Image Filtering

1
Image Filtering
2
Problem!

• Noise is a problem, even in images!

Gaussian Noise
Salt and Pepper Noise
3
What to do?

• Try to think of ways to remove the noise!

Gaussian Noise
Salt and Pepper Noise
4
Smoothing

• The most obvious and easy way to remove the noise
  is by averaging.
• We can average pixels of a specific neighborhood
  by convolution with an averaging filter.

5
Smoothing

• Convolution is a simple mathematical operation.

6
Smoothing

• Smoothing using 3x3 mean filter.

Gaussian Noise
Salt and Pepper Noise
7
Gaussian Filter

• Gaussian filter is more natural!

8
Gaussian Filter

• Gaussian mask with s1.

9
Gaussian Filter

• Gaussian filter is more natural!

Gaussian Noise
Gaussian 5x5 Filter
10
Median Filter

• Median filter operation.

11
Median Filter

• Removing Salt and Pepper Noise using Median
  Filtering.

Salt and Pepper Noise
3x3 Median Filter
12
Edge Detection

• Finding the edges of images is a very useful
  tool.
• How do we find the edges in images?

13
Edge Detection

• Yes youre right, derivatives!!

14
Edge Detection

• We use image filtering to find edges.
• There are many masks Prewitt, Sobel, Roberts

Roberts
Sobel
Prewitt
15
Edge Detection

• The gradient magnitude and gradient angle can be
  calculated as follows