数字图像处理课件 16-17周-lec-10-Image Segmentation.pptx

Image Segmentation;Psychology: The Gestalt school;Content;Introduction to image segmentation;Introduction to image segmentation;Introduction to image segmentation;Introduction to image segmentation;Introduction to image segmentation;Introduction to image segmentation;Content;11;12;Content;The goals of segmentation;The goals of segmentation;Content;Performance indicators;Performance indicators;Performance indicators;Content;Segmentation as clustering;;EE465: Introduction to Digital Image Processing Copyright Xin Li2004;24;25;Toy Example of Kmeans Clustering;;;;;K-means Clustering: Step 5;32;K-means Properties;K-Means Example 1;K-Means Example 2;K-Means Example 3;Apply K-means to RGBXY features;slide from Naotoshi Seo; K-means – common extensions K-means → GMM;Content;;The mean shift algorithm seeks modes or local maxima of density in the feature space;;;;;;;Cluster: all data points in the attraction basin of a mode Attraction basin: the region for which all trajectories lead to the same mode;Find features (color, gradients, texture, etc) Initialize windows at individual feature points Perform mean shift for each window until convergence Merge windows that end up near the same “peak” or mode;;More results;More results;Content;;;Measuring affinity;;Graph cut;Minimum cut;Minimum cut;;;Normalized cut;Normalized cut;Normalized cut;Normalized cut;;;Example Segmentations;;Example results;Results: Berkeley Segmentation Engine;Content;Region Growing;Felzenszwalb and Huttenlocher’s Algorithm;Content;Torr Vision Group, Engineering Department;Torr Vision Group, Engineering Department;Torr Vision Group, Engineering Department;Torr Vision Group, Engineering Department;Torr Vision Group, Engineering Department;End