialog. 2. Confirm the userid’s imported model type in “Target Type” dialog. 3. Confirm that the userid is imported by absolute elevation or relative elevation. 4. Select the “Elevation Attribute Item”. 5. After setting the userid, you can click [Add] button, and then you can see the information of this userid in the imported list under the dialog. 6. You can set the userid one by one or import a lot of userid at one time. Parameters explication:

Feature Name: It lists all the userid of this workspace. Select this layer and then you can set for the data of this layer. If you don't set it, the system will not process this layer.

Target Type: the model type after importing userid. There are four choices including housing (volume model), area object (area model, without thickness), line object (line model) and point object (point model).

Relative Elevation: Elevation attribute of data in this layer is relative coordinate. (For example: corresponding attribute option is height of the house, house floors and so on), i.e. the height is calculated from ground, you can select this item.

Absolute Elevation: Elevation attribute of data in this layer is absolute coordinate. I.e. the height is calculated from sea level, you can select this item.

Elevation Attribute Item: If the data of this layer has affiliated elevation attribute,

47

please select elevation attribute item here. If you want to change objects into buildings and there is only floor attribute, but no height attribute of this buildings, select floor attribute, tick in “Calculate Height By Floor”, then appoint height of every floors (meter) in “Height of Each”. If the layer hasn't elevation attribute, select the first choice (blank) of the lists, and appoint the height of house in “Elevation Modify Value”.

Calculate Height By Floor: For the buildings, if you tick in this item, it will calculate the building height by floor attribute.

Height of Each: Height of buildings’ each floor. Elevation Modify Value: If there isn’t elevation attribute item for one layer, this

selection will enable you to set the height of this layer which is calculated from ground plane. ⑸ Click [OK] button, the system will model automatically according to model data file and DEM file, and the model data after modeling will be imported to the appointed workspace. ⑹ If you import model successfully, you will see the pop-up dialog, this dialog will enable you to set visible scope of DEM and userid of model in workspace. If you want to display the whole workspace, you can click [Cancel] button. If you want to set the visible scope of DEM, you can tick in “Visible Scope”, the middle white district in the left is DEM range, set visible scope by drawing a box with mouse in this range (When the mouse is in the range of DEM, you can see XY coordinate corresponding with current mouse position real-time at the bottom of the left.).

If you want to set visible userid, tick in “Visible UserID” firstly, then click [Select] button, you will see the pop-up dialog:

48

I. Click [Add] button, you can add a new layer which includes all the visible userid.

II. Input the name of visible layer, then click [OK] button, you wil see the new layer in “All Layer” list box, select this layer (blue), all the userid in workspace is listed in “Optional” dialog, select the userid which you want to display, click [Add] button, it will be added to the “After Sel” list box. One layer can includes a lot of userid.

49

Parameters setting:

: Create a visible layer.

: If you select a layer in “All Layer” list box, then click this button, you

can delete this layer.

: If you select a layer in “All Layer” list box, then click this button, you

can modify the name of this layer.

: Add a userid in visible layer.

: If you select a userid in “After Sel” list box, then click this button, you

can delete this userid in visible layer.

: Add all the userid in the workspace to the visible layer.

: Delete all the userid in “After Sel” list box.

7. Click [OK] button.

50

8. You can see the userid which you set, click [OK] button, the system will display the workspace. §6.1.3 CAD format—dxf data (*.dxf) You can perform this process with *.dxf data file (1) You need to check if the lines of building are closed in CAD .If they are not, you must make them close. The picture is shown as follow:

(2) Execute the 3DCM.EXE file and enter the systematic interface of the program. (3) Click [file│new] submenu, you will see pop-up “workspace” dialog as follows. Choose “transfer data in workspace” and click [browse], you can choose working directory of new workspace. Then you need to input the name of workspace in “Name”

51

(4) Click [create],system will pop up dialog ,in which you can choose corresponding DEM file of new workspace. (5) After opening DEM file, system will pop up dialog. You can choose dxf format date. (6) Then system pop up dialog as follow, you can set parameter of models.

