Lecture 03,04,05 - Intensity Transformation and Spatial Filtering PDF

Intensity transformation and spatial filtering
Digital Image Processing Lecture 3 Intensity Transformation& Spatial Filtering

Lecturer: Ha Dai Duong Faculty of Information Technology
I. Introduction
Spatial Domain vs. Transform Domain

Spatial Domain
Image plane itself, directly process the intensity values of the image plane Process the transform coefficients, not directly process the intensity values of the image plane
Transform (Frequency) domain

Digital Image Processing
I. Introduction
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 )
3
I. Introduction
Spatial Domain Process
II. Intensity transformation function

Intensity transformation function s = T (r )
II. Intensity transformation function

Some basic Functions
intensity
transformation
II.1. Negative
Image negatives s = L 1 r
II.1. Negative
Small lesion
II.2. Log Transform

Log Transformations s = c log(1 + r )
II.2. Log Transform
10
II.3. Power Law
s = cr
11
II.3. Power Law
12
II.3. Power Law
13
II.4. Piecewise-Linear Transform..

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
14
II.4. Piecewise-Linear Transform
15
16
II.5. Bit Plane Slicing
17
II.5. Bit Plane Slicing
18
III. Histogram processing

Histogram Equalization Histogram Matching Local Histogram Processing Using Histogram Statistics Enhancement
for
Image
19

Histogram h(rk ) = nk rk is the k th intensity value nk is the number of pixels in the image with intensity rk
nk MN nk : the number of pixels in the image of Normalized histogram p (rk ) = size M N with intensity rk
20
10

No. of pixels
6
2 4 3 2
3 2 2 4
3 4 3 2
2 3 5 4
5 4 3 2 1
4x4 image Gray scale = [0,9]

Gray level
0 1 2 3 4 5 6 7 8 9
histogram
21
22
11
III.1. Histogram Equalization

As the low-contrast images histogram is narrow and centered toward the middle of the gray scale, if we distribute the histogram to a wider range the quality of the image will be improved. We can do it by adjusting the probability density function of the original histogram of the image so that the probability spread equally
23

Histogram transformation
s

s = T(r)
Where 0 r 1 T(r) satisfies
sk= T(rk) T(r)
(a). T(r) is single-valued and monotonically increasingly in the interval 0r1 (b). 0 T(r) 1 for 0r1
rk
r
24
12

2 conditions of T(r)
Single-valued (one-to-one relationship) guarantees that the inverse transformation will exist Monotonicity condition preserves the increasing order from black to white in the output image thus it wont cause a negative image 0 T(r) 1 for 0 r 1 guarantees that the output gray levels will be in the same range as the input levels. The inverse transformation from s back to r is r = T -1(s) ; 0 s 1
25

Let

pr(r) denote the PDF of random variable r ps(s) denote the PDF of random variable s
If pr(r) and T(r) are known and T-1(s) satisfies condition (a) then ps(s) can be obtained using a formula :
ps(s) = pr(r)
dr ds
26
13

A transformation function is a cumulative distribution function (CDF) of random variable r :
s = T ( r ) = pr ( w )dw
0

CDF is an integral of a probability function (always positive) is the area under the function Thus, CDF is always single valued and monotonically increasing Thus, CDF satisfies the condition (a) We can use CDF as a transformation function
27

ds dT ( r ) = dr dr r d p ( w ) dw = r dr 0 = pr ( r )
p s ( s ) = pr ( r ) = pr ( r )
dr ds 1 pr ( r )
Substitute and yield
= 1 where 0 s 1
28
14

Ps(s):

As ps(s) is a probability function, it must be zero outside the interval [0,1] in this case because its integral over all values of s must equal 1. Called ps(s) as a uniform probability density function ps(s) is always a uniform, independent of the form of pr(r)
29

Discrete Transformation Function

The probability of occurrence of gray level in an image is approximated by n where k = 0 , 1 , ..., L- 1 p r ( rk ) = k n The discrete version of transformation
s k = T ( rk ) = =
j=0
pr ( rj ) = 0 , 1 , ..., L- 1
30
nj n
where k
j=0
15

Thus, an output image is obtained by mapping each pixel with level rk in the input image into a corresponding pixel with level sk in the output image In discrete space, it cannot be proved in general that this discrete transformation will produce the discrete equivalent of a uniform probability density function, which would be a uniform histogram
31

