A Graph Cuts Based Interactive Image Segmentation Method
Liu Jia①②③; Wang Hong-qi①
①Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;②National Key Laboratory of Microwave Imaging Technology, Beijing 100190, China; ③Graduate University, of Chinese Academy of Sciences, Beijing 100039, China
Abstract:Interactive image segmentation methods have recently gained more and more attentions. A new interactive segmentation method is proposed based on the graph cuts. It combines several image features like texture, color and edge together through a probabilistic model. Texture and color features are modeled with histograms. Dimensionality reduction in feature space is achieved with a fisher discriminant criterion based on texton. The global optimal segmentation can be efficiently computed via graph cuts. Efficiency and accuracy of the method is demonstrated on aerial image segmentation and some other applications.