COMP 9517 WK6
2021 T1
Introduction
Introduction
Introduction
• Segmentation approaches • Region based
• Contour based
• Template matching based
• Splitting and merging based • Global optimisation based
Introduction
• Issues and challenges
• So far there is no single segmentation method working well for all problems
• Special domain knowledge of the application is typically essential for the development of successful computer vision methods for segmentation
Introduction
• Results from several popular segmentation techniques • Active contours
• Level sets
• Graph-based merging
• Mean shift
• Normalised cuts • Binary MRF
Outline
Thresholding
• Fine if regions have sufficiently different intensity distributions • Problematic if regions have overlapping intensity distributions
K-Means Clustering
• Could work if the number of clusters is known a priori
• Problematic if the number of clusters Is not know a priori
Feature Based Classification
Region Splitting and Merging
Region Splitting and Merging
Connected Components
Connected Components Algorithm
Region Splitting
Region Splitting
Region Splitting
Region Splitting
Region Merging
Region Merging
Region Merging
Watershed Segmentation
Watershed Segmentation
Watershed Segmentation
Maximally Stable Extremal Regions
Mean Shifting
Mean Shifting
Mean Shifting
Mean Shifting
Superpixel Segmentation
Superpixel Segmentation
Superpixel Segmentation
Conditional Random Field
Conditional Random Field
Conditional Random Field
Active Contour Segmentation
Active Contour Segmentation
Active Contour Segmentation
Active Contour Segmentation
Level Set Segmentation
Evaluation Metrics
Evaluation Metrics
Basic set operations
Dilation of binary images
Definition of binary erosion
Structuring elements
Opening of binary images
Closing of binary images
Morphological edge detection
Binary object outlines
Ultimate erosion and reconstruction