Lecture 03

Image Enhancement
Histogram Equalization
Contents

In this lecture we will look at image enhancement point

processing techniques:
 What is histogram?
 Histogram equalization
 Histogram matching

Image Enhancement-Histogram Equalization 2

What is the histogram of a digital image?

 The histogram of a digital image with gray values

r0 , r1 ,, rL1 is the discrete function

nk
p(rk ) 
n
nk: Number of pixels with gray value rk
n: total Number of pixels in the image

 The function p(rk) represents the fraction of the total

number of pixels with gray value rk.

Image Enhancement-Histogram Equalization 3

What is the histogram of a digital image?

 Histogram provides a global description of the

appearance of the image.

 If we consider the gray values in the image as

realizations of a random variable R, with some
probability density, histogram provides an
approximation to this probability density. In other
words,

Pr( R  rk )  p(rk )

Image Enhancement-Histogram Equalization 4

Some typical histograms

 The shape of a histogram provides useful information

for contrast enhancement

Dark image

Image Enhancement-Histogram Equalization 5

Some typical histograms

Bright image

Low contrast
image

Image Enhancement-Histogram Equalization 6

Some typical histograms

High contrast image

Image Enhancement-Histogram Equalization 7

Histogram equalization

 The histogram equalization is an approach to enhance

a given image. The approach is to design a
transformation T(.) such that the gray values in the
output is uniformly distributed in [0, 1].

 Let us assume for the moment that the input image to

be enhanced has continuous gray values, with r = 0
representing black and r = 1 representing white.

 We need to design a gray value transformation s =

T(r), based on the histogram of the input image, which
will enhance the image.

Image Enhancement-Histogram Equalization 8

Histogram equalization

we assume that:
 T(r) is a monotonically
increasing function for
0 r 1 (preserves
order from black to
white).
 T(r) maps [0,1] into
[0,1] (preserves the
range of allowed Gray
values).

Image Enhancement-Histogram Equalization 9

Histogram equalization

 Let us denote the inverse transformation by r T -1(s) .

We assume that the inverse transformation also
satisfies the above two conditions.

 We consider the gray values in the input image and

output image as random variables in the interval [0, 1].

 Let pin(r) and pout(s) denote the probability density of

the Gray values in the input and output images.

Image Enhancement-Histogram Equalization 10

Histogram equalization

 If pin(r) and T(r) are known, and r T -1(s) satisfies

condition 1, we can write (result from probability
theory): dr  
pout ( s)   pin (r ) 
 ds  r T 1 ( s )
 One way to enhance the image is to design a
transformation T(.) such that the gray values in the
output is uniformly distributed in [0, 1], i.e. pout (s)  1,
0  s 1
 In terms of histograms, the output image will have all
gray values in “equal proportion” .
 This technique is called histogram equalization.

Image Enhancement-Histogram Equalization 11

Histogram equalization

 Next we derive the gray values in the output is

uniformly distributed in [0, 1].
 Consider the transformation
r
s  T (r )   pin ( w)dw , 0  r 1
0

 Note that this is the cumulative distribution function

(CDF) of pin (r) and satisfies the previous two
conditions.
 From the previous equation and using the fundamental
theorem of calculus,
ds
 pin (r )
dr

Image Enhancement-Histogram Equalization 12

Histogram equalization

 Therefore, the output histogram is given by

 1 
pout ( s )   pin (r )    1r T 1 ( s )  1, 0  s 1
 pin (r )  r T 1 ( s )

 The output probability density function is uniform,

regardless of the input.
 Thus, using a transformation function equal to the CDF
of input gray values r, we can obtain an image with
uniform gray values.
 This usually results in an enhanced image, with an
increase in the dynamic range of pixel values.

Image Enhancement-Histogram Equalization 13

How to implement histogram equalization

 Step 1:For images with discrete gray values, compute:

nk
pin (rk )  0  rk  1 0  k  L 1
n
L: Total number of gray levels
nk: Number of pixels with gray value rk
n: Total number of pixels in the image
 Step 2: Based on CDF, compute the discrete version
of the previous transformation :
k
s k  T (rk )   pin (r j ) 0  k  L 1
j 0

Image Enhancement-Histogram Equalization 14

Example

 Consider an 8-level 64 x 64 image with gray values (0,

1, …, 7). The normalized gray values are (0, 1/7, 2/7,
…, 1). The normalized histogram is given below:

 NB: The gray values in output are also (0, 1/7, 2/7, …, 1).

Image Enhancement-Histogram Equalization 15

Example

# pixels Fraction
of # pixels

Gray value Normalized gray value

Image Enhancement-Histogram Equalization 16

Example
k
 Applying the transformation,s k  T (rk )   pin (r j ) we have
j 0

Image Enhancement-Histogram Equalization 17

Example

 Notice that there are only five distinct gray levels ---
(1/7, 3/7, 5/7, 6/7, 1) in the output image. We will
relabel them as (s0, s1, …, s4 ).
 With this transformation, the output image will have
histogram

Image Enhancement-Histogram Equalization 18

Example
Histogram of output image

# pixels

Gray values
 Note that the histogram of output image is only
approximately, and not exactly, uniform. This should not be
surprising, since there is no result that claims uniformity in
the discrete case.
Image Enhancement-Histogram Equalization 19
Example Original image and its histogram

Image Enhancement-Histogram Equalization 20

Example
Histogram equalized image and its histogram

Image Enhancement-Histogram Equalization 21

Example

Image Enhancement-Histogram Equalization 22

Example

Image Enhancement-Histogram Equalization 23

Example

Image Enhancement-Histogram Equalization 24

Comments

 Histogram equalization may not always produce

desirable results, particularly if the given histogram is
very narrow. It can produce false edges and regions. It
can also increase image “graininess” and “patchiness.”

