图像分割中基于过渡区技术的统计调查  

A Survey on Transition Region-Based Techniques for Image Segmentation

作　　者：章毓晋[1] 

机构地区：[1]清华大学电子工程系,北京100084

出　　处：《计算机辅助设计与图形学学报》2015年第3期379-387,共9页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61171118);教育部高等学校博士学科点专项科研基金(SRFDP-20110002110057)

摘　　要：第一个基于过渡区的图像分割方法在四分之一个世纪前提出后得到了许多图像技术同行的关注,至今已有很多应用,并得到进一步的研究且取得了一系列研究成果.借助谷歌学术工具对与此相关的文献进行了网络搜索,并以搜索到的文献为基础对近年来相关的研究进展进行全面的统计和回顾:包括对引用文献进行了逐年统计,对其中有进一步研究内容的文献进行了分类分析,对研究过程中出现的几个问题进行了剖析讨论,还指出了若干进一步研究的可能方向.从中不仅可了解到基于过渡区图像分割方法的发展情况、基本特点、涉及领域以及存在的瓶颈,还可看出一些未来深入研究的趋势和方向.The first image segmentation method based on the determination of transition region proposed a quarter of a century ago has attracted many attentions from the community of image techniques. Till now, a lot of applications have been reported, and further researches have been conducted and some research results have been obtained. This paper uses Google Scholar as a tool for searching related literatures, and makes a complete review of progress for this method on the basis of the search results. This includes the yearly counts of citing articles, the classification and analysis of further research articles, the discussions on appearing problems, and the potential directions for further researches. From it, not only the information for the developments, basic principle, related areas and existing problems of the method are presented, but also the tendency for further research tasks is provided.

关 键 词：图像分割 过渡区 灰度变化幅度 灰度变化频率 方法比较 文献统计 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]