Example
before
after
Histogram equalization
32
16

Example
before after Histogram equalization
The quality is not improved much because the original image already has a broaden gray-level scale
33

No. of pixels
6
2 4 3 2
3 2 2 4
3 4 3 2
2 3 5 4
5 4 3 2 1
4x4 image Gray scale = [0,9]

Gray level
0 1 2 3 4 5 6 7 8 9
histogram
34
17

Example
0
0 0 0 0
Gray Level No.of pixels
1
0 0 0 0
2
6 6
6/ 16
3
5 11
4
4 15
5
1 16
6
0 16
7
0 16
8
0 16
9
0 16
n
k
j=0
s=
j =0
nj n
11 / 15 / 16 / 16/ 16/ 16/ 16/ 16 16 16 16 16 16 16
sx9
3.3 3
6.1 6
8.4 8
9
35

Example
3 8 6 3 6 3 3 8 6 8 6 3 3 6 9 8
No. of pixels
6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9
Output image Gray scale = [0,9]

Gray level Histogram equalization

36
18
III.2. Histogram Matching

Generate a processed image that has a specified histogram

Let pr ( r ) and pz ( z ) denote the continous probability density functions of the variables r and z. pz ( z ) is the specified probability density function. Let s be the random variable with the probability s = T (r ) = ( L 1) pr ( w)dw
0 r
Define a random variable z with the probability G ( z ) = ( L 1) pz (t )dt = s

0 z
37
s = T ( r ) = ( L 1) p r ( w ) dw
0
G ( z ) = ( L 1) p z ( t ) dt = s
0
z = G 1 ( s ) = G 1 [T (r ) ]
38
19

Procedure
1.
Obtain pr(r) from the input image and then obtain the values of s
s = ( L 1) pr ( w)dw
0
2.
Use the specified PDF and obtain the transformation function G(z)
G ( z ) = ( L 1) pz (t )dt = s
0
3.
Mapping from s to z
z = G 1 ( s )
39

Example
Assume an image has a gray level probability density function pr(r) as shown. Pr(r) 2 1 0 1 2 r
40
2 r + 2 ;0 r 1 pr ( r ) = ; elsewhere 0
p r ( w ) dw = 1
20

Example
We would like to apply the histogram specification with the desired probability density function pz(z) as shown. Pz(z) 2 1 0
2z pz ( z ) = 0
;0 z 1 ; elsewhere
z
p
0
( w )dw = 1
41

Example
s=T(r) 1
1. Obtain the transformation function T(r)
s = T ( r ) = pr ( w )dw
0
= ( 2 w + 2 )dw
0
= w 2 + 2w
0
r 0
= r + 2r
2
42
21

Example
2. Obtain the transformation function G(z)
G ( z ) = ( 2 w )dw
0
= z2
z 0
= z2
43

Example
3. Obtain the inversed transformation function G-1
G ( z ) = T (r ) z 2 = r 2 + 2r z = 2r r 2
We can guarantee that 0 z 1 when 0 r 1
44
22

Procedure in discrete cases

1.
Obtain pr(rj) from the input image and then obtain the values of sk, round the value to the integer range [0, L-1]
sk = T (rk ) = ( L 1) pr (rj ) =
j =0
( L 1) k nj MN j =0
2.
Use the specified PDF and obtain the transformation function G(zq), round the value to the integer range [0, L-1].
G ( zq ) = ( L 1) pz ( zi ) = sk
i =0
3.
Mapping from sk to zq
zq = G 1 ( sk )
45

Example
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.
46
23

Example
Obtain the scaled histogram-equalized values,
s0 = 1, s1 = 3, s2 = 5, s3 = 6, s4 = 7, s5 = 7, s6 = 7, s7 = 7.
Compute all the values of the transformation function G,
G ( z0 ) = 7 pz ( z j ) = 0.00
j =0
0
s0 s2
G ( z1 ) = 0.00 0 G ( z5 ) = 4.55 5
G ( z3 ) = 1.05 1 G ( z7 ) = 7.00 7
0 G ( z4 ) = 2.45 2 s1
G ( z2 ) = 0.00 G ( z6 ) = 5.95
s7
47
6 s3
s4 s5 s6

