Digital Image Processing

LECTURE 04
IMAGE ENHANCEMENT-II
(HISTOGRAM PROCESSING)

Maryiam Zahoor
 A Note About Grey Levels

 So far when we have spoken about image grey level values we have said they are
in the range [0, 255]
 Where 0 is black and 255 is white
 There is no reason why we have to use this range
 The range [0,255] stems from display
 For many of the image processing operations in this lecture grey levels are
assumed to be given in the range [0.0, 1.0]
Image Histograms
 Image Histograms

 The histogram of an image shows us the distribution of grey levels in the image
 Massively useful in image processing, especially in segmentation
Frequencies

Grey Levels
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram Examples (cont…)

Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram Examples (cont…)

Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram Examples (cont…)

Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram Examples (cont…)

Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Histogram Examples (cont…)

 Histogram Examples (cont…)
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

 A selection of images and

their histograms
 Notice the relationships
between the images and
their histograms
 Note that the high contrast
image has the most
evenly spaced histogram
 11
Histogram Equalization

The intensity levels in an image may be viewed as

random variables in the interval [0, L-1].
Let pr (r ) and ps ( s) denote the probability density
function (PDF) of random variables r and s.
Histogram Equalization
Histogram Equalization
Background (Probability
Distribution)
Histogram Equalization
Histogram Equalization
Histogram Equalization
 Histogram Equalization(Summary)

 Spreading out the frequencies in an image (or equalising

the image) is a simple way to improve dark or washed
out images
 The formula for histogram
equalisation is given where
 rk: input intensity
 sk: processed intensity
 k: the intensity range
Discrete values:
(e.g 0 – 255)
k
 nj: the frequency of intensity j
sk  T (rk )  ( L  1) pr (rj )
 MN: number of pixels/ sum of all frequencies
j 0
k nj
L 1 k
 ( L  1)   nj k=0,1,..., L-1
j  0 MN MN j 0
Example 2
Example 2
Example 2
Example
Histogram Equalization-Examples
Histogram Equalization-Examples
Histogram Equalization-Examples
Histogram Equalization-Examples
Images taken from Gonzalez & Woods, Digital Image Processing (2002)

Function
Equalisation Transformation
Histogram Equalization-Examples
Histogram Equalization-Examples
Histogram Equalization-Examples
Histogram Equalization-Examples
Histogram Specification/Matching
Histogram Specification/Matching
Histogram Specification/Matching
Histogram Specification/Matching
Example
Cont.
Cont.
Cont.
Cont.
Histogram Matching Example
Histogram Matching
Example(continued)
Histogram Matching Example
(continued)
Local Histogram Processing

 Define a neighborhood and move its center from pixel to

pixel

 At each location, the histogram of the points in the

neighborhood is computed. Either histogram equalization or
histogram specification transformation function is obtained

 Map the intensity of the pixel centered in the neighborhood

 Move to the next location and repeat the procedure

Local Histogram Processing:
Example