New Approach for Image Segmentation Based on Graph Cuts
Jun Kong, Kun Gao, Min Jiang, Chenhua Liu, Yan Li · International Journal of Signal Processing Image Processing and Pattern Recognition · 2017
The paper proposed a novel image segmentation algorithm based on the improved affinity propagation algorithm and graph cuts.Firstly, the similarity matrix of the improved affinity propagation algorithm is constructed by using three fundamental features of the image, and each feature is assigned weight according to their distribution in the image.So the improved affinity propagation algorithm can be implemented to cluster the image into numbers of high quality regions.Secondly, these high quality regions are represented by suitable models and these models are selected as labels to construct the data term and smooth term of the energy function.According to the energy function, corresponding weights are assigned to the edges of the graph.Finally, the mincut/max-flow algorithm is used to search for the minimum cut of the weighted graph and get the final segmentation results.The segmentation results of the proposed algorithm are evaluated through probabilistic rand index and global consistency error methods.It is shown that the presented segmentation method provides effective results in terms of both accuracy and computational efficiency.