7 elements of remote sensing process 1.Energy Source (A) 2.Radiation & Atmosphere (B) 3.Interaction with Targets (C) 4.Recording of Energy by Sensor (D)

7 elements of remote sensing process

1. Energy Source (A)

2. Radiation & Atmosphere (B)

3. Interaction with Targets (C)

4. Recording of Energy by Sensor (D)

5. Transmission & Reception (E)

6. Interpretation and Analysis (F)

7. Application (G)

Extract meaningful information from imagery

6. Interpretation and Analysis (F) - the processed image is interpreted, visually and/or digitally, to extract information about the target which was illuminated.

4.3 Digital Image Processing

Common image processing image analysis functions:

A. Preprocessing

B. Image Enhancement

C. Image Transformation

D. Image Classification and Analysis

Background

• DIP - manipulation & interpretation of

images

• Began in 1960’s

• 1972 - Landsat 1 launched

• Access to low cost, efficient computers

• Access to imagery

Digital spatial image - made up of a grid of cells, each containing a value or measurement and representing an area of the Earth’s surface.

Pixel

Digital Number (DN) - value stored within a pixel

of an image, represents amount of light

reflected back to sensor.

digital format – images are represented in a

computer as arrays of pixels.

Multispectral images - multiple layers representingdifferent parts of the EMS.

4.2 Elements of Visual Interpretation

• Identifying targets

– Based on how they reflect/emit radiation

• Based on;

– Visual elements – tone, shape, pattern,

texture, shadow, association.

4.3 Digital Image Processing

Common image processing image analysis functions:

A. Preprocessing

B. Image Enhancement

C. Image Transformation

D. Image Classification and Analysis

1. Pre-Processing (Image Rectification)

• Initial processing of raw data prior for analysis

• Correct for distortion due to characteristics of

imaging system & imaging conditions.

1. Pre-Processing (Image Rectification)

• Procedures include:

a. geometric correction - correct for geometric distortion due

to Earth's rotation, curvature, platform motion, relief

displacement, (such as oblique viewing).

b. radiometric correction - correct for uneven sensor

response over image, random noise, atmosphere.

c. geo-referencing - ground control points (GCP's) used to

register image to a precise map.

2. Image Enhancement

• Solely to improve appearance of imagery.

• Increasing visual distinction

• Un-enhanced images usually appear very dark -

little contrast - difficult to visually interpret.

• Various procedures applied to image data in order to

more effectively display data for visual

interpretation.

2. Image Enhancement

A. Contrast stretching

– Histograms

– Increase tonal distinction

B. Spatial filtering

– Enhance/suppress features

A. Contrast stretching

• Radiometric enhancement - manipulate

brightness and contrast of pixels to amplify

differences between features.

• Changes made to pixels without

consideration of values of surrounding

pixels.

– adjust brightness and contrast controls

– apply preset contrast stretches

– manually adjusting image histograms

A. Contrast stretchingRadiometric Enhancement

• Not all values will be used or spread out to fill the entire range of 256 values.

• Need to manipulate the relative brightness and contrast of the pixels to amplify the differences between features.

Lanier.img (4-3-2)

Swipe

• Computers - ideal for manipulating

and analyzing large continuous

data sets displayed as grayscale.

• Used to distinguish between slight

spectral variations and enhance

them.

Landsat 7 image with no contrast stretching - histogram for the near infrared

band.

Some features, like agricultural areas, can be distinguished.

Applying a histogram stretch produces a simple classification of urban, agricultural, and mixed use areas.

A. Contrast stretchingRadiometric resolution

• Dynamic range or number of possible data

values (Digital numbers) in each band of

the image.

• The range of DN’s is usually referred to by

the number of bits into which the recorded

energy is divided.

• 28 = 256 is most common

0 = black255 = white

A sensor measures the electromagnetic energy within its range.

Total intensity of the energy from zero to the maximum is broken down into 256 brightness values for 8-bit data.

A. Contrast stretchingLinear grey-level stretching

• Lower threshold value is chosen so that all pixel values

below threshold are mapped to zero.

• Upper threshold value is chosen so that all pixel values

above threshold are mapped to 255.

• All other pixel values are linearly interpolated to lie

between 0 and 255.

– Lower and upper thresholds are usually chosen to be values close to the

minimum and maximum pixel values of the image.

Two types of histogram stretches

Landsat TM image - Olympic Pennisula, NW Washington

Vis red band