Firstly, in “LayerName” combo box ,you can choose the layer which you want to import. Next you need to choose target type of this layer in “Target Type” combo box. then you need to choose “Absolute Elev” or “Relative Elev”. After setting, the information of this layer will be display in inferior list box. Then set other layers in turn. System support importing multiple layers at the same time. Parameter declaration:

LayerName：It lists all the layer name of DXF file. Select one layer, and then you can set for the data of this layer. If you don't set it, the system will not process this layer.

Target Type：

52

1. Build Volume From Roof: the data of this layer contains Z value, which is the elevation of roof point.

2. Build Volume from ground: There isn't Z value for the data in this layer; but there is plane coordinate information. You should define volume's height yourself. (The height of all the volumes in this layer is the same. If you want to make them different, you can divide data into several layers according to their elevation, and then import them.)

3. Surface: Interpolate a 3D surface on DEM with 2D information in this layer. If you want to appoint that this surface is lower (or higher) than DEM (for example: water face), input its height that lower or higher in “Elve Offset”

4. Line: Interpolate a 3D line on DEM with 2D information in this layer. If you want to appoint that this face is lower (or higher) than DEM (for example: water face), input its height that lower or higher in“Elve Offset”

5. Point: Interpolate a 3D point on DEM with 2D information in this layer. If you want to appoint that this face is lower (or higher) than DEM (for example: water face), input its height that lower or higher in “Elve Offset”

6. Annotation: Interpolate a 3D point on DEM as the position of bottom center of the annotation with annotation information in this layer. If you want to appoint that this annotation is lower (or higher) than DEM, input its height that higher in “Elve Offset”

Relative Elev --Z value in DXF and elevation interpolated on DEM will be used as object's elevation coordinate in the process of modeling.

Absolute Elev -- Z value in DXF will be used as object's elevation coordinate in the process of modeling.

Elev Offset -- Appoint value which area object is higher or lower than DEM, if you need to higher area object, input positive, or else input negative.

Building Height -- Appoint the height of the house when you build volume from ground. The unit is meter. ⑸ clicking [OK],system will create modeling according to data file and DEM file automatically. At the same time, system export the data of modeling into appointing workspace.

53

§6.2 Modeling Important UserID After modeling large number of buildings, you can begin to map the profile texture of building and model ground objects. And you need to model a few buildings in 3DCM workspace directly. Next this tutorial will introduce how to modeling important UserID in 3DCM workspace. §6.2.1 building There are multiple methods to modeling building. This tutorial have introduced three methods. This chapter will introduce other methods. (1) If the building have very complex geometric character and peculiar body, you can collect feature points of geometric models by using JX4 photogrammetric workstation. then you can get ASC files and import them in 3DCM to model buildings automatically. Such as the modeling building explained in the 6.1.1 section of this chapter can adopt this method to get geometric model. Then you can model peculiar house roof by using 3ds max. (2)If the building have simple geometric character and regular diameter, you can import the gws files and dxf files in 3DCM workspace to model buildings automatically. As chapter one tells, about the data source format of dxf and gws, the automatic creating modeling can collect geometrical model according to feature points, as to comparatively complex formal roof, use 3DS method to create model (3)According to practical situation. You can add some outhouses into workspace by using 3DCM.Please refer to 3DGIS manual in detail. §6.2.2 Road The road is the emphasis in ground objects modeling. Dividing big work space depends mainly on main roads. Before modeling ground objects, you must modeling main roads in order to ensure the seamless-connection among small work space. 1. The general steps of modeling road ① Collect the material of field survey and photos of roads. The field survey must depend on large-scale terrain map and orthophotoimage. About street with zebra crossing, vegetation(including trees, treelawn, flower and grass) and attached facility of roads , you must survey and take photos one section by one section of road. ② Collect other materials of roads modeling. That include materials of the region-divided, DEM,DOM (have been overlayed the large-scale terrain map) ③ Draw polygon of divided workspaces ④ Model vegetation according to field survey and photos. ⑤ Model road, bridges and affiliated facility. 2.Making zebra crossing of street Different from the texture character of general surface model, the zebra crossing has special design character. This type of character includes two aspects. One is symmetric character of the two-way street. The other is the change of the number of roadway. the symmetric character of two-way street: The two-way street has symmetric character in general. So in process of texture making,

