image enhancement ii
TRANSCRIPT
-
8/13/2019 Image Enhancement II
1/68
Digital Image Processing
Lecture 4: Image Enhancement II
Dr. Muhammad Amjad Iqbal
-
8/13/2019 Image Enhancement II
2/68
Pixel Operations
for Image Enhancement
-
8/13/2019 Image Enhancement II
3/68
1/10/2014 3
Common Pixel Operations
Image Negatives
Log Transformations
Power-LawTransformations
-
8/13/2019 Image Enhancement II
4/68
1/10/2014 4
Image Negatives
Reverses the gray level order For L gray levels the transformation function is
s =T (r ) = (L - 1) - r
-
8/13/2019 Image Enhancement II
5/68
1/10/2014 5
Image Scaling
s =T (r ) = a.r (a is a constant)
-
8/13/2019 Image Enhancement II
6/68
1/10/2014 6
Log Transformations
Function of s = cLog(1+r )
-
8/13/2019 Image Enhancement II
7/68
1/10/2014 7
Log Transformations
Properties of log transformations –For lower amplitudes of input image the range of gray
levels is expanded
–For higher amplitudes of input image the range of gray
levels is compressed
Application: – This transformation is suitable for the case when the
dynamic range of a processed image far exceeds the
capability of the display device (e.g. display of theFourier spectrum of an image)
– Also called “dynamic-range compression / expansion”
-
8/13/2019 Image Enhancement II
8/68
1/10/2014 8
Log Transformations
-
8/13/2019 Image Enhancement II
9/68
1/10/2014 9
Power-Law Transformation
-
8/13/2019 Image Enhancement II
10/68
1/10/2014 10
Power-Law Transformation
For γ < 1: Expands values of dark pixels,compress values of brighter pixels
For γ > 1: Compresses values of dark pixels,expand values of brighter pixels
If γ=1 & c=1: Identity transformation (s = r)
A variety of devices (image capture, printing, display) respondaccording to power law and need to be corrected
Gamma (γ) correction
The process used to correct the power-law responsephenomena
-
8/13/2019 Image Enhancement II
11/68
1/10/2014 11
Power-Law Transformation
-
8/13/2019 Image Enhancement II
12/68
1/10/2014 12
Gamma Correction
-
8/13/2019 Image Enhancement II
13/68
1/10/2014 13
Power Law Example
-
8/13/2019 Image Enhancement II
14/68
1/10/2014 14
Power Law Example (cont…)
γ = 0.6
00.1
0.2
0.3
0.4
0.5
0.60.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
Old Intensities
T
r a n s f o r m e d I n t e n
s i t i e s
-
8/13/2019 Image Enhancement II
15/68
1/10/2014 15
Power Law Example (cont…)
γ = 0.4
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
Original Intensities
T r a n s f o r m e d I n t e
n s i t i e s
-
8/13/2019 Image Enhancement II
16/68
1/10/2014 16
Power Law Example (cont…)
γ = 0.3
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
Original Intensities
T r a n s f o r m e d I n t e
n s i t i e s
-
8/13/2019 Image Enhancement II
17/68
1/10/2014 17
Power Law Example (cont…)
• The images to theright show amagnetic resonance
(MR) image of afractured humanspine
• Different curves
highlight differentdetail
s = r 0.6
s =
r 0 . 4
-
8/13/2019 Image Enhancement II
18/68
1/10/2014 18
Power Law Example
-
8/13/2019 Image Enhancement II
19/68
1/10/2014 19
Power Law Example (cont…)
γ = 5.0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8 1
Original Intensities
T r a n s f o r m e d I n t e n s i t i e s
-
8/13/2019 Image Enhancement II
20/68
1/10/2014 20
Power Law Transformations (cont…)
• An aerial photo
of a runway isshown
• This timepower law
transforms areused to darkenthe image
• Different curves
highlightdifferent detail
s = r 3.0
s =
r 4 . 0
i i i f i
-
8/13/2019 Image Enhancement II
21/68
1/10/2014 21
Piecewise-Linear Transformation
Contrast StretchingGoal:
– Increase the dynamic range of the gray levels for low
contrast images
Low-contrast images can result from
– poor illumination
–lack of dynamic range in the imaging sensor
–wrong setting of a lens aperture during image
acquisition
C S hi E l
-
8/13/2019 Image Enhancement II
22/68
1/10/2014 22
Contrast Stretching Example
-
8/13/2019 Image Enhancement II
23/68
1/10/2014 23
Piecewise-Linear Transformation: Contrast Stretching
-
8/13/2019 Image Enhancement II
24/68
1/10/2014 24
Thresholding
-
8/13/2019 Image Enhancement II
25/68
1/10/2014 25
Gray Level Slicing
•Highlighting a specific range ofGray levels in an image
• First approach
– Display a high value for all the
gray levels in the range of
interest
– Low value for all other gray
levels
– This will produce a Binary Image
-
8/13/2019 Image Enhancement II
26/68
1/10/2014 26
Gray Level Slicing
• 2nd approach
– Brightens the desired range of
Gray Levels but preserves the
Gray Levels of rest of the pixels
-
8/13/2019 Image Enhancement II
27/68
Contrast Stretching
-
8/13/2019 Image Enhancement II
28/68
Contrast Stretching through Histogram
C
If r max and r min are the maximum and minimum gray levelof the input image and L is the total gray levels of output
image The transformation function for contrast stretching
will be
-
8/13/2019 Image Enhancement II
29/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
30/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
31/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
32/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
33/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
34/68
Histogram Equalization
-
8/13/2019 Image Enhancement II
35/68
Histogram Equalisation(Summary)
Spreading out the frequencies in an image (orequalising the image) is a simple way to improvedark or washed out images
The formula for histogramequalisation is given where
– r k : input intensity
– sk : processed intensity
– k : the intensity range(e.g 0.0 – 1.0)
– n j: the frequency of intensity j
– n: the sum of all frequencies
)( k k r T s
k
j
jr r p1
)(
k
j
j
n
n
1
-
8/13/2019 Image Enhancement II
36/68
Example
-
8/13/2019 Image Enhancement II
37/68
Example: cdf
-
8/13/2019 Image Enhancement II
38/68
Example
Notice that the minimum value (52) is now 0 and the maximum value (154) is now 255.
