A Simultaneous Cartoon-Texture Image Segmentation and Image Decomposition Method
A Simultaneous Cartoon-Texture Image Segmentation and Image Decomposition Method作者机构：School of Computer Science and Technology Baoji University of Arts and Science College of Computer Science Sichuan University School of Information Science and Technology Tibet University School of Electronics Engineering and Computer Science Peking University
出 版 物：《Chinese Journal of Electronics》 (电子学报(英文))
年 卷 期：2020年第29卷第5期
页 面：906-915页
核心收录：
学科分类：0808[工学-电气工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 0701[理学-数学]
基 金：supported by the National Natural Science Foundation of China (No.61971005) the Natural Science Basic Research Plan in Shaanxi Province of China (No.2018JM6043)
主 题：Image segmentation Image decomposition Variational model Dictionary learning
摘 要：Image segmentation and image decomposition are fundamental problems in image processing.Image decomposition methods for separating images into cartoon and texture components can effectively serve different image processing tasks because different components can be respectively treated in more effective way. However, image decomposition methods are currently simply taken as an independent preprocessing step, and particularly in image segmentation different effects of cartoon and texture components have not been considered.This paper presents a novel simultaneous cartoon-texture image segmentation and image decomposition method to boost the performance of both segmentation and decomposition. We design a fast alternating optimization algorithm to solve the proposed model. Experimental results demonstrate the outstanding performance of the proposed method on both image segmentation and image decomposition.
同方期刊数据库