SPECTRAL CONTRAST ENHANCEMENT Course: Introduction to RS & DIP Mirza Muhammad Waqar Contact: [email protected] +92-21-34650765-79 EXT:2257 RG610.

SPECTRAL CONTRAST ENHANCEMENT

Course: Introduction to RS & DIP

Mirza Muhammad WaqarContact:

[email protected]+92-21-34650765-79 EXT:2257

RG610

2

Contents

Geographical Information System Remote Sensing & Satellite Image Processing Color Space Landsat 7 spectral bands Spectral Reflectance Curves Image Interpretation Spectral Ratioing Indices

3

Overview

One of the strength of image processing is that it gives us the ability To enhance the view of an area by manipulating the

pixels value.

Contrast enhancement does not change the values in the image rather simply adjust the colors associated with these color values.

Image Enhancement

The alteration of the appearance of an image in such a

way that the info contained in that image is more

readily interpreted visually in terms of a particular

need

It alters the visual impact of the image to improves

the info contents for the interpreter These operations improve the interpretability of an

image by changing the contrast between the features in the scene

To improve the appearance of an image for human visual analysis

No single standard method can be said to be the

best, it depends upon the need of the user

The characteristic of each image in terms of

distribution of pixel values over 0-255 range will

change from one area to another , thus

enhancement tech suited for one image may not

be good for other image covering different type of

area

Image Enhancement

Image Histogram

Histogram greatly helps to deduce the

appearance of an image

In a dark image, the gray levels would be clustered

towards the lower end

In a uniformly bright image, the gray levels would be

clustered towards the upper end

In a well contrasted image, the gray levels would be well

spread out over much of the range

Image Enhancement

Methods of improving visual interpretability of an

image

By altering the contrast of an image ( contrast stretching)

Converting from black and white to color representation

Contrast is simply the range and the distribution

of the pixel values over the 0-255 gray scale

Perception of Colors

Conversion to color is desirable as the eye is more sensitive to variations in hue than change in the brightness

Contrast Enhancement/ Stretching

Sensors record reflected or emitted radiant flux exiting from earth

surface materials

Ideally one material would reflect tremendous amount of energy in a

certain wavelength while another much less in the same wavelength

This would result in contrast between the two types of materials

In some cases different materials would often reflect similar amount

of radiant flux throughout the visible and IR portion of EM spectrum

resulting in a relatively low contrast image

Contrast Enhancement

Sensor on board have to be capable of detecting upwelling

radiance levels ranging from low (from oceans) to very high

(over snow)

For particular area to be imaged ,it is unlikely that full

dynamic range of the sensor will be used ,thus the

corresponding image is dull or over bright-over or under

exposed

11

Why we need Contrast Enhancement

Quite often the useful data in a digital image populate only a small portion of available range of digital values. Commonly 8 bit or 256 levels

Contrast enhancement involves changing the original values so that more of the available range is used.

It increases the contrast among the features and their background.

12

Types of Contrast Enhancement

1. Histogram Equalization Stretch2. Standard Deviation Stretch3. Gaussian Stretch4. Gamma Correction5. Level Slice6. Constant Value7. Invert Stretch8. Percentage LUT9. Piecewise Linear Contrast Stretch10. Linear Stretch11. Logarithm Stretch

13

Linear Contrast Enhancement

This technique involves the translation of the image pixel values from the observed range of digital number to the full range of the display device (e.g. 8 bit)

LINEAR STRETCH-MIN MAX

134 179 96 140 153

41 13 0 32 51

130 165 100 135 145

57 35 25 50 65

180 215 135 200 205

205 205 225 220 225

30 25 25 120 205

The uniform expansion of the of input digital numbers to full range )-255) is called linear stretch

BV OUT=255(BV IN-MIN) / (MAX-MIN)

MIN=25, MAX=225

16

Histogram Equalization Stretch

This stretch assign more display values (range) to the frequently occurring portion of the histogram.

In this way, the detail in those areas will be better enhanced having high frequency relative to those areas having low frequency value in the histogram.

Histogram Conversion

The histogram of the original image is converted to

other types of histograms as specified by user

Histogram Stretch- Image values are assigned to the

display levels on the basis of their frequency of occurrence

More display values ( more radiometric details) are

assigned to the frequently occurring portion of histogram

Special Stretch- To analyze specific features in greater

radiometric detail s by assigning the display range

exclusively to a particular range of image values

Histogram Stretch

Histogram Equalization

After

Before

20

Standard Deviation Stretch

Standard deviation stretch trim all pixels that have a digital number beyond the range the defined standard deviation; Then perform the linear stretch for the remaining

pixels

Standard Deviation 1: 67 % Standard Deviation 2: 95 % Standard Deviation 3: 99 %

21

Gaussian Stretch

This histogram involve the fitting of the observed histogram to normal or Gaussian histogram.

This stretch adjust the range of lookup table values so that the output histogram is approximately a normal distribution.

22

Level Slice Stretch

It will slice the input image into user defined number of classes.

The output image will have only limited number of variations depending upon the user defined number of classes.

Density/ Level Slicing

Representation of a range of contiguous gray levels of gray scale image by a single color

Used to separate the data into “n” intervals or “slices” based on the histogram from one wavelength band.

All data within a slice are displayed as one digital number or color in the output image

The Gray level in the output image corresponds to the number of slices

Used frequently with thermal images, i.e. different temperature ranges can be shown with different slices

Level Slicing

After

Before

25

Invert Contrast Stretch

This contrast enhancement technique invert the current lookup table values.

This has the effect of producing a photographic negative of the image.

This technique is often used to extract information from the shadow.

Gray-Level Thresholding

Used to “segment "an input image into two classes. Purpose is to develop a binary mask for one category, so that

processing can be applied to each class independently

Original NIR Image

Mask Image for Water

Set threshold here

Contrast Enhancement

If the range of gray levels could be altered so as

to fit the full range of the black and white axis,

then the contrast between the dark and bright

areas of the image would be improved

Does not modify the original data unless new file

is saved

Questions & Discussion