Initial Image Image After Equalization
-
8/13/2019 Image Enhancement II
39/68
Example
-
8/13/2019 Image Enhancement II
40/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
41/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
42/68
-
8/13/2019 Image Enhancement II
43/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
44/68
Equalisation Transformation Function
I m a g e s t a k e n f r o m G o n z a l e z & W o o d s , D i g i t a l
I m a g e P r o c e s s i n g ( 2 0 0 2 )
-
8/13/2019 Image Enhancement II
45/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
46/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
47/68
Histogram Equalization-Examples
-
8/13/2019 Image Enhancement II
48/68
-
8/13/2019 Image Enhancement II
49/68
Edge detection
S i l Fil i f I Sh i
-
8/13/2019 Image Enhancement II
50/68
Spatial Filtering for Image Sharpening
Background: to highlight fine detail in an image or toenhance blurred detail
Applications: electronic printing, medical imaging, industrialinspection, autonomous target detection (smartweapons)......
Foundation:
Blurring/smoothing is performed by spatial averaging(equivalent to integration)
Sharpening is performed by noting only the gray level
changes in the image that is the differentiation
S ti l Filt i f I Sh i
-
8/13/2019 Image Enhancement II
51/68
Spatial Filtering for Image Sharpening
Operation of Image Differentiation Enhance edges and discontinuities (magnitude of
output gray level >>0)
De-emphasize areas with slowly varying gray-levelvalues (output gray level: 0)
Mathematical Basis of Filtering for Image Sharpening
First-order and Second-order Derivatives
Approximation in Discrete-space Domain
Implementation by Mask filtering
-
8/13/2019 Image Enhancement II
52/68
1st Derivative
• The formula for the 1st derivative of a function is as follows:
• It’s just the difference between subsequent values and
measures the rate of change of the function
)()1( x f x f x
f
-
8/13/2019 Image Enhancement II
53/68
First and Second Order Derivatives
-
8/13/2019 Image Enhancement II
54/68
First and Second Order Derivatives
l f
-
8/13/2019 Image Enhancement II
55/68
Example for Discrete Derivatives
Let’s consider a simple 1 dimensional example
l f i i i
-
8/13/2019 Image Enhancement II
56/68
Example for Discrete Derivatives
E l f Di D i i
-
8/13/2019 Image Enhancement II
57/68
Example for Discrete Derivatives
-
8/13/2019 Image Enhancement II
58/68
Various Situations encountered by
Derivatives• Flat Segment f’=0 and f’’=0
• Step f’=0 and f’’=0
-
8/13/2019 Image Enhancement II
59/68
• Ramp
•Ramp or step in 1D profile normally characterize an edge in the image
•f’’ is non-zero on on-set and end of Ramp and produces thin edges
•f’ is non-zero along whole Ramp and produces thick edges
-
8/13/2019 Image Enhancement II
60/68
• Thin Line
• Isolated Point
C i b t f" d f´
-
8/13/2019 Image Enhancement II
61/68
Comparison between f" and f´
f´ generally produces thicker edges in an image
f" has a stronger response to fine detail
f´ generally has a stronger response to a gray-level step
f" produces a double response at step changes in gray
level For image enhancement, f" is generally better suited
than f´
Major application of f´ is for edge extraction; f´ used
together with f" results in impressive enhancement effect
Appl ing First and 2nd Deri ati e sing 1D
-
8/13/2019 Image Enhancement II
62/68
Applying First and 2nd Derivative using 1D
Kernel
E h b 2 d D i i (E l )
-
8/13/2019 Image Enhancement II
63/68
Enhancement by 2nd Derivative (Example)
L l i f I E h t
-
8/13/2019 Image Enhancement II
64/68
Laplacian for Image Enhancement
L l i f I E h t
-
8/13/2019 Image Enhancement II
65/68
Laplacian for Image Enhancement
Laplacian for Image Enhancement
-
8/13/2019 Image Enhancement II
66/68
Laplacian for Image Enhancement
Laplacian for Image Enhancement (Example)
-
8/13/2019 Image Enhancement II
67/68
Laplacian for Image Enhancement (Example)
Laplacian for Image Enhancement (Example)
-
8/13/2019 Image Enhancement II
68/68
Laplacian for Image Enhancement (Example)