54

the polygon of roadway in different direction must have the same size, position and length. the change of number of roadways: The number of roadways is changeable in some specific district, such as changing from three-roadway .to four-roadway. In order to ensuring the continuity, you need to use particular method to make textures. Next this tutorial will introduce this method. a) The making of texture Make the road texture (showed as follow) in image processing software separately.

b) The making of model Make road model in 3DCM

c) The texture mapping Map the prepared road texture by the default way in 3D window.

55

The default road texture a) the modeling requirements Draw nodes of roads in order.

b) Click [Edit│Set Default Texture]. Set default texture. 3.the mapping of road texture The mapping method of road texture which have zebra crossing or central line: Click [texture mapping] command in 3D view, then click Road Tex in the pop-up window. Select four points in the road with mouse (you’d better select withershins). Click the green button in the up-left, we can see the pop-up success or warning information of texture mapping, then select the needed texture file in the texture mapping window. Click OK (as follows :)

56

The clicking position and sequence in road texture mapping

Corresponding road texture

The effect after road texture mapping is done

§6.2.3 bridge There are various bridges in modern city, such as skywalk and flyover. In general, they are produced by using 3DS MAX. Because the bridge has complex structure, you need to

57

notice some problems in process of field survey and taking photos. ① You need to take photographs of the bridge face, side face and panoramic scenograph .You need to refer to them when you model the bridge. ② In order to make sure the model is accord with the object in radio, you need to note breadth ,length and height radio of bridge.. ③ About some things that need to be described clearly, such as the name of bridge, guidepost, billboard, you need to take photographs of them respectively. ④ About pier, balustrade, steps and so on, you need to take high quality photography to produce texture. Because the skywalk and large-scale flyover have different spatial character, they are modeled by different methods. Next this tutorial will introduce these methods. Skywalk

Modeling: Using general Box modeling. Mapping: About pier, use default Box mapping. If a box have two kinds or more mapping, you need to use Edit Mesh, respectively. large-scale flyover

58

This type of bridge have big volume and abundant bended radian. So it is some difficult to modeling. Modeling: Use Loft-lofting to model bridge body., You had better do tersely in process of drawing magic square of bridge body. Mapping: For models that are produced by Loft-lofting you’d better not use Edit Mesh to mapping. Because the mapping will become clutter by using this method. You can use default Box mapping. And you had better make the texture a whole.

§6.2.4 lawn 1.In 2D window, draw line model and snap node along road(if there have pavements, snap node of it’s inside) 2.Choose the finished line models, then choose[Edit│Dense Line Node ] and [Edit│Simplify Geometry]. 3.choose [Edit│Convert Geometry Type], convert line model to surface model. 4.In the 3D window, click , then choose texture file and choose repeat by length. Attention: Like the road modeling, the lawn need to be modeled in small section. §6.2.5 water system The water system modeling is the same as the lawn modeling. §6.2.6 sight facility There are many types of sight facility. You need to produce them according to practical

59

situation (1) Modeling lawn, road, pavement, shoulder, parterre and so on. In order to ensure seamless-connection and convenient operation, you can draw large-area ground sight by using CAD according to large-scale terrain map and actual photos. This method can create the dxf file. Then you need to import the dxf file into workspace of 3DCM to model surface and volume. And then you can map texture. A few models can be created in workspace according to practical situation. Steps: (1) Draw according large-scale terrain map and create the dxf file.

(2)Import the dxf file into 3DCM workspace.

60

the result is as follow:

(3) Map textures of surface and volume.