Image Enhancement-Histogram Equalization 25

Comments

Image Enhancement-Histogram Equalization 26

Histogram matching
 Histogram equalization yields an image whose pixels
are (in theory) uniformly distributed among all gray
levels.
 Sometimes, this may not be desirable. Instead, we
may want a transformation that yields an output image
with a pre-specified histogram. This technique is called
histogram specification.
z=H(r)

Input Uniform Output

image s=T(r) image v=G(z) image

Image Enhancement-Histogram Equalization 27

Histogram matching
 Suppose, the input image has probability density in p(r). We
want to find a transformation z H (r) , such that the
probability density of the new image obtained by this
transformation is pout(z) , which is not necessarily uniform.
 First apply the transformation
r
s  T (r )   pin ( w)dw , 0  r 1 (*)
0
This gives an image with a uniform probability density.
 If the desired output image were available, then the
following transformation would generate an image with
uniform density:
z
v  G ( z )   pout ( w)dw, 0  z 1 (**)
0
Image Enhancement-Histogram Equalization 28
Histogram matching
 From the gray values nwe can obtain the gray values
z by using the inverse transformation, z  G-1(v)

 If instead of using the gray values nobtained from (**),

we use the gray values s obtained from (*) above (both
are uniformly distributed ! ), then the point
transformation
z=H(r)= G-1[ v=s =T(r)]

will generate an image with the specified density out

p(z), from an input image with density in p(r) !

Image Enhancement-Histogram Equalization 29

Histogram matching
 For discrete gray levels, we have
k
s k  T (rk )   pin (r j ) 0  k  L 1
j 0
k
vk  G ( z k )   pout ( z j )  sk 0  k  L 1
j 0

 If the transformation zk G(zk) is one-to-one, the

inverse transformation sk G-1 (sk) , can be easily
determined, since we are dealing with a small set of
discrete gray values.
 In practice, this is not usually the case (i.e., ) zk 
G(zk) is not one-to-one) and we assign gray values to
match the given histogram, as closely as possible.
Image Enhancement-Histogram Equalization 30
Algorithm for histogram matching

 Step 1: Equalize input image to get an image with

uniform gray values, using the discrete equation:
k
s k  T (rk )   pin (r j ) 0  k  L 1
j 0

 Step 2: Based on desired histogram to get an image

with uniform gray values, using the discrete equation:
k
vk  G ( z k )   pout ( z j )  sk 0  k  L 1
j 0

 Step 3:
1 1
  
z G ( v=s ) z G [T ( r )]

Image Enhancement-Histogram Equalization 31

Example

 Consider an 8-level 64 x 64 previous image.

# pixels

Gray value

Image Enhancement-Histogram Equalization 32

Example

 It is desired to transform this image into a new image,

using a transformation Z=H(r)= G-1[T(r)], with
histogram as specified below:

# pixels

Gray values

Image Enhancement-Histogram Equalization 33

Example

 Step 1: The transformation T(r) was obtained earlier

(reproduced below):

Image Enhancement-Histogram Equalization 34

Example

 Step 2: we compute the transformation G as before.

Image Enhancement-Histogram Equalization 35

Example

 Step 3: Computer z=G-1 (s=v),Notice that G is not

invertible.
G-1(0) = ?
G-1(1/7) = 3/7
G-1(2/7) = 4/7
G-1(3/7) = ?
G-1(4/7) = ?
G-1(5/7) = 5/7
G-1(6/7) = 6/7
G-1(1) = 1

Image Enhancement-Histogram Equalization 36

Example

 Combining the two transformation T and G-1 , compute

z=H(r)= G-1[v=s=T(r)]

Image Enhancement-Histogram Equalization 37

Example

 Applying the transformation H to the original image

yields an image with histogram as below:

 Again, the actual histogram of the output image does

not exactly but only approximately matches with the
specified histogram. This is because we are dealing
with discrete histograms.
Image Enhancement-Histogram Equalization 38
Example

Original image and its histogram

Histogram specified image and its histogram

Image Enhancement-Histogram Equalization 39
Example

Desired histogram

Image Enhancement-Histogram Equalization 40

Example

Image Enhancement-Histogram Equalization 41

Example

Image Enhancement-Histogram Equalization 42

Example

Image Enhancement-Histogram Equalization 43

Example

Image Enhancement-Histogram Equalization 44

Example

Image Enhancement-Histogram Equalization 45

Summary

 We have looked at histogram equalization and

histogram matching

 Next time we will start to look at neighbourhood

operations – in particular filtering and convolution

Image Enhancement-Histogram Equalization 46