Assignment -4

1. Image Restoration (Noise Removal)1. Arithmetic Mean2. Geometric Mean3. Local Noise Reduction Filter4. Median Filter5. Adaptive Median Filter

2. Color Image Processing (Pseudo-color image processing)1. Intensity Slicing2. Intensity to Color Transformation

Due Date: May 11th , 11:59 PM

Image Restoration

Input

1. Noise Image

Gaussian Bipolar (Salt and Pepper)

Filtering

Noise image

Filter (3X3)

Output image

Filtering

Noise image

Filter (3X3)

Output image

Zero padding

Filtering

Noise image

Filter (3X3)

Output image

Filter:MeanMedian…

Roi:Region of interest

Filtering

Noise image

Kernel (3X3)

Output image

Filter:MeanMedian…roi

Arithmetic Mean Filter

Gaussian Mean: 0, Var: 100 Filter Size: 7

Geometric Mean Filter

Gaussian Mean: 0, Var: 100 Filter Size: 9

Local Noise Mean Filter

Gaussian Mean: 0, Var: 100 Filter Size: 9

Median Filter

Bipolar: Salt/Pepper probability: 0.01

Filter Size: 3

Median Filter

Bipolar: Salt/Pepper probability: 0.5

Filter Size: 7

Adaptive Median Filter

Bipolar: Salt/Pepper probability: 0.5

Filter Size: 7

Median vs. Adaptive Median

Median Adaptive Median

Color Image Processing

1. Color Image Processing (Pseudo-color image processing)1. Intensity Slicing2. Intensity to Color Transformation

1. Intensity Slicing

• Input: greyscale image, number of slices

Intensity Slicing

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

Intensity Slicing

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Red

Green

Blue

Intensity Slicing

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Color3 Color4 Color5 Color6(120,32,92)Color1

(80,200,130)Color2

Red

Green

Blue

Intensity Slicing

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

(120,32,92)Color1

(80,200,130)Color2 Color3 Color4 Color5 Color6

30

Lies in interval 1

Red

Green

Blue

Example

10 Slices

8 Slices

2. Intensity to Color Transformations

• Input: greyscale image, number of slices, (phase1, phase2, phase3)

Color Transformations

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Red

Green

Blue

Color Transformations

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Red

Green

Blue

255*Sin (k + phase1)

255*Sin (k + phase2)

255*Sin (k + phase3)

Color Transformations

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Red

Green

Blue

255*Sin (k + phase1)

255*Sin (k + phase2)

255*Sin (k + phase3)

50Lies in interval 2

Center = 63.5Color = 255*Sin (63.5 + phase1)255*Sin (63.5 + phase2)255*Sin (63.5 + phase3)

Color Transformations

Greyscale image (h,w,1)

Output color image ((h,w,3)

Slices = 5

0 255~42 ~85 ~127 ~170 ~212

Red

Green

Blue

255*Sin (k + phase1)

255*Sin (k + phase2)

255*Sin (k + phase3)

50Lies in interval 2

Center = 63.5Color = 255*Sin (63.5 + phase1)255*Sin (63.5 + phase2)255*Sin (63.5 + phase3)

Examples

Slices = 20Phase1 = 20Phase2=70Phase3=270

Assignment -4

1. Image Restoration – 75 Pts2. Color Image Processing – 75 Pts

Total: 150 Pts.

Submission Instructions

Must use the starter code available in Github Submission allowed only through Github You will receive an email with invitation to join

Github classroom Start by reading the readme.md file.

Instructions are available here Github will automatically save the last commit

as a submission before the deadline