Scribd
Upload
48 views5 pages

Adaptive Histogram

Adaptive histogram is an image processing technique that increases contrast locally by splitting images into blocks and calculating histogram equalization for each block separately. This improves local contrast and edge definition in individual image areas. The technique computes histograms for individual parts of an image, conserves edges within local areas, and improves contrast locally. Adaptive histogram equalization applies a modification function to each pixel based on the histogram of a square neighborhood around it.

Uploaded by

Muhammad Abbas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
48 views5 pages

Adaptive Histogram

Adaptive histogram is an image processing technique that increases contrast locally by splitting images into blocks and calculating histogram equalization for each block separately. This improves local contrast and edge definition in individual image areas. The technique computes histograms for individual parts of an image, conserves edges within local areas, and improves contrast locally. Adaptive histogram equalization applies a modification function to each pixel based on the histogram of a square neighborhood around it.

Uploaded by

Muhammad Abbas
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Adaptive Histogram

Introduction:
Adaptive histogram is a digital image processing technique employed to increase the
contrast of images. Adaptive Histogram is different from Normal Histogram equalization in
some manners that adaptive histogram increases the contrast locally. It splits the image into
separate blocks and calculates histogram equalization for every portion. Thus, Adaptive
Histogram Equalization calculates many histograms, each one equivalent to a individual part of
the image. It improves the local contrast and definitions of edges in all individual areas of the
image.

Benefits using Adaptive Filtering:


1. It computes the Histogram [1] Equalization of individual parts of the image.
2. It conserves the edges in individual areas of the image.
3. It improves the contrast locally.
Explanation:
This is enhanced by adaptive histogram, which applies a modification function to each pixel that
is obtained from the neighborhood area. It was initially created for use in cockpit displays for
aircraft. In its most basic form, as seen in the picture below, each pixel is converted based on the
histogram of a square surrounding it [2]. The transformation functions are derived from the
histograms in the same way as traditional histogram equalization. The neighborhood's pixel
values' cumulative distribution function is inversely proportional to the transformation function.

Because their neighborhood wouldn't totally be contained within the image, pixels close to the
image boundary need to be handled differently. This is true, for instance, of the pixels in the
illustration to the left of or above the blue pixel. The image can be made larger by mirroring the
pixel lines and columns regarding the image boundaries. The border's pixel lines should not be
copied directly because that would result in a neighborhood histogram with a sharp peak.
Properties:

1. One of the parameters of the method is the size of the neighborhood area. It makes up a
distinctive length scale where contrast is increased at smaller scales and decreased at
bigger scales.
2. The result value of a pixel under Adaptive Histogram Equalization is proportional to its
rank among the pixels in its neighborhood due to the nature of histogram equalization.
This makes it possible to implement the algorithm effectively on specialized hardware
that can compare the central pixel to every other pixel nearby. One way to calculate an
unnormalized result value is to add 2 for each pixel that has a smaller value than the
centre pixel and 1 for each pixel that has an equal value.
3. The transformation function will apply a small range of pixel values to the entire range of
the final image when the image region comprising a pixel's neighborhood is
homogeneous regarding intensities; as a result, the histogram of this region will be
strongly peak. Due to this, Adaptive Histogram Equalization overamplifies small levels
of noise in the image's relatively homogeneous regions. [2]

Here is the code for Adaptive Histogram:


This is the MATLAB code:
clear all
close all
clc
I = imread('pout.tif');
figure
subplot(1,2,1)
imshow(I)
subplot(1,2,2)
imhist(I,64)

This output shows original image and its histogram.

This is the original image.


This is the MATLAB code:

clear all
close all
clc
f = imread('pout.tif');
k = adapthisteq(f);
figure
subplot(1,2,1)
imshow(k)
subplot(1,2,2)
imhist(k,64)

This output shows the adaptive histogram on an image.


Conclusion:
Now the conclusion is that as I have utilized code on image so we can say that Adaptive
histogram is superior to ordinary Histogram when the image has particularly dark or bright spots.
But Adaptive histogram manages to overamplify the contrast in near-constant regions of the
image since the histogram in such regions is highly concentrated. As a result, Adaptive
histogram may cause noise to be amplified in the near-constant region. I have added explanation
of MATLAB code and output also.

References

[1] geeksforgeeks, "Adaptive Histogram Equalization in Image Processing Using MATLAB".

[2] Wikipedia, "Adaptive histogram equalization," [Online]. Available:


https://en.wikipedia.org/wiki/Adaptive_histogram_equalization.

[3] Mathworks, "Adaptive Histogram Equalization," Mathworks, [Online]. Available:


https://www.mathworks.com/help/images/adaptive-histogram-equalization.html.

You might also like