IMAGE PROCESSING TECHNIQUES FOR THE

ENHANCEMENT OF DIFFERENT TYPES OF IMAGES

PROJECT SYNOPSIS
OF MAJOR PROJECT

BACHELOR OF TECHNOLOGY
ELECTRONICS AND COMMUNICATION ENGINEERING

Under the Guidance of

Mr. Abhishek Panda
(ASSISTANT PROFESSOR)

Submitted by:
1. Priyanka Mishra (152004)
2. Divya Sharma (152062)
3. Priyanka Kumari (152015)
4. Nisha Kumari (152001)

January 2019

NETAJI SUBHAS INSTITUTE OF TECHNOLOGY,

BIHTA, PATNA.
Introduction:

In this project, the contrast of the image has to be changed with the help of histogram
equalization. Histogram equalization is used to enhance contrast. It is not necessary that
contrast will always increase in this. There may be some cases where this technique goes
wrong. This method usually increases the global contrast of many images, especially when
the usable data of the image is represented by close contrast value. Through this
adjustment, the intensities can be better distributed on the histogram. This allows for areas
of local contrast to gain a higher contrast. Histogram equalization accomplishes this by
effectively spreading out the most frequent intensities values.

This method is useful in image with background and foreground that are both bright or
both dark. In particular, the method can leads to better views of bones structures in x-rays
images.

Contrast is the difference between the maximum and minimum pixel intensity.

There are basically two methods that are used to enhance the contrast of an image, the
first one is the contrast stretching and the other is the histogram equalization.

Noise in an image:

Real world signals usually contains departures from the ideal signal that would be
produced by the model of the signal production process. Such departure are known to be
as noise.

Many image processing packages contain operators to artificially add noise to an image.
Deliberately corrupting an image with the noise allows the user to test the resistance of an
image processing operator to noise and assess the performance of various noise filters.

Noise are generally be grouped into two categories:

1. Independent noise.

2. Noise which is dependent on the image data.

Image independent noise can often be described by an additive noise model, where the
recorded image f(i,j) is the sum of the true image s(i,j) and the noise n(i,j):

f(i,j) = s(i,j) + n(i,j)

The noise n(i,j) is often zero-mean and described by its variance σn^2, the impact of the
noise on the image is often described by signal to noise ratio (SNR), which is given by:

SNR = (σs/σn) = [(σf/σn)^2-1]^1/2

The other is the image dependent noise, these types of noise are more complex than the
data independent noise, so it is called as those noise are assumed to be the data of the
image taken.

Sources of image noise:

• Error occurs in image signal while an image is sent electronically from one place
to another.
• Sensor heat while clicking an image.
• ISO factor, an ISO number indicates that how quickly a camera’s sensor absorbs
light higher ISO used mare chance of noticeable noise.
• Memory cell failure.

Types of noise:

• Salt and pepper noise.

• Gaussian noise.
• Speckle noise.
• Uniform noise.

Salt and pepper noise: It is also known as short noise, impulse noise and spike noise. Its
appearance is randomly scattered white or black or both pixels over the image. There are
only two possible values exists that is a and b and the probability of each is less than
0.2.

Reason for salt and pepper noise:

• By memory cell failure.

• By malfunctioning of camera’s sensor cells.
• By synchronization errors in image transmission.

Filtering techniques:

• It can be filtered by mean filtering.

 • By using median filtering.
• By using Gaussian filtering.

Gaussian noise: It is caused by random fluctuations in the signals, its modelled by

random values and to an image. This noise has a probability density function(PDF) of
the normal distribution. It is also known as Gaussian distribution.

P(z) = (1/√2πσ)*e^-(z-μ)^2/2σ^2

Sources of gaussian noise:

• In digital image arises during acquisition.

• Transmission.

Filtering techniques:

• Mean filtering.
• Median filtering.
• Gaussian filtering.

Speckle noise: This noise can be modelled by random values multiplied by pixel values
of an image.

g(n,m) = f(n,m)*u(n,m) +∑(n,m)

where g(n,m) is the observed image, u(n,m) is the multiplicative component and ∑(n,m)
is the additive component of the speckle noise.

Statement about the problem:

Given a grayscale image, its histogram consists of the histogram of its grey levels; that is,
a graph indicating the number of times each grey level occurs in the image. We can infer
a great deal about the appearance of an image from its histogram, as the following
examples indicate:
• In a dark image, the grey levels (and hence the histogram) would be clustered at
the lower end.
• In a uniformly bright image, the grey levels would be clustered at the upper end.
• In a well contrasted image, the grey levels would be well spread out over much
of the range.
Why this particular topic is chosen:

A Graphical User Interface (GUI) is a set of techniques and mechanism used to create an
interactive communication between a program and a programmer. GUI eliminates the need
to learn a language or type commands to run the application by providing point and click
control of software application.

Objective and scope of the project:

An image lacks contrast when there are no sharp differences between black and white.
Brightness refers to the overall lightness and darkness of the image.

In the process of adjusting the contrast, pixel values below the specified values are
mapped to be black and the value above the specified value is mapped to be white. The
result is a linear mapping of a subset of pixel values to the entire range of display
intensities. This produces an image of higher contrast by darkening pixels whose value is
below a specified value and lightening the pixels above that specified value.

Methodology:

To enhance the contrast of the selected image, a technique called histogram equalization is
used. Suppose the image has L different gray level 0, 1, 2,......, (L-1) and the gray level i
occurs ni times in the image. Let the total number of pixels in the image is n.

So, to change the contrast of the image, the formula given below is used:

(n0+n1+n2+.......+ni) (L-1)
n

and this number is rounded by the nearest integer.

There are many methods used to enhance the contrast of the image, some of those
methods are mentioned below:

1. classical histogram equalization(CHE): The CHE is a global operation ,here the

equalization is applied to whole image. For a given image X, the probability density
function P(Xk) is defined as:

P(Xk) = n^k/n
For k = 0, 1 , 2 , 3..., (L-1), n^k represents the number of times that the level (Xk) appears
in the input image X and n is the total number of samples in the input image. The
cumulative density function is defined as:

C(X) = ∑j=0 to k P(Xj)

Transform function f(x) based on the cumulative density function as

f(x) = X0 + (XL-1-X0)C(x)

then the output image is expressed as

Y = f(X)

Algorithm for the CHE:

➢ Start the program.

➢ Read the image from the current folder.
➢ Find the size of the image.
➢ Get the histogram of the image.
➢ Calculate new values via general histogram equalization formula.
➢ Built new image by replacing original gray value with the new gray value.
➢ Get the histogram of the new image.
➢ Stop.

Software to be used:

This project is done only on software named MATLAB with version MATLAB 8.1 release
R2013a. There is no use of hardware for its implementation.

What contribution would the project make:

The contrast enhancement plays a vital role in this era. The contrast enhancement gives a
good visualization and a better finishing.

• It is used in thresholding.
• It gives a better view of bones structure in an x-ray image.
• Histogram equalization often produces unrealistic effect in photographs.
• It is very useful for scientific images like thermal, satellite images, x-rays images.
Advantages:

A key advantage of the method is that it is a fairly straight forward technique and an
invertible operator. So, in theory, if the histogram equalization function is known, then the
original histogram can be recovered.

Disadvantages:

A disadvantage of this method is that it is indiscriminate. It may increase the contrast of

background noise, while decrease the usable signal.