EE4328, Section 005 Introduction to Digital Image Processing Image Segmentation Zhou Wang Dept. of Electrical Engineering The Univ. of Texas at Arlington Fall 2006
Concepts and Approaches What is Image Segmentation? Image Segmentation Methods Thresholding Boundary-based Region-based: region growing, splitting and merging Partition an image into regions, each associated with an object but what defines an object? From Prof. Xin Li
From [Gonzalez & Woods] Thresholding Method thresholding From Prof. Xin Li histogram single threshold multiple thresholds From [Gonzalez & Woods]
From [Gonzalez & Woods] Thresholding Method Global Thresholding: When does It Work? From [Gonzalez & Woods]
From [Gonzalez & Woods] Thresholding Method Global Thresholding: When does It NOT Work? A meaningful global threshold may not exist Image-dependent global thresholding From [Gonzalez & Woods]
Thresholding Method true object boundary Thresholding T = 4.5
Thresholding Method Solution Spatially adaptive thresholding Localized processing Split
spatially adaptive threshold selection Thresholding Method spatially adaptive threshold selection Thresholding T = 4 Thresholding T = 7 Thresholding T = 4 Thresholding T = 7
merge local segmentation results Thresholding Method merge local segmentation results merge merge merge merge
Boundary-Based Method edge detection boundary detection classification and labeling image segmentation From Prof. Xin Li
Boundary-Based Method Advanced Method: Active Contour (Snake) Model Iteratively update contour (region boundary) Partial differential equation (PDE) based optimization From Prof. Xin Li
Region-Based Method: Region Growing Key: similarity measure Start from a seed, and let it grow (include similar neighborhood) Key: similarity measure From [Gonzalez & Woods]
Region-Based Method: Split and Merge Iteratively split (non-similar region) and merge (similar regions) Example: quadtree approach From [Gonzalez & Woods]
Region-Based Method: Split and Merge Example: Quadtree Split and Merge Procedure Iteration 1 split merge original image 4 regions 4 regions (nothing to merge) Split Step split every non-uniform region to 4 Merge Step merge all uniform adjacent regions
Region-Based Method: Split and Merge Example: Quadtree Split and Merge Procedure Iteration 2 split merge from Iteration 1 13 regions 4 regions Split Step split every non-uniform region to 4 Merge Step merge all uniform adjacent regions
Region-Based Method: Split and Merge Example: Quadtree Split and Merge Procedure Iteration 3 split merge from Iteration 2 10 regions 2 regions Split Step split every non-uniform region to 4 Merge Step merge all uniform adjacent regions final segmentation result
Hard Problem: Textures Similarity measure makes the difference From Prof. Xin Li