Example
rk 0 1 2 3 4 5 6 7
s0 = 1, s1 = 3, s2 = 5, s3 = 6, s4 = 7, s5 = 7, s6 = 7, s7 = 7.
rk zq 03 1 4 25 36 47 57 67 77
48
24

Example
49

Example
50
25
III.3. 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
51
III.3. Local Histogram Processing
52
26
III.4. Using Histogram Statistics

Average Intensity
m = ri p (ri )
i =0
L 1 i =0
L 1
1 = MN
M 1 N 1 x =0 y =0
f ( x, y )
un (r ) = (ri m) n p (ri )
Variance
2 L 1 i =0
= u2 (r ) = (ri m) 2 p(ri )
1 MN
M 1 N 1 x =0 y =0
[ f ( x, y ) m ]
53

Local average intensity msxy = ri psxy (ri )
i =0 L 1
sxy denotes a neighborhood

Local variance
L 1 i =0
s2 = (ri ms )2 ps (ri )
xy xy xy
54
27

E f ( x, y ), if msxy k0 mG and k1 G sxy k2 G g ( x, y ) = otherwise f ( x, y ), mG : global mean;
G : global standard deviation
k0 = 0.4; k1 = 0.02; k2 = 0.4; E = 4
55
IV. Using Arithmetic/Logic Operators

Arithmetic/Logic operations perform on pixel by pixel basis between two or more images, except NOT operation which perform only on a single image Logic operation performs on gray-level images, the pixel values are processed as binary numbers:

Light represents a binary 1, and Dark represents a binary 0

56
28
IV.1. Logic Operators
57
IV.2. Image Subtraction
g(x,y) = f(x,y) h(x,y)

Extraction of the differences between images
58
29
IV.3. Image Averaging

Consider a noisy image g(x,y) formed by the addition of noise (x,y) to an original image f(x,y)
g(x,y) = f(x,y) + (x,y)

If noise has zero mean and be uncorrelated then it can be shown that if
g ( x, y )
= image formed by averaging K different noisy images
1 g ( x, y ) = K
g ( x, y )
i =1 i
59

Then
2 g ( x, y ) =
2 g ( x , y ) , 2 ( x , y )
1 2 ( x, y ) K
= variances of g and
if K increase, it indicates that the variability (noise) of the pixel at each location (x,y) decreases.
60
30

Thus
E{g ( x, y )} = f ( x, y )
E{ g ( x , y )}
= expected value of g (output after averaging) = original image f(x,y)
61
V. 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
g ( x, y ) =
s = a t = b
w(s, t ) f ( x + s, y + t )
62
31
V. Spatial Filtering
Example
M=3, N=3 3x3 Or M=5, N=5 5x5
63
V.1. Smoothing 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.
64
32

Linear filters
The general implementation for filtering an M N image with a weighted averaging filter of size m n is given g ( x, y ) =
s = a t = b
w(s, t ) f ( x + s, y + t )
s = a t = b
w(s, t )
n = 2b + 1.
where m = 2a + 1,
65

Linear filters
Averaging Filter Masks
66
33

a c e b d f

Example
Averaging Filter Masks a). original image 500x500 pixel b). - f). results of smoothing with square averaging filter masks of size n = 3, 5, 9, 15 and 35, respectively. Note:

Big mask is used to eliminate small objects from an image. The size of the mask establishes the relative size of the objects that will be blended with the background.
67

Example
original image
result after smoothing with 15x15 averaging mask
result of thresholding
we can see that the result after smoothing and thresholding, the remains are the largest and brightest objects in the image.
68
34

Order-Statistics Filters (Nonlinear Filters)

The response is based on ordering (ranking) the pixels contained in the image area encompassed by the filter Example

Median filter : R = median{zk |k = 1,2,,n x n} Max filter : R = max{zk |k = 1,2,,n x n} Min filter : R = min{zk |k = 1,2,,n x n}
Note: n x n is the size of the mask
69

Median Filter
Median: X= {x1,x2,, x2b+1}, x is called the median of X if x greater than or equal b elements and less than or equal b other elements in X; For example: X={3,2,3,2,3,4,5,6,5} (b=4) -> Median is 3 Peplaces the value of a pixel by the median of the gray levels in the neighborhood of that pixel (the original value of the pixel is included in the computation of the median) Quite popular because for certain types of random noise (impulse impulse noise salt and pepper noise) , they provide excellent noisenoise noisereduction capabilities, capabilities with considering less blurring than linear smoothing filters of similar size
70
35

