ABSTRACT: to remotely sensed images with Gaussian or non-Gaussian

 This paper attempts to undertake the
study of two types of the contrast enhancement. Contrast enhancement procedures
can be grouped into two broad categories. Namely, linear and non-linear. In
linear enhancement techniques, the mathematical implementation is on the basis
of uniform slope, while in non-linear method, the enhancement is done using a
varying slope. In linear enhancement techniques, the commonly used the min-max
stretch, percentile stretch and piecewise, while non-linear techniques are
histogram equalization, adaptive equalization, log arithmetic contrast

linear contrast enhancement, non-linear contrast enhancement

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now


         The operation of contrast enhancement brings
about a change in the original values so that the full range of the brightness
scale may be utilized. This helps in increasing the distinguishability between
objects and their background. It is observed that contrast stretch is a
subjective in nature and hence often a trial and error procedure. One of the
most important quality factors in satellite images comes from its contrast.
Contrast is created by the difference in luminance reflected from two adjacent
surfaces. In visual perception, contrast is determined by the difference in the
color and brightness of an object with other objects. The contrast enhancement
has been two types,

 They are,

Linear contrast enhancement

Non-linear contrast enhancement


       The type referred a contrast stretching,
linearly expands the original digital values of the remotely sensed data into a
new distribution. By expanding the original input values of the image, the
total image of sensitivity of the display device can be utilized. Linear
contrast enhancement also makes subtle variations within the data obvious
.these  types of  enhancement are best applied to remotely
sensed images with Gaussian or non-Gaussian histograms, meaning, all the
brightness values fall within a narrow range of histogram and only mode is apparent.
There are three methods of linear contrast enhancement.

linear contrast enhancement


       This technique is simplest enhancement. When
using the minimum-maximum linear contrast the original minimum and maximum
values of the data are assigned a newly specified set of values that utilize
the full range of available brightness values. The min-max contrast stretch or
enhancement greatly improves the contrast of most of the original brightness values,
but, there is loss contrast of the extreme high and low ends. This method is
applying with respect to image application type,



a.     Percentile linear contrast


       Many times, it is observed that the tails of
input histogram are long.

   This virtually leads to no improvement in
the image quality even after min-max stretch has been performed. This can be
carried out on a percentage basis, where 1% or %5 of the data on both the ends
of cut off. A standard deviation from the mean is often used to push the tails
of the histogram beyond the original minimum and maximum values.


b.    Piece-wise linear contrast


The above two techniques of linear enhancements are suitable when the
histogram of the input image is unimodel in nature. It has only one the peak.
However, this may not be normal case, and histogram is multi-model in nature
indicating that there are more than one object in the image. A piece-wise
linear contrast occupying difference gray scale ranges can be enhanced.



 Nonlinear contrast enhancement often involves
histogram equalization through the use of an algorithm .the nonlinear contrast
stretch method has one major disadvantage. Each value in the input images can
have several values in the output images, so that objects in the original scene
lose their correct relative brightness value. There are three methods of
nonlinear contrast enhancement.


Histogram equalization


 Histogram equalization is widely
used for contrast Manipulation in digital image processing due it is simple implementation.
It requires minimum information from the analyst. In this enhancement, the
original histogram is readjusted to produce a uniform population density of
pixels along the horizontal grey value axis. The technique ensures that the
entropy of the images an indicator of the image information is increased. In
this method, first of all the target number of pixel in each brightness level
of the equalized histogram is calculated by the dividing the total number of
pixels in the images with the total number of brightness levels is designed.




         Adaptive histogram equalization where you can
divide the image into several rectangular domains, compute an equalizing
histogram and modify levels so that they match across boundaries. Depending on
the nature of the nonlinear uniformly of the image. Adaptive histogram
equalization uses the histogram equalization mapping function supported over a
certain size of local window to determine each enhanced value. It acts of local

contrast enhancement


 This transformation is a very
useful nonlinear contrast enhancement. Here the output pixel brightness values
will be generated from input pixel brightness values following some logarithmic

log (BVin)

This transformation only highlights
then features lying in the darker region of the histogram. Lists same of the
advantages and important characteristics of logarithmic contrast enhancement,
as given below.


Ø It
enhances low contrast edges, thus making low contrast details more distinct

Ø It
provides a contrast to signal to noise ratio.

Ø It
usually provides a more equal distribution of grey values.

Ø It
transforms multiplicative noise into additive noise.




    In this paper, studies take the contrast
enhancement and its types. The first contrast enhancement is linear contrast enhancement
another enhancement is nonlinear contrast enhancement .in first enhancement
from the linear contrast enhancement studies are explained and experiments
max-min methods, percentile contrast method and piecewise contrast method the
best method in linear contrast enhancement. And second case of enhancement
study is explained and experiments are carried out for different techniques in
histogram equalization method, adaptive equalization and logarithmic contrast
enhancement is the best study is remote processing techniques in the nonlinear
contrast enhancement.