UNIVERSITY TEKNIKAL MALAYSIA MELAKA DEVELOPMENT OF AUTOMATIC QUALITY CONTROL SYSTEM USING VISION SYSTEM This report submitted in accordance with requirement of the Universiti Teknikal Malaysia Melaka (UTeM) for the Bachelor Degree of Manufacturing Engineering (Robotics and Automation) with honours by CHIN GEE SING B050710006 FACULTY OF MANUFACTURING ENGINEERING 2011
24
Embed
UNIVERSITY TEKNIKAL MALAYSIA MELAKAeprints.utem.edu.my/5294/1/...System_Using_Vision_System_-_24_pages.pdfUNIVERSITY TEKNIKAL MALAYSIA MELAKA. DEVELOPMENT OF AUTOMATIC QUALITY CONTROL
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UNIVERSITY TEKNIKAL MALAYSIA MELAKA
DEVELOPMENT OF AUTOMATIC QUALITY CONTROL
SYSTEM USING VISION SYSTEM
This report submitted in accordance with requirement of the Universiti Teknikal
Malaysia Melaka (UTeM) for the Bachelor Degree of Manufacturing Engineering
(Robotics and Automation) with honours
by
CHIN GEE SING
B050710006
FACULTY OF MANUFACTURING ENGINEERING
2011
DECLARATION
I hereby, declared this report entitled “Development of Automatic Quality Control
System Using Vision System” is the results of my own research except as cited in
references.
Signature :
Author’s Name : CHIN GEE SING
Date : 18 MAY 2011
APPROVAL
This report is submitted to the Faculty of Manufacturing Engineering of UTeM as a
partial fulfillment of the requirements for the degree of Bachelor of Manufacturing
Engineering (Robotics and Automation) with honours. The member of the supervisor
committee is as follow:
……………………………………….
Supervisor
i
ABSTRAK
Mesin sistem visi banyak digunakan dalam industri untuk tujuan pemeriksaan.
Sistem menyusun automatik dan sistem pemeriksaan visi yang dilaksanakan untuk
integrasi telah membina sistem automasi. Kecekapan ia memeriksa dan menyusun
secara teratur telah member banyak faedah untuk mengatasi kelemahan sistem
konvensional yang dijalankan oleh pekerja. Integrasi konveyor ke sistem yang
dicadangkan adalah untuk menghantarkan objek untuk diperiksa. Kemudian, sistem
pemeriksaan dimulakan dari pengesanan sensor fotoelektrik untuk mengaktifkan
“webcam” menangkap gambar melalui satu PLC. Gambar yang ditangkap akan
didigitalkan oleh “webcam” dan menghantar ke hos komputer untuk pemprosesan
gambar dengan menggunakan perisian Vision Builder Automated Inspection (VBAI).
Konfigurasi untuk perisian VBAI akan dipelajarikan sebelum pemeriksaan visi
dijalankan. Sementara itu, grafik antara muka pengguna direkakan dengan
menggunakan perisian Labview tujuan untuk pengguna mengawasi and
mengawalkan sistem tersebut. Objek yang diperiksa akan disusun secara teratur oleh
robot. Ia adalah satu robot enam paksi yang mempunyai ruang kerja yang berbentuk
separuh-bola. Namun, integrasi elektronik seperti papan “relay” adalah diperlukan
untuk menyelesaikan masalah ketidaksamaan voltan antara PLC dan prosesor Rhino.
Sebagai kesimpulan, projek ini telah dijalankan untuk mengklasifikasikan objek
dengan menggunakan pemeriksaan secara visual dan eksperimental validasi ke atas
prestasi sebenarnya.
ii
ABSTRACT
Machine vision systems are widely used in manufacturing industries for inspection
purpose. The implementation of automated sorting system to machine vision system
has develops an integration of automation system. Its efficiency to inspect parts and
sort according to order offers advantages which against the limitation of conventional
human-operated system. The integration of a conveyor to the proposed system feeds
the inspected parts for inspection purposes. Then, the inspection system is started
from a detection of photoelectric sensor to trigger USB webcam by image acquisition
via a Programmable Logic Controller (PLC). The webcam digitized the image data in
order to send to host computer for image processing by using Vision Builder
Automated Inspection (VBAI) software. The configuration of VBAI is studied on a
personal computer before vision inspection is carried out. On the other hand, a
graphical user interface (GUI) is designed by using Labview for person-in-charge to
monitor and control the entire system. After classification from VBAI software, the
inspected parts are arranged accordingly by Rhino robot. It is a six axis robot arm
which consists of a hemisphere of work space. Though, integration of electronic
hardware such as relay boards are necessary to solve the different voltage signal
between PLC and Rhino controller. As a result, this project is performed to classify
parts by using developed visual-based inspection and experimental validated its
actual performance.
iii
ACKNOWLEDGEMENT
I would like to thank for peoples who guided and contributed in various ways to this
PSM I report. This report would not have been possible without the expert advice
and suggestion of my PSM supervisor, Dr. Zamberi bin Jamaludin. Thank you Dr.
Zamberi for the invaluable information and suggestions that you provided and your
willingness for contribute your precious time to this effort. Besides that, I also thanks
to laboratory technician, En. Faizul and Pn. Norhafisah for giving me technical
support.
Thanks also to the Universiti Teknikal Malaysia Melaka (UTeM) for offering PSM
as a compulsory subject for students to gain skills on organizing a project.
Beyond the professional involved in this report, my coursemate and family were also
quite influential in this work. Thanks to my coursemate for sharing their precious
opinion, experience, comments and resources. Thank you for being there to listen to
my gripe about my work. Thanks for my beloved family unending support and
motivation.
