图像纹理分类方法研究进展和展望  

Texture Classification:State-of-the-art Methods and Prospects

作　　者：刘丽[1] 赵凌君[2] 郭承玉 王亮[3] 汤俊[1] LIU Li;ZHAO Ling-Jun;GUO Cheng-Yu;WANG Liang;TANG Jun(College of Information System and Management, National University of Defense Technology, Changsha 410000;College of Electronic Science and Engineering, National University of Defense Technology, Changsha 410000;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing 100190)

机构地区：[1]国防科学技术大学信息系统与管理学院,长沙410000 [2]国防科学技术大学电子科学与工程学院,长沙410000 [3]中国科学院自动化研究所模式识别国家重点实验室,北京100190

出　　处：《自动化学报》2018年第4期584-607,共24页Acta Automatica Sinica

基　　金：湖南省自然科学基金杰出青年基金(2017JJ1007)资助~~

摘　　要：纹理分类是计算机视觉和模式识别领域的一个重要的基本问题,也是图像分割、物体识别、场景理解等其他视觉任务的基础.本文从纹理分类问题的基本定义出发,首先,对纹理分类研究中存在的困难与挑战进行阐述;接下来,对纹理分类方面的典型数据库进行全面梳理和总结;然后,对近期的纹理特征提取方法的发展和现状进行归类总结,并对主流纹理特征提取方法进行了详细的阐述和评述;最后,对纹理分类发展方向进行思考和讨论.Texture is a fundamental characteristic of many types of images. Texture classification is one of the essential tasks in the field of computer vision and pattern recognition. It is also the basis of other complex vision tasks, such as image segmentation, object recognition and scene understanding. In this paper, we first address the importance of texture classification and summarize the difficulties and challenges in the development of texture feature extraction approaches.Then we discuss the existing texture databases which are generally acknowledged as public evaluation bases for texture classification methods. Next, we review recent achievements in the study of texture feature development and provid detail discussion on prominent texture feature descriptors. Finally, we point out the future directions of texture classification.

关 键 词：纹理分类 特征提取 深度学习 局部特征描述 计算机视觉 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]