文献类型：会议论文

英文题名：GRAPH-BASED IMAGE SEGMENTATION WITH BAG-OF-PIXELS

作者：Chen, Zhi-Hua[1];Xiao, Xiao-Long[1];Liu, Yi[1];Zhang, Jing[1,2];Yuan, Yu-Bo[1]

机构：[1]E China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[2]Nanjing Univ, State Key Lab Novel Software Technol, Nanjing, Jiangsu, Peoples R China

会议论文集：International Conference on Machine Learning and Cybernetics (ICMLC)

会议日期：JUL 14-17, 2013

会议地点：Tianjin, PEOPLES R CHINA

语种：英文

外文关键词：Image Segmentation; Image Content; Graph-Based; Bag-of-Pixels

摘要：Image segmentation can help us with a better understanding of the image content, and thus plays an important role in many fields of image processing including image retrieval, object recognition and image compression. We propose a novel method based on an efficient Graph-Based segmentation after analyzing its irrationalities. The proposed method with bag-of-pixels is simple and effective. In the evaluation using a large public image database that contains a lot of natural images, our method outperforms the efficient graph-based segmentation method, yields better segmentation results and shortens the running time.