1. Image Enhancement course: biomedical image processing vibhor kumar
Biomedical Image Analysis
Rangaraj M. Rangayyan
course: biomedical image processing
vibhor kumar
Hannu Laaksonen

2. Topics to be covered
Convolution mask Operations
. unsharp masking
. Sobtracting Laplacian
2) High Frequency Emphasis
3) Homomorphic filtering for Enhancement
4) Adaptive Contrast enhancement

3. Convolution Mask Operators - Unsharp masking
The Generalized equation of unsharp masking is
f(m,n) = [g(m,n) - µg(m,n)] +  g(m,n)
blurred image
Is calculated as average of the pixels in the window taken around the pixel(m,n)
The weight  can be changed according to desired effect.
For e.g. For a 3X3 convolution mask the unsharp masking is given by
- 1 8 2

4. Subtracting Laplacian
The degraded image may be expressed in a Taylor series as g t 2 2
t2
2g
g(x,y,) = g(x,y,0) +  (x,y,t) …... g t
Taking = k2g using
diffusion model
We get fe = g - k2g
For k = 1 and mask 3X3 we get subtracting Laplacian as:
Unlike Laplacian subtracting Laplacian maintain the intensity information while making the image sharp

5. (a)
(b)
original lena image
Laplacian
Unsharp masking
subtracting laplacian
(d)
(c)

6. High-frequency Emphasis
Highpass filtering are useful in detecting edges but for enhancing the images it is necessary to maintain the intensity information.
High-emphasis filter does the image enhance keeping the intensity information.
The Butterworth high-emphasis filter can termed as: 2 D0
D(u,v) 2n
1 + (sqrt(2)-1 )
1 +
H(u,v) =
Filter gain
Frequency

7. Original shape image
the ideal high pass filter
The Butter worth highpass filter
the Butterworth high-emphasis filter

8. Enhancement using Homomorphic filtering
linear filtering and enhancement
inverse transform
transform
input image
filtered image

9. Original Image
log transform of original image
Homomorphic filtering including a Butterworth high-emphasis filter
Butterworth high imphasis filter only
Graphical Models and Image Processing, 54(3): ,May 1992

10. Adaptive Contrast Enhancement
Adaptive-neighborhood contrast enhancement:
non overlapping regions segmentation
Overlapping regions segmentation
seed fill region growing:
The region consists of spatially connected pixels that falls that fall within the specified gray level devaiation from seed pixel.
every time data is devided into back ground and foreground pixels
The growth tolerance threshold  is highly important factor

11. Adaptive Contrast Enhancement
contrast enhancment can be done using the formula
fe = b (1+Ce)/ (1-Ce)
increased contrast
mean background value
Part of mammogram with a cluster of calcification
adaptive-neighborhood contrast enhancement
gamma correction
unsharp masking
IEEE Transcation on Medical imaging 11(3): ,1992

12. Topics covered
Convolution mask Operations
. unsharp masking
. Sobtracting Laplacian
2) High Frequency Emphasis
3) Homomorphic filtering for Enhancement
4) Adaptive Contrast enhancement