61

(2) If the objects are trees and so on, you can model them after modeling roads and pavement. (3)If the object is street lamp, garbage bin, billboard ,waiting room and so on you need to import the 3DS model. §6.2.7 scarp

Modeling scarp is to draw vertical surface. Steps 1.In 2D window, draw vertical surface according to A surface (snap nodes ). Attention: you need to know Z-coordinate of A surface nodes. And use the Z-coordinate as the base height of vertical surface.

62

2.choose B surface and vertical surface and click “Change Display Mode” button. 3.According B surface, set the remainder nodes of vertical surface. Choosing [Display|Display NodeEdit Tool], you can edit nodes. Attention: You need not to tick in “Link Up and Down” in Node Edit Tool dialog.

4. The result is as follow:

§6.3 Map Texture The essence of real 3D scene texture process is to increase object reality in visualization through texture image detail presentation. It is related to all types of object texture image processing and mapping in 3D scene. The general requirements (such as size, hue, naming, etc.) of building texture production in 3D scene are the basic requirements of all

63

texture image production in the scene. For the specialization of managing methods, some types of object have some special requirements. The concrete classification and producing requirements are as follows: ① Building profile texture management: the mapping operation is executed in 3D texture window of 3DCM. ② Tree texture surface management: Directly insert point model into scene and present trees with texture surface. ③ 3DS model texture management: one demand of 3DS texture production is that file name doesn’t exceed 8 characters. ④ Lawn texture management: map texture repeatedly. 1. Click [Display│ Display 3Dview] menu item, we can see a pop-up independent window to display and edit 3D model. Attention: we must select an object model in the main window before 3D view is startup. Pressing right key and dragging it in this window can rotate the model arbitrarily gTexture Mapping, Scale, Rotate , Set As Top Window, Change Display Mode. 2. Click toolbar button, we can see a pop-up “texture mapping” dialog,

3. Select the ID number of surface in the pop-up dialog ,then choose the corresponding texture for this surface in “Tex File”

64

§6.4 Inserting 3DS Model After modeling the building and ground objects, you can import all 3DS model into 3DCM workspace and adjust them according to scale and position of other objects in 3DCM workspace. These 3DS model are made by using 3DS MAX

§6.5 Assemble Object You can assemble many isolated models whose attribute are the same into the whole by assembling object, its purpose is to reduce the work amount of importing attribute and query the whole attribute, for example: at the time of modeling, one building is divided into many isolated models, but at the time of querying attribute, you are interested in the attribute of the whole building, not one part of the building, so you must assemble object. After modeling for all the models, you can input attribute of userid, such as buildings, main road, and 3DS model. Before you input attribute, you must assemble volume model of buildings and userid whose attribute are the same into group models in 3DCM, and define the uniform type name, for example: buildings can be defined as "architecture". The following will introduce the process of assembling object; you can tackle the process of assembling volume according to the actual situation: For the large main buildings, you must assemble them according to the amount of volume models that the actual building includes, and input attribute according to the actual situation; If there isn’t formal name for a group of unimportant buildings, you can assemble some buildings into a group model and input model data. Assembling of 3DCM can be operated in 2D windows. Steps:

① Click [Display | Display 2D View] to display 2D edit windows;

65

② Set the selection mode as “rectangle or polygon select”, select userid as “body” type in the “Select Current Userid” dialog;

③ Select the userid which you want to assemble; ④ Click [Edit | Compose] submenu; ⑤ Select “building” userid in the “Group Object Setting” dialog to complete the

creation of group object.

Assemble houses, which belong to a building but are disassembled to several volume models, to a group object. It can be used for attributes linked. Attention: When you assemble houses, make sure that you've selected merged objects rightly. You cannot miss any of them or select redundant houses.

