University of Duhok Image Processing Lab

College of Engineering Class: Fourth Year

Electrical and Computer Department 2019-2020

Experiment No. 5
Spatial Domain Processing – Image Histogram

Object:
The purpose of this experiment is to develop and illustrate MATLAB formulations representative of
processing techniques in spatial domain.

Introduction:
The term spatial domain refers to the image plane itself, and methods in this category are based on direct
manipulation of pixels in an image.

Procedure:
1. Image Complement: This process is used for obtaining a photographic negative, it is particularly
useful for enhancing white or gray detail embedded in a large, predominantly dark region.

2. IM = imcomplement(MyImage)

Example:
MyImage = imread(‘D:\cameraman.jpg’);
IM = imcomplement(MyImage);
imshow(MyImage) figure, imshow(IM)

(1 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

Image Complement

3. Image histograms: An image histogram is a chart that shows the distribution of intensities in an
image. It plays a basic role in image processing, in areas such as enhancement, compression,
segmentation, and description.
h = imhist(MyImage) Example:
MyImage = imread(‘D:\cameraman.jpg’);
Figure, imhist(MyImage)

Image histogram

(2 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

There are other ways to plot the image histogram, depending on image processing applications.
bar graph:
bar(horz, vert)
where vert is a row vector containing the points to be plotted,
horz is a vector of the same dimension as vert that contains the increments of the horizontal scale. When
plotting a bar graph, it is customary to reduce the resolution of the horizontal axis by dividing it into bands

Example:
MyImage = imread(‘D:\cameraman.jpg’); h = imhist(MyImage) horz =
1:10:256; % reduce the horizontal resolution by 10 vert =
h(1:10:256); bar(horz, vert) axis([0 255 0 2000]) set(gca, 'xtick',
0:50:255) set(gca, 'ytick', 0:250:2000)

Bar Histogram

stem graph: is similar to a bar graph.

stem(horz, vert)

(3 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

Example:
MyImage = imread(‘D:\cameraman.jpg’);
h= imhist(MyImage)
horz = 1:10:256; % reduce the horizontal resolution by 10

vert = h(1:10:256);

stem(horz, vert) axis([0 255 0 2000]) set(gca, 'xtick', 0:50:255)

set(gca,'ytick', 0:250:2000)

(4 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

Stem Histogram

plot: The set of points are plotted by linking them with straight lines.
Example:
MyImage = imread(‘D:\cameraman.jpg’);
h = imhist(MyImage) plot(h) % Use
the default values.
axis([0 255 0 2000])
set(gca,'xtick', 0:50:255)
set(gca,'ytick', 0:250:2000)

Plot Histogram

(5 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

4. Histogram Equalization: it is the technique for adjusting image intensities to enhance contrast by
spreading the levels of the input image over a wider range of the intensity scale. It is used for images that
have poor contrast, images that look like they're too dark, or if they're too bright. g = histeq(f, nlev) where f
is the input image and nlev is the number of intensity levels specified for the output image.

Example:
f = imread(‘D:\cameramandark.jpg’);
imshow(f) figure, imhist(f)
ylim('auto') g = histeq(f, 256); figure,
imshow(g) figure, imhist(g)
ylim('auto')

(6 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

Histogram Equalization

(7 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

5. Histogram Matching (Specification): It is an automatic enhancement technique in which the

required histogram is derived from a user-specified target distribution. g = histeq(f, hspec) where f is the
input image, hspec is the specified histogram.
The target histogram distribution hspec will be either corresponding to a specified mathematical function
with the correct properties or will extract from an existing reference image.
Example:
I = imread(' D:\cameramandark.jpg’);
hspec = 0:255; %Define ramp like pdf as desired output histogram
Im = histeq(I, hspec); %Supply desired histogram to perform matching
subplot(2,3,1), imshow(I); subplot(2,3,2), imshow(Im); subplot(2,3,3),
plot(hspec); subplot(2,3,4), imhist(I); subplot(2,3,5), imhist(Im);

Histogram Matching

(8 - 5)
University of Duhok Image Processing Lab
College of Engineering Class: Fourth Year
Electrical and Computer Department 2019-2020

Lab Work:
Execute all four Image Histograms ways to plot the image cameraman dark and see the
differences between it and the original cameraman image

Define a (sign) like PDF as the desired output histogram to enhance (cameramandak) image Hint

use linspace to generate lineary spaced vector from 0 to 3.14

I = imread('D:\cameramandark.jpg');
hspec1 = linspace(0,3.14,256); %Define ramp like pdf as desired output histogram hspec
= sin(hspec1);
Im = histeq(I, hspec); %Supply desired histogram to perform matching
subplot(2,3,1), imshow(I); subplot(2,3,2), imshow(Im);
subplot(2,3,3), plot(hspec); subplot(2,3,4), imhist(I); subplot(2,3,5),
imhist(Im);

(9 - 5)