Lecture 2. Intensity Transformation and Spatial Filtering

Lecture 2.
Intensity Transformation
and Spatial Filtering
Prof. Dr. Noorullah Shariff

Professor and Head
Department of Electronics and Communication Engineering
Secab Institute of Engineering and Technology
Vijayapura
Spatial Domain vs. Transform Domain
► Image Processing techniques in the

 Spatial domain directly manipulate pixels of a image
 Transform domain
► Transform a image into transform domain
► Doing the processing their
► Obtaining the inverse transform to bring back image in spatial
domain
10/22/20 2
Spatial Domain Process
g ( x, y )  T [ f ( x, y )])
f ( x, y ) : input image
g ( x, y ) : output image
T : an operator on f defined over
a neighborhood of point ( x, y)
10/22/20 3
g ( x, y )  T [ f ( x, y )])
f ( x, y ) : input image
g ( x, y ) : output image
T : an operator on f defined over
a neighborhood of point ( x, y)
► Works in the neighbourhood of every pixel
10/22/20 4
► Two principal image processing technique
classifications in spatial domain
 Intensity transformation methods
►Operate on single pixels
 Spatial filtering methods
►Works in the neighbourhood of every pixel
►Generally N X N where N > 1
► Intensity transformation can be considered

as special case of speial filtering with 1 x 1
10/22/20 5
10/22/20 6
10/22/20 7
Intensity transformation function

s  T (r )
10/22/20 8
Some Basic Intensity Transformation
Functions
10/22/20 9
Image Negatives
Image negatives
s  L 1 r
10/22/20 10
Example: Image Negatives
Small
lesion
10/22/20 11
Log Transformations
Log Transformations
s  c log(1  r )
10/22/20 12
Example: Log Transformations
10/22/20 13
Power-Law (Gamma) Transformations

s  cr
10/22/20 14
Example: Gamma Transformations
10/22/20 15
Cathode ray tube
(CRT) devices have an
intensity-to-voltage
response that is a
power function, with
exponents varying
from approximately
1.8 to 2.5
sr 1/2.5
10/22/20 16
10/22/20 17
10/22/20 18
Piecewise-Linear Transformations
► Contrast Stretching
— Expands the range of intensity levels in an image so that it spans
the full intensity range of the recording medium or display device.
► Intensity-level Slicing
— Highlighting a specific range of intensities in an image often is of
interest.
10/22/20 19
10/22/20 20
Highlight the major
blood vessels and
study the shape of the
flow of the contrast
medium (to detect
blockages, etc.)
Measuring the actual

flow of the contrast
medium as a function
of time in a series of
10/22/20
images 21
Bit-plane Slicing
10/22/20 22
Bit-plane Slicing
10/22/20 23
Bit-plane Slicing
10/22/20 24
Histogram Processing
► Histogram Equalization
► Histogram Matching
► Local Histogram Processing
► Using Histogram Statistics for Image Enhancement
10/22/20 25
Histogram Processing
Histogram h( rk )  nk
rk is the k th intensity value
nk is the number of pixels in the image with intensity rk
nk
Normalized histogram p( rk ) 
MN
nk : the number of pixels in the image of
size M  N with intensity rk
10/22/20 26
10/22/20 27
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.
10/22/20 28
s  T (r ) 0  r  L 1
a. T(r) is a strictly monotonically increasing function

in the interval 0  r  L -1;
b. 0  T (r )  L -1 for 0  r  L -1.
10/22/20 29
s  T (r ) 0  r  L 1
a. T(r) is a strictly monotonically increasing function

in the interval 0  r  L -1;
b. 0  T (r )  L -1 for 0  r  L -1.
T (r ) is continuous and differentiable.
ps ( s)ds  pr (r )dr
10/22/20 30
r
s  T (r )  ( L  1)  pr ( w)dw
0
ds dT (r ) d  r 
dr

dr
 ( L  1)

dr  0
pr ( w) dw

 ( L  1) pr ( r )
pr (r )dr pr (r ) pr (r ) 1
ps ( s )    
ds  ds   ( L  1) pr (r )  L  1
 
 dr 
10/22/20 31
Example
Suppose that the (continuous) intensity values
in an image have the PDF
 2r
 , for 0  r  L-1
pr (r )   ( L  1) 2
 0, otherwise

