A simultaneous cartoon and texture segmentation method within the fuzzy framework

详细信息    查看全文

• 作者：Yafeng Li ^{liyafeng770729@126.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Image segmentation ; Image decomposition ; Variational model ; Wavelet ; Fuzzy method
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：12 July 2016
• 年：2016
• 卷：197
• 期：Complete
• 页码：161-170
• 全文大小：4650 K

文摘

This paper presents a new image segmentation method called simultaneous cartoon and texture segmentation (SCTS). The proposed method takes advantage of the decomposition of images into the cartoon and texture components and their respective properties. In the proposed model, each region is represented by a fuzzy membership function (FMF), and two data fidelity terms are jointly defined to measure the conformity of cartoon and texture components within image regions. In order to efficiently solve the proposed model, a fast alternative iteration algorithm for SCTS is presented. Most importantly, the proposed method has good selectivity to patterns of cartoon, texture and noise, which makes it increase the robustness and produce better results than the classical methods. Experimental results show that the proposed method has very promising segmentation performance.