Finally, thanks for the peoples who were there to lend a hand along the way. Thank
you.
iv
DEDICATION
I dedicate this report to my beloved parents, without their patience, support
understanding and most of all loves, the completion of work would have not been
possible.
v
TABLE OF CONTENT
Abstract
Abstrak
Dedication
Acknowledgement
Table of Content
List of Tables
List of Figures
List Abbreviations
1. INTRODUCTION
1.1 Background
1.2 Problem Statement
1.3 Objectives
1.4 Scope
1.5 Expected Result
1.6 Content
2. LITERATURE REVIEW
2.1 Introduction
2.2 Vision System
2.3 Machine Vision Component
2.3.1 Vision Sensor Types
2.3.1.1 CCD
2.3.1.2 CMOS
2.3.1.3 Comparison between CCD and CMOS
2.3.2 Image Processing Technique
2.3.2.1 Smoothing Technique
2.3.2.1.1 Lowpass Filter
2.3.2.1.2 Local Average
2.3.2.1.3 Gaussian Filter
i
ii
iii
iv
v
ix
x
xiii
1
1
2
2
3
3
3
5
5
5
7
9
9
14
17
18
20
20
21
21
vi
2.3.2.1.4 Median Filter
2.3.2.2 Edge Detection Technique
2.3.2.2.1 Prewitt Operator
2.3.2.2.2 Sobel Operator
2.3.2.2.3 Roberts Operator
2.4 Sorting System
2.4.1 Robot
2.4.2 Types of Robot Manipulator
2.5 Conveyor
2.5.1 Traditional Conveyor
2.5.2 Evolution of Conveyor
2.6 Measurement Devices
2.6.1 Analogue-to-Digital Converter
2.6.2 Photoelectric Sensor
2.7 Programmable Logic Controller (PLC)
2.8 Summary
3. METHODOLOGY
3.1 Introduction
3.2 Overall Research Strategies
3.3 Project Structure
3.3.1 Image Processing Procedures
3.4 Characterization of System Components
3.4.1 Conveyor
3.4.2 Rhino Robot Arm
3.4.3 Webcam
3.4.4 Personal Computer
3.4.5 Photoelectric Sensor
3.4.6 Programmable Logic Controller (PLC)
3.5 Summary
4. EXPERIMENTAL SETUP AND PROCEDURES
4.1 Introduction
21
22
23
24
24
25
26
27
29
30
30
31
32
33
36
38
39
39
39
43
45
46
47
48
49
49
50
52
52
55
55
vii
4.2 Experimental Setup
4.2.1 Hardware Component
4.2.2 Software Component
4.2.2.1 Logic of Electrical Control System
4.2.2.2 PC Access
4.2.2.3 Graphical User Interface (GUI)
4.2.2.4 Image Processing Configuration
4.2.2.5 Pick and Place Configuration
4.3 Experimental Setup Procedures
4.4 Summary
5. RESULT AND DISCUSSION
5.1 Introduction
5.2 Results
5.2.1 Ladder Diagram
5.2.2 Development of I/O Server
5.2.3 Graphical User Interface Design and Data Binding
5.2.4 Vision Inspection
5.2.5 Sorting
5.3 Discussion
5.3.1 Circuit Board for Relay
5.3.2 Matched Timing and Sequence between Robot Arm
and Image Acquisition
5.3.3 Numeric Indicators
5.4 Suggestion
5.4.1 Robot Feedback
5.4.2 Robot Efficiency
5.5 Summary
6. CONCLUSION
REFERENCES
55
55
58
58
60
62
63
66
67
68
69
69
69
70
71
72
74
75
75
76
77
77
78
78
79
79
80
82
viii
APPENDICES
A Ladder Diagram
B Rhino Robot Programming
C Block Diagram of GUI
85
86
90
92
ix
LIST OF TABLES
2.1 Feature and Performance Comparison of CCD and
CMOS
2.2 Correction Factor for Direct Reflective Photoelectric
Sensor
3.1 Gantt Chart for FYP I
3.2 Gantt Chart for FYP II
4.1 Step for PLC Configuration
4.2 Configuration of PC Access
4.3 GUI Configuration of Labview
4.4 Image Processing Configuration on VBAI
4.5 Rhino Robot Configuration using RoboTalkWindow
5.1 Condition for Combination of Colour and Shape
18
35
53
54
59
61
62
64
66
70
x
LIST OF FIGURES
2.1 Basic Functions of Machine Vision System
2.2 The Coordinate Convention of Pixels That Represent Digital
Images
2.3 The Basic Components of a Vision System
2.4 The Example of CCD Image Sensor
2.5(a) CCD Pixel
2.5(b) Architecture
2.6 Basic Principle of The Charge-Coupled Device
2.7 CMOS Sensor is Implemented in Low-End Web-Cam
2.8 CMOS Vision Sensor Floorplan
2.9 Schematic of The RGB Colour Cube
2.10 The Level of Histogram of Gray Scales
2.11 Calculating The Median Value of a Pixel Neighbourhood
2.12 A 3 × 3 Region Mask
2.13 Prewitt Horizontal and Vertical Operators
2.14 Sobel Convolution Kernels
2.15 The Computation Method of a Pixel
2.16 Roberts Cross Convolution Masks
2.17 The Closed-Loop Control System
2.18 Cartesian Robot
2.19 Cylindrical Robot
2.20 Spherical Robot
2.21 Articulated Robot
2.22 Traditional Conveyor
2.23 Steps in Analog-to-Digital Conversion of Continuous Analog
Signals from Process
2.24 Thru-Beam Photoelectric Sensor Detection
2.25 A Reflection with Reflector Photoelectric Sensor Detection
Method
2.26 A Direct Reflection Photoelectric Sensor Detection Method