Find the transformation function for equalizing

the image histogram.
10/22/20 32
Example
r
s  T (r )  ( L  1)  pr ( w)dw
0
r 2w
 ( L  1)  dw
0 ( L  1) 2
2
r

L 1
10/22/20 33
Continuous case:
r
s  T (r )  ( L  1)  pr ( w)dw
0
Discrete values:
k
sk  T (rk )  ( L  1) pr (r j )
j 0
k nj L 1 k
 ( L  1)   nj k=0,1,..., L-1
j  0 MN MN j 0
10/22/20 34
Example: Histogram Equalization
Suppose that a 3-bit image (L=8) of size 64 × 64 pixels (MN = 4096)
has the intensity distribution shown in following table.
Get the histogram equalization transformation function and give the
ps(sk) for each sk.
10/22/20 35
0
s0  T (r0 )  7 pr (rj )  7  0.19  1.33 1
j 0
1
s1  T (r1 )  7 pr (rj )  7  (0.19  0.25)  3.08 3
j 0
s2  4.55  5 s3  5.67  6
s4  6.23  6 s5  6.65  7
s6  6.86  7 s7  7.00  7
10/22/20 36
10/22/20 37
10/22/20 38
10/22/20 39
Figure 3.22
(a) Image from Phoenix Lander. (b) Result of histogram
equalization. (c) Histogram of image (a). (d) Histogram of image
(b). (Original image courtesy of NASA.)
Question
Is histogram equalization always good?
No
10/22/20 41
Histogram Specification(Matching)
►Histogram equalization does not always provide
Images taken from Gonzalez & Woods, Digital Image Processing (2002)
the desirable results.
• Image of Phobos (Mars moon) and its histogram.

• Many values near zero in the initial histogram
Histogram equalization
Histogram Specification (cont...)
Histogram specification (cont.)
►In these cases, it is more useful to specify
the final histogram.

►Problem statement:
 Given pr(r) from the image and the target
histogram pz(z), estimate the transformation
z=T(r).
• The solution exploits histogram
equalization.
Histogram specification (cont…)
►Equalize the initial histogram of the image:
r
s  T (r )  ( L  1)  pr ( w) dw
0
G ( z )  T (r )
►Equalize the target histogram: r
s  G ( z )  ( L  1)  pz ( w) dw
0
►Obtain the inverse transform: z  G 1 ( s )  G 1 (T (r ))

In practice, for every value of r in the image:
► get its equalized transformation s=T(r).
► perform the inverse mapping z=G-1(s), where s=G(z) is the
equalized target histogram.
Histogram Matching: Example
Assuming continuous intensity values, suppose that an image has

the intensity PDF
 2r
 , for 0  r  L -1
pr (r )   ( L  1) 2
 0, otherwise

Find the transformation function that will produce an image
whose intensity PDF is
 3z 2
 , for 0  z  ( L -1)
pz ( z )   ( L  1) 3
 0, otherwise

10/22/20 46
Histogram Matching: Example
Find the histogram equalization transformation for the input image

2
r r 2w r
s  T (r )  ( L  1)  pr ( w)dw  ( L  1)  dw 
0 0 ( L  1) 2
L 1
Find the histogram equalization transformation for the specified histogram

z z 3t 2 z3
G ( z )  ( L  1)  pz (t )dt  ( L  1)  dt  s
0 0 ( L  1) 3
( L  1) 2
The transformation function

1/3
1/3  2 r
2
 2 1/3
z   ( L  1) s 
2
 ( L  1)   ( L  1)r 
 L  1 
10/22/20 47
Histogram specification (cont…)
The discrete case:

►Equalize the initial histogram of the image:

k
( L  1) k
sk  T (rk )  ( L  1) pr (r j )   nj
j 0 MN j 0 G ( z )  T (r )
►Equalize the target histogram:
q
sk  G ( zq )  ( L  1) pz (ri )
i 0
►Obtain the inverse transform: zq  G 1 ( sk )  G 1 (T (rk ))