§6.6 Input attribute and multimedia data Attribute data is another part of content of 3D model data. It is involved in two parts of content in 3DCM, one is attribute of one type of object, and the other is the multimedia attribute of one special object. For the former object, you need to create attribute table for the special userid, then input attribute for the single object and link attribute; for the latter, you only need to link the multimedia for the special objects. 1. Attribute structure design

Data structure design -- All the attribute data must be set according to promissory data types and field length, or it can not process attribute data at the time of merging work space and putting work space to library. Steps: 1. Click [Attribute | Attribute Struct] submenu, you will see the pop-up “attribute data structure design ” dialog:

66

Parameters setting: ◇ Input table name in table name edit box, such as building; ◇ Input field name, data type, field length in field attribute; ◇ Tick in Object ID Field check box (default).

2. Select “add” to complete design of this field, repeat it until all the fields have been designed. 3. Press “Create” to complete the design of this table, and press “Related Classes” button, you will see the pop-up dialog:

Select corresponding userid name of this table, for example: house, buildings, and then Press “Create Empty Record” button. 4. Lastly select “Done” of attribute data structure design dialog. An empty table has been

67

created, all the userid after this must be created with this table.

2. Link object attribute

Steps: 1. Select the object, which needs to link attribute, then click [Attribute | Query Object] submenu, you will see a pop-up dialog to display object’s attribute information.

2. Double-click left mouse's button in the position where attribute needs modifying. Then modify the attribute.

If there are a lot of attributes of objects to be modified, don't close the attribute dialog. Click the next object, which needs to be modified, and the information of attribute dialog will be updated automatically. Then edit its attribute directly. Attention： If attribute cannot be saved after modifying, it may be that there is no corresponding blank record for this object. You need to design attribute structure again. Preparing work: Attribute structures design of all data.

§6.7 Merge sub-workspace After the measure area’s 3-D scene viewing model have been finished and attribute imported, merge all the sub-workspaces as following steps: on the base of one workspace, add the other workspaces to it to form a independent workspace of 3-D model data in this measure area. Before partition of the sub-workspace, the meeting border area has been made,, after merging the workspaces , no aperture will appear ultimately in this area, quality check

68

must be very careful and expert used to look through and check during this area. Once aperture discovered, modification of the meeting border is not allowed in independent sub-workspace, appoint person to modify the aperture in the workspace after merge.

§6.8 Data quality check Data quality check play a great part in the whole process of data production, whether

its content be comprehensive and check method be reasonable or not will relates to the whole data quality. 3 types data are involved in data quality check, which refers to topographic data, model data and 3-D viewing model data. Topographic data check contain DEM and DOM data quality check, model data check refers to building geometrical model data check, 3-D viewing model data refers to building view data and ground viewing model.

§6.8.1 Topographic data check

DEM check ① Whether DEM mesh space be 5 meter ② Whether DEM data be clipped correctly ③ Whether DEM order reserve one digit after decimal fraction, and precision be 0.8

meters ④ Whether split exist between bordering chart breadth ⑤ After margining the bordering chart breadth DEM border, whether all

corresponding mesh points have the same altitude. ⑥ Whether DEM data be edited and express ground variety diagnostic topographic

surface data During the process of collecting topographic feature points, when man-made surface features tangency with the terrain, adjust feature points to make them tangency with where the topographic varies, avoiding 3-D surface features tangency under the terrain during 3-D modeling construction. Use the DEM you have produced as building datum plane to construct building 3-D modeling, if some profile’s two ending elevations of the slow terrain building are no morn than 1/3 of the profile’s height, DEM points in this area must be re-measured.

DOM check ① Whether the resolution be 0.6 meter ② Whether DOM data be clipped correctly ③ Make sure neighboring DOM breadth overlap 50 pixels ④ Make the neighboring DOM breadth the same hue

§6.8.2 Model data check

Geometrical model data check refers to checking the geometrical model of building, this should be done carefully before producing building viewing model. It will increase the workload once you discover geometrical model error after building viewing model.. During the process, much attention should be paid to as follows:

(1) Integrality of data

69

(2) Geometrical precision the feature points (3) Reality of model (4) Topological structure and precision of building

§6.8.3 3D viewing model data check Finishing the 3D model, you should make strict quality check with the 3D model produced, which includes as follows:

1、 Integrality of data 2、 Reality of data 3、 Visual effect of data 4、 Definition of texture 5、 Logical error of data

Discovering error during check, quality personnel should note some point’s coordinate where it has been discovered. Make a record of error by recording on the document or scratch the screen, send them to be modified by data producer. §6.8.4 Methods of quality check In the process of making 3D viewing model, you perform three-class censorship, which are worker’s self-check, leader’s check, quality check expert’s check. The flow chart of quality check is shown below

70

Chart 6-8-4-1 flow chart of quality check

① Operator finishing the check of one important production, quality experts of this unit take a record of the quality check by document，according to which operator modify it ,leaving the experts to confirm all quality problems have been modified and sign.

② Check personnel appointed by production consign side check and make a record with the data, which has been confirmed and signed by quality check personnel of production unit. The quality problem will be noted, it also should be confirmed by both the operator and the check personnel. Operator re-modify to make the check personnel confirm, then transfer it to quality check personnel of production consign side, that re-confirm the quality problem and sign

③ Transfer the data signed by check personnel of production consign side to data quality principal of production consign side to check, he (or she ) check and accept by signing to confirm no quality problem .

In every class quality check, quality check personnel must judge the type and record according to the quality check manual with every quality problem, meanwhile three types of judgement are made to the data: re-produce, re-modify and check-and- accept

71

Re-produce: If serious quality problem excesses 5%, or absolute quantity excesses 10; to ordinary problem, if 10% or absolute quantity excesses 30. Re-modify: If quantity of existing quality problem doesn’t excesses the upper limit, operator modify it according to the record taken by quality check personnel. Check and accept: Modifying all quality problems, correlative quality check personnel and quality principal sign to confirm. On the basis of check censorship above, collocate check personnel as follow

① Every production unit collocate a full time check personnel at least, he (she) can check the data all operator have finished.

② Production consign side collocate one or two check personnel in every production unit, who check the data signed and confirmed by production quality check personnel;

③ Production consign side needs at least one data quality check personnel to check and accept the data finally.

72

Chapter 7 Building database Finishing produce the terrain data of the measure areas and 3D viewing model , the next work is using 3DDB to build database with the data produced, which is on purpose to realizing mass data dynamic seamless elevation in 3D GIS.

§7.1 building database with DEM data You only need to build database with DEM data processed in the pretreatment part. No modification is needed. DEM data contains two cases, one is huge scope divided into many files recording DEM data with breadth, the other is only one DEM data. Now we will introduce to you how to build DEM database with the two cases separately. §7.1.1 Original DEM data consists of many breadths of data

Operation steps 1. Execute 3DDB.exe file, enter into systematic interface 2. Click [file│file system manage│DEM manage], enter into DEM file manage system 3. Click [DEM file system│create database] menu, you will see the pop-up dialog as

follow

Parameter explication: □ Database path: Path for stocking DEM data □ Database name: Build name for DEM database □ Plane coordinate unit:: DEM data plane coordinate unit □ Original data mesh space: Mesh space of DEM data □ Elevation byte count: How many bytes every DEM elevation value uses (choose 2 or

4) □ Zoom out factor: Elevation need to multiply zoom out rate if you use 2 bytes to show

height □ Overlap precision threshold value: The biggest restrict value of overlap precision of

data layer between blocks

4. Click the button , you will get the popup browsing files dialog, choose saving

path of database, click [OK] button. Return to building database interface

73

5. Import parameter information, click [OK] button, build database 6. Click [DEM file system│Data Layer Manage] menu, you will get pop-up dialog as

follow

Parameter explication

□ DEM database layer list: record names of all the layers from the bottom to the top, users can choose one layer at a time

□ Create new layer: provide the user the operation of creating any new layer. Two ways for creation: submit data exteriorly and extract data from the bottom. New layer created can only lie on the top, that is, only the top layer can be created. The number of layer is limited to 12.