Median Filter
Example
71
V.2. Sharpening Spatial Filters

Foundation Laplacian Operator Unsharp Masking and Highboost Filtering Using First-Order Derivatives for Nonlinear Image Sharpening - The Gradient
72
36
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
73

Foundation
74
37
V.3. Method based on Laplace Operator

Laplace Operator
The second-order isotropic derivative operator is the Laplacian for a function (image) f(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 2 f f = 2 + 2 x y
2
2 f = f ( x + 1, y ) + f ( x 1, y ) + f ( x, y + 1) + f ( x, y 1) - 4 f ( x, y )
75

Masks
76
38

Image sharpening in the way of using the Laplacian:
where g(x,y) sharpened image
77

Example
78
39
V.4. Unsharp Masking and Highboost Filtering
79
80
40
81

Example
82
41

Example
83
V.5. Method based Derivatives (Gradient)
on
First-Order
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 ) M ( x, y ) = mag(f ) = g x 2 + g y 2

84
42
on
First-Order
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 z4 z7 z2 z5 z8 z3 z6 z9
85
M ( x, y ) =| z8 z5 | + | z6 z5 |
on
First-Order
Roberts Cross-gradient Operators M ( x, y ) | z9 z5 | + | z8 z6 |
z1 z4 z7
z2 z5 z8
z3 z6 z9
Sobel Operators M ( x, y ) | ( z7 + 2 z8 + z9 ) ( z1 + 2 z2 + z3 ) | + | ( z3 + 2 z6 + z9 ) ( z1 + 2 z4 + z7 ) |
86
43
on
First-Order
87

on
First-Order
Example
88
44
V.6. Combining Spatial Enhancement Methods

Solve:
1. Laplacian to highlight fine detail 2. Gradient to enhance prominent edges 3. Gray-level transformation to increase the dynamic range of gray levels
89

Exampe
90
45

Exampe
91
46

Lecture 03,04,05 - Intensity Transformation and Spatial Filtering PDF

Uploaded by

Document Information

Original Description:

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 03,04,05 - Intensity Transformation and Spatial Filtering PDF

Uploaded by

Copyright:

Available Formats

Intensity transformation and spatial filtering

Digital Image Processing Lecture 3 Intensity Transformation& Spatial Filtering

Spatial Domain vs. Transform Domain

Transform (Frequency) domain

Digital Image Processing

Intensity transformation and spatial filtering

Spatial Domain Process

Digital Image Processing

Spatial Domain Process

Digital Image Processing

Intensity transformation and spatial filtering

II. Intensity transformation function

Digital Image Processing

II. Intensity transformation function

Some basic Functions

Digital Image Processing

Intensity transformation and spatial filtering

Digital Image Processing

Digital Image Processing

Intensity transformation and spatial filtering

II.2. Log Transform

Digital Image Processing

II.2. Log Transform

Digital Image Processing

Intensity transformation and spatial filtering

II.3. Power Law

Digital Image Processing

II.3. Power Law

Digital Image Processing

Intensity transformation and spatial filtering

II.3. Power Law

Digital Image Processing

II.4. Piecewise-Linear Transform..

Digital Image Processing

Intensity transformation and spatial filtering

II.4. Piecewise-Linear Transform

Digital Image Processing

Digital Image Processing

Intensity transformation and spatial filtering

II.5. Bit Plane Slicing

Digital Image Processing

II.5. Bit Plane Slicing

Digital Image Processing

Intensity transformation and spatial filtering

III. Histogram processing

Digital Image Processing

III. Histogram processing

Intensity transformation and spatial filtering

III. Histogram processing

4x4 image Gray scale = [0,9]

III. Histogram processing

Digital Image Processing

Intensity transformation and spatial filtering

III.1. Histogram Equalization

Digital Image Processing

III.1. Histogram Equalization

sk= T(rk) T(r)

Digital Image Processing

Intensity transformation and spatial filtering

III.1. Histogram Equalization

III.1. Histogram Equalization

Intensity transformation and spatial filtering