Example: Histogram Matching
Suppose that a 3-bit image (L=8) of size 64 × 64 pixels (MN = 4096)
has the intensity distribution shown in the following table (on the
left). Get the histogram transformation function and make the output
image with the specified histogram, listed in the table on the right.
10/22/20 49
► Obtain the scaled histogram-equalized values,
0
s0  T (r0 )  7 pr (rj )  7 pr (r0 )  1.33
j 0
1
s1  T (r1 )  7 pr (rj )  7 pr (r0 )  7 pr (r1 )  3.08
j 0
► Rounding to the nearest integer:
s0  1.33  1 s1  3.08  3 s2  4.55  5 s3  5.67  6
s4  6.23  6 s5  6.65  7 s6  6.86  7 s7  7.00  7
Example
s0  1.33  1 s1  3.08  3 s2  4.55  5 s3  5.67  6

s4  6.23  6 s5  6.65  7 s6  6.86  7 s7  7.00  7
Compute all the values of the transformation function G,
0
G ( z0 )  7 pz ( z j )  0.00  0
j 0
G ( z1 )  0.00  0 G ( z2 )  0.00 0
G ( z3 )  1.05  1 s0=1 G( z4 )  2.45  2 s1=3
G ( z5 )  4.55  5 s2=5 G ( z6 )  5.95 6 s3=s4=6
G ( z7 )  7.00  7 s5= s6= s7=7
Notice that G(z) is not strictly monotonic. We must resolve this
ambiguity by choosing, e.g. the smallest value for the inverse
mapping.
10/22/20 51
10/22/20 52
10/22/20 53
10/22/20 54
Example: Histogram Matching
10/22/20 55
Original image Histogram equalization

Histogram equalization
Specified histogram
Transformation function
and its inverse
Resulting histogram
10/22/20 59
10/22/20 60
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
10/22/20 61
Local Histogram Processing: Example
10/22/20 62
Using Histogram Statistics for Image
Enhancement
Average Intensity L 1 M 1 N 1
1
m   ri p (ri ) 
MN
  f ( x, y )
x 0 y 0
i 0
L 1
un (r )   (ri  m) n p (ri )
i 0
Variance L 1 M 1 N 1
1
  u2 (r )   (ri  m) p (ri )   f ( x, y )  m 
2 2 2

i 0
MN x 0 y 0
10/22/20 63
Enhancement
Local average intensity

L 1
msxy   ri psxy (ri )
i 0
sxy denotes a neighborhood
Local variance
L 1
 2
s xy   (ri  msxy ) psxy (ri )
2
i 0
10/22/20 64
Enhancement: Example
 E  f ( x, y ), if msxy  k0 mG and k1 G   sxy  k 2 G

g ( x, y )  
 f ( x, y ), otherwise
mG : global mean;  G : global standard deviation

k0  0.4; k1  0.02; k 2  0.4; E  4
10/22/20 65
Figure 3.33
(a) Original image. (b) Result of local enhancement based on
local histogram statistics. Compare (b) with Fig. 3.32(c).
Spatial Filtering
A spatial filter consists of (a) a neighborhood, and (b) a

predefined operation
Linear spatial filtering of an image of size MxN with a filter

of size mxn is given by the expression
a b
g ( x, y )    w(s, t ) f ( x  s, y  t )
s  a t  b
10/22/20 67
Spatial Filtering
10/22/20 68
Spatial Correlation
The correlation of a filter w( x, y ) of size m  n

with an image f ( x, y ), denoted as w( x, y ) f ( x, y )
a b
w( x, y ) f ( x, y )    w(s, t ) f ( x  s, y  t )
s  a t  b
10/22/20 69
Spatial Convolution
The convolution of a filter w( x, y ) of size m  n

