Download presentation
Presentation is loading. Please wait.
Published byAdela Short Modified over 6 years ago
1
Morphological Transformations and Histogram Equalization
Dr. Rongzhong Li CSC391/691 Fall 2016 CameraCalibration Example Images are from: OpenCV-Python Documentation
2
Input + Neighbor + Rule = Output
Cellular Automaton Input Neighbor Rule = Output 1 1 && || 1 1 Growth of Crystal, Replication of bacteria, Complex rules=> randomness, Stephen Wolfram Conway's Game of Life
3
Kernel Shape 1 1 1 Complexity? Optimization?
4
Morphological Transformations
Transform(Origin, Kernel) Erosion Dilation Opening = Erosion + Dilation Closing = Dilation + Erosion Morphological Gradient = diff( Dilation, Erosion) Top Hat = diff(Origin, Opening) Black Hat = diff(Origin, Closing) Original image Kernel 1
5
Erosion A pixel in the original image (either 1 or 0) will be considered 1 only if all the pixels under the kernel is 1, otherwise it is eroded (made to zero).
6
Dilation Opposite of erosion. A pixel element is '1' if at least one pixel under the kernel is '1'.
7
1st Level Combination Opening = Erosion + Dilation Closing = Dilation + Erosion
8
Morphological Gradient = diff(Dilation, Erosion) Top Hat
2nd Level Combination Morphological Gradient = diff(Dilation, Erosion) Top Hat = diff(Origin, Opening) Black Hat = diff(Origin, Closing)
9
Generating Examples from Scratch
Reduce problem complexity
10
Results of Combinations
11
Histogram Threshold 4 128
12
Histogram Threshold 50
13
Histogram Equalization
50 150
14
Enhancing Details
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.