□ Delete current layer: provide the user the operation of delete the layer selected. Only the top floor of database can be deleted, if any layer in the middle is selected, non-use state turns on

□ Submit data: provide the user the operation of re-submit data with the layer selected. This operation cannot be done with the data extracting the bottom but with the data layer created from external submitting data. Button will be gray if you create data layer with the way of extracting from the bottom.

□ Substrate extraction: provide the user the operation of extracting the substrate data to the upper layer. 0 layer is the bottom layer, it cannot be used to the substrate extraction button. Substrate extraction and submit data are two ways of make layer data

□ Query layer information: provide the user the operation of layer parameter information query with the layer selected.

Notes: DEM data layer manage dialog is a integration environment about pyramid database, In which user can do all the operations about the layer manage in the dialog. Application

74

of different button have some conditions. ◆ If bottom layer (0 layer) is selected, [DEM data layer manage│delete

current layer] and [DEM data layer manage │re-create current layer] button is gray, it is unable.

◆ If ground layer is selected as extracting layer, [DEM data layer manage│delete current layer] button is gray, it is un-useable

◆ If middle layer is selected, [DEM data layer manage│delete current layer] button is gray, it is unable

◆ If the database opened is blank (that is, there is no layer name record in the DEM data layer list) [DEM data layer manage │ ground layer extraction][DEM data layer manage │ submit data][DEM data layer manage│query layer information] buttons are gray, they are unable. Now system is waiting the user to execute [create new layer] operation, form original layer (0 layer)

§7.1.2 original DEM data as one breadth

Operation steps 1. Execute 3DDB.exe file, enter into the systematic interface 2. Click [file│file system manage│DEM manage] menu, enter into DEM file manage

system 3. Click [DEM file system│special create mode│one file create..] menu , you will get

popup dialog as follow :

Parameter explication:

□ Raw data name: Original DEM path and name needed to be created □ Database path: Path for stocking database □ Elevation byte counts: How many bytes every elevation value uses (choose 4 or 2) □ Zoom factor: If you use 2 bytes for elevation description, the elevation needs to

75

multiply the zoom out rate □ Overlap precision threshold value: The biggest limit value of overlap precision of data

layer between blocks 4. File in information parameter, click [OK] button, finish database creation

§7.2 creating digital orthographic image database In this part, you only need to create database for the digital orthographic image data , which have been dealt with in the pre-process part, no modification is needed. Original orthographic image can be bmp format or tiff format. Operation steps: 1. Execute 3DDB.exe,enter into the systematic interface 2. Click [file│file system manage│DOM manage] menu, enter into the DOM file

manage system 3. Click [DOM file system │create database] menu, you will get popup dialog as follow:

Parameter explication: □ Correlative DEM database path: DOM refer to the DEM database path (created from

the chapter above) □ Basic information of image database:

Raw image resolution: Resolution of original DOM Interpolate image: Whether the 0 layer data be the original interpolate image Overlap pixel value: Overlap pixel numbers of DOM database between blocks Database path: Stocking path for the new DOM data Database name: Name of the new DOM database built

Notes: Parameters got from DEM database will be condition of dividing the DOM database. They can make DOM database and DEM database be the same structure. Users import parameter to check data in the process of submitting original DOM image.

4. Click the button on the right of the [correlative DEM database path] submenu,

you will get the popup browsing file dialog, choose the saving path for DOM.

5. Click button on the right of [database path] submenu, you will get the popup

76

browsing file dialog, choose the saving path, click [OK] button to build DOM database.