with an image f ( x, y ), denoted as w( x, y ) f ( x, y )
a b
w( x, y ) f ( x, y )    w(s, t ) f ( x  s, y  t )
s  a t  b
10/22/20 70
Figure 3.35
Illustration of 1-D correlation and convolution of a kernel, w, with a function f consisting of a
discrete unit impulse. Note that correlation and convolution are functions of the variable x, which
acts to displace one function with respect to the other. For the extended correlation and
convolution results, the starting configuration places the rightmost element of the kernel to be
coincident with the origin of f. Additional padding must be used.
10/22/20 72
10/22/20 73
Figure 3.58
Transfer functions of ideal 1-D filters in the frequency domain (u
denotes frequency). (a) Lowpass filter. (b) Highpass filter. (c)
Bandreject filter. (d) Bandpass filter. (As before, we show only
positive frequencies for simplicity.)
Table 3.7
Summary of the four principal spatial filter types expressed in
terms of lowpass filters. The centers of the unit impulse and the
filter kernels coincide.
Smoothing Spatial Filters
Smoothing filters are used for blurring and for noise

reduction
Blurring is used in removal of small details and bridging of

small gaps in lines or curves
Smoothing spatial filters include linear filters and nonlinear

filters.
10/22/20 76
Spatial Smoothing Linear Filters
The general implementation for filtering an M  N image

with a weighted averaging filter of size m  n is given
a b
  w(s, t ) f ( x  s, y  t )
g ( x, y )  s  a t  b
a b
  w(s, t )
s  a t  b
where m  2a  1, n  2b  1.
10/22/20 77
Two Smoothing Averaging Filter Masks
10/22/20 78
10/22/20 79
Example: Gross Representation of Objects
10/22/20 80
Order-statistic (Nonlinear) Filters
— Nonlinear
— Based on ordering (ranking) the pixels contained in the

filter mask
— Replacing the value of the center pixel with the value

determined by the ranking result
E.g., median filter, max filter, min filter
10/22/20 81
Example: Use of Median Filtering for Noise Reduction
10/22/20 82
Sharpening Spatial Filters
► Foundation
► Laplacian Operator
► Unsharp Masking and Highboost Filtering
► Using First-Order Derivatives for Nonlinear Image

Sharpening — The Gradient
10/22/20 83
Sharpening Spatial Filters: Foundation
► The first-order derivative of a one-dimensional function f(x)

is the difference
f
 f ( x  1)  f ( x)
x
► The second-order derivative of f(x) as the difference
2 f
 f ( x  1)  f ( x  1)  2 f ( x)
x 2
10/22/20 84
10/22/20 85
Sharpening Spatial Filters: Laplace Operator
The second-order isotropic derivative operator is the

Laplacian for a function (image) f(x,y)
 2
f  2
f
 f  2  2
2
x y
2 f
 f ( x  1, y )  f ( x  1, y )  2 f ( x, y )
x 2
2 f
 f ( x, y  1)  f ( x, y  1)  2 f ( x, y )
y 2
 2 f  f ( x  1, y )  f ( x  1, y )  f ( x, y  1)  f ( x, y  1)
- 4 f ( x, y )
10/22/20 86
10/22/20 87
Image sharpening in the way of using the Laplacian:
g ( x, y )  f ( x, y )  c  2 f ( x, y ) 
where,
f ( x, y ) is input image,
g ( x, y ) is sharpenend images,
c  -1 if  2 f ( x, y ) corresponding to Fig. 3.37(a) or (b)
and c  1 if either of the other two filters is used.
10/22/20 88
10/22/20 89
Unsharp Masking and Highboost Filtering
► Unsharp masking
Sharpen images consists of subtracting an unsharp (smoothed)
version of an image from the original image
e.g., printing and publishing industry
► Steps
1. Blur the original image
2. Subtract the blurred image from the original
3. Add the mask to the original
10/22/20 90
Unsharp Masking and Highboost Filtering
Let f ( x, y ) denote the blurred image, unsharp masking is

g mask ( x, y )  f ( x, y )  f ( x, y )
Then add a weighted portion of the mask back to the original
g ( x, y )  f ( x, y )  k * g mask ( x, y ) k 0
when k  1, the process is referred to as highboost filtering.
10/22/20 91
Unsharp Masking: Demo
10/22/20 92
Figure 3.55
(a) Unretouched “soft-tone” digital image of size469×600pixels

(b) Image blurred using a 31×31 Gaussian lowpass filter with σ = 5.
(c) Mask. (d) Result of unsharp masking using Eq. (3-65) with k = 1. (e) and (f)
Results of highboost filtering with k = 2 and k = 3, respectively.
Unsharp Masking and Highboost Filtering: Example
10/22/20 94
Image Sharpening based on First-Order Derivatives
For function f ( x, y ), the gradient of f at coordinates ( x, y )

is defined as
 f 
 g x   x 
f  grad( f )      
 g y   f 
 y 
The magnitude of vector f , denoted as M ( x, y )

