DWT and MRF Model-based Unsupervised Texture Segmentation

摘要：

A new approach to unsupervised texture segmentation is proposed in this paper, which combines the advantages of both Discrete Wavelet Transform (DWT) theory and Markov Random Field (MRF) model. This algorithm based on DWT-NNC-MRF-MAP uses DWT and Nearest Neighborhood Clustering (NNC) to determine the number of texture classifications, uses DWT and MRF model to establish our texture segmentation model, and uses ICM criterion to realize the texture segmentation. Finally, the validity and practicability of our new algorithm are verified by the experiment results.

关键词： texture segmentation MRF model DWT

作者: Cui Wang Haiyan Sun

作者单位: LIMB of the Ministry of Education, School of Mathematics and Systems Science, Beihang University, Beijing

会议类型: 国际会议

会议名称: 2010 International Conference on Probability and Statistics of the International Institute for General Systems Studies(国际一般系统理论研究会中国概率统计学会第二届学术会议IIGSS-CPS2010)

会议地点: 南京

会议语种:英文

页码: 531-534

在线出版日期: 2010-07-29（万方平台首次上网日期，不代表论文的发表时间）