§7.3 Creating database of 3D visualization model In general, the area to produce 3D visualization model data are divided into many measure areas, in the final mass data browse, it is not merge all the measure areas into one for mass data that can make the system unworkable, and the mass data seamless ramble impossible. To avoid the effort caused by mass data, we adopt the way that build one workspace for every measure area, forming a project by creating database to transfer mass data dynamic and realize mass data seamless ramble, also it is convenient to manage and modify data. Operation steps: 1. Execute 3DDB.exe file, enter into the systematic interface 2. Click [file│file system manage│model manage], enter into model manage system 3. Click [MODEL file manage｜model data submit…], you will get popup dialog as follow:

Parameter explication: □ Create new project: set the path and name for the project □ Open the project: open a file which contains project information, it exists in the path

“project\WORKSTRU”, many “*.ifo” files are there to choose, any one will be ok. □ Attribute Check: Check if there are errors in workspace. □ Add: Add a workspace to the project, at present no data is inputted. Click

to input workspace after choose the workspace name added. Click it, you

will get the popup dialog as follow, import workspace name. The name can be any but

77

not the same as the ones existed in the workspace

□ Delete: choose a workspace in the list of workstruct, clicking “delete” will delete a

workspace in which the data existed will be deleted with the workspace at the same time , but the texture related with the workspace and 3ds model will remain

□ input: Input model to a workspace added by the button , the button

will be light when the workspace selected is accord with the condition

above. Click it, you will get the popup file dialog, choose the corresponding workspace to confirm

□ Input by batch :Input a series of workspace to the project, which are selected from the following dialog

□ Input by batch flies: on the basis of a pre-define workspace list files, submit to the

project with them. This workspace list files have the postfix “*bgs”, the formats are as follow:

CGS_IMP_BATCH_1.0 Full path name of workspace one Full path name of workspace two

……………..

□ Update workspace range and build workspace index: build index for all the workspaces in the project, the index will be used in 3D GIS enterprise edition 4.0 to browse the project. Origin is default. Space between columns and space between rows need to be adjusted according to the workspace range. In general, “workspace range/100” is ok. If the space is below 50, choose default (50).

4. If no project is created, create one first

78

5. Click 『Batch Input』, you will get the popup dialog as follow:

6. Two ways for inputting model data to the database ■ Choose several cgs files to input at random

Click button “add”, you will get the popup dialog for file open , choose several files to input ■ Input all the cgs files in a catalog

Click the button “add catalog”, you will see the popup dialog to choose path , choose all the files to submit

7. Submit data show the files submitted after the process is finished

8 Click the button , build index for

the updating workspace range, you will get the popup dialog as follow:

79

9. Click OK, you will get the popup dialog as follow:

Until now, 3D visualization model database is built successfully.

80

Chapter 8 Frequently question answer

§8.1 how to adjust multi neighboring surface node at the same

time ?

In the operation of “set up” ——“option” ,choose “show the object selected in the 3D

window ” , press the button “ctrl” to choose multi neighboring surface . In the perspective

view, drag to choose the nodes (if click to choose, only the nodes of a surface can be

selected), at present, modify the coordinate of node in the modify toolbar for nodes.

Chapter 9 Appendix

§Appendix one Ground model field note table

3D viewing model note table

serial number of workspace: missionary: weather: data: month day year

Photo

Serial

Geometrical

direction

Model name Geometrical

size

remark

81

§Appendix two Texture ,building attributes field note table

serial number of workspace: missionary: weather: data: month day year

Photo serial

Number

Building

number

Surface

number or

direction

Building

name

Number of

floor

Windows

of every

layer

Roof shape

Plane (p)

Roof

color

Main

institution and

organization

new building

or not

New house

Sketch or not

Remark

(door)

§Appendix three Surface feature classes and corresponding table

Surface feature classes and corresponding table

point

Sight establishment P independent tree

Tree beside the road

shrub

gloriette

sculpture

garden

chair

person

Public establishment P

mast

telephone booth

street lamp

82

waiting room for car

ash bin

bicycle shed

publicize window

basketball frame

bridge

car

Line

center line of load

surface

Sight establishment S lawn

spread terra

concrete terra

fence

Public establishment S sign board

plaza

playground

parking lot

Main loads

Spur track

Water track lake

Natatorium

volume

Body

Sight establishment V parterre

step

lawn

terra brick

Public establishment V parking lot in the open air

sidewalk

83

Group

Important building

Common building

Trunk road

Hypo-road

84