Gradient Image M ( x, y )  mag(f )  g x 2  g y 2
10/22/20 95
The magnitude of vector f , denoted as M ( x, y )

M ( x, y )  mag(f )  g x 2  g y 2
M ( x, y ) | g x |  | g y |
z1 z2 z3
M ( x, y ) | z8  z5 |  | z6  z5 |
z4 z5 z6
z7 z8 z9
10/22/20 96
Roberts Cross-gradient Operators

M ( x, y ) | z9  z5 |  | z8  z6 |
Sobel Operators
M ( x, y ) | ( z7  2 z8  z9 )  ( z1  2 z2  z3 ) |
z1 z2 z3  | ( z3  2 z6  z9 )  ( z1  2 z4  z7 ) |
z4 z5 z6
z7 z8 z9
10/22/20 97
10/22/20 98
Example
10/22/20 99
Example:
Combining
Spatial
Enhancement
Methods
Goal:
Enhance the
image by
sharpening it
and by bringing
out more of the
skeletal detail
10/22/20 100
Example:
Combining
Spatial
Enhancement
Methods
Goal:
Enhance the
image by
sharpening it
and by bringing
out more of the
skeletal detail
10/22/20 101

Lecture 2. Intensity Transformation and Spatial Filtering

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 2. Intensity Transformation and Spatial Filtering

Uploaded by

Copyright:

Available Formats

Lecture 2.

Prof. Dr. Noorullah Shariff

► Image Processing techniques in the

► Intensity transformation can be considered

Intensity transformation function

Measuring the actual

► Local Histogram Processing

► Using Histogram Statistics for Image Enhancement

a. T(r) is a strictly monotonically increasing function

a. T(r) is a strictly monotonically increasing function

T (r ) is continuous and differentiable.

Find the transformation function for equalizing

the desirable results.

• Image of Phobos (Mars moon) and its histogram.

the final histogram.

►Obtain the inverse transform: z  G 1 ( s )  G 1 (T (r ))

Assuming continuous intensity values, suppose that an image has

Find the histogram equalization transformation for the input image

Find the histogram equalization transformation for the specified histogram

The transformation function

The discrete case:

►Equalize the initial histogram of the image:

►Obtain the inverse transform: zq  G 1 ( sk )  G 1 (T (rk ))

s0  1.33  1 s1  3.08  3 s2  4.55  5 s3  5.67  6

Original image Histogram equalization

Local average intensity

sxy denotes a neighborhood

 E  f ( x, y ), if msxy  k0 mG and k1 G   sxy  k 2 G

mG : global mean;  G : global standard deviation

A spatial filter consists of (a) a neighborhood, and (b) a

Linear spatial filtering of an image of size MxN with a filter

The correlation of a filter w( x, y ) of size m  n

The convolution of a filter w( x, y ) of size m  n

Smoothing filters are used for blurring and for noise

Blurring is used in removal of small details and bridging of

Smoothing spatial filters include linear filters and nonlinear

The general implementation for filtering an M  N image

— Based on ordering (ranking) the pixels contained in the

— Replacing the value of the center pixel with the value

E.g., median filter, max filter, min filter

► Unsharp Masking and Highboost Filtering

► Using First-Order Derivatives for Nonlinear Image

► The first-order derivative of a one-dimensional function f(x)

► The second-order derivative of f(x) as the difference

The second-order isotropic derivative operator is the

Image sharpening in the way of using the Laplacian:

2. Subtract the blurred image from the original

3. Add the mask to the original

Let f ( x, y ) denote the blurred image, unsharp masking is

when k  1, the process is referred to as highboost filtering.

(a) Unretouched “soft-tone” digital image of size469×600pixels

For function f ( x, y ), the gradient of f at coordinates ( x, y )

The magnitude of vector f , denoted as M ( x, y )

The magnitude of vector f , denoted as M ( x, y )

Roberts Cross-gradient Operators

You might also like