纹理特征遥感影像分割研究Texture-based segmentation for remotely sensed imagery
林小平,毛政元,刘建华
摘要(Abstract)：
影像分割是面向对象的分类思想应用于遥感影像信息自动提取的基础,纹理是影像的基本特征,是影像分析、理解和识别的重要信息,纹理特性的有效表达和抽取,是基于纹理影像分割的前提。本文系统概述了近年来各类文献中使用频率较高的基于纹理的遥感影像分割方法,并以基于统计的方法、基于纹理结构的方法、基于模型的方法和基于空间/频率的方法四种基本类型为主线,对每一类分割方法的特点进行了分析和总结,在此基础上指出了基于纹理特征遥感影像分割的研究趋势。
关键词(KeyWords)： 遥感影像;分割;纹理;小波变换
基金项目(Foundation): 国家973计划项目(2006CB708306);; 国家自然科学基金项目(40871206)
作者(Author): 林小平,毛政元,刘建华
DOI: 10.16251/j.cnki.1009-2307.2010.05.071
参考文献(References)：
- [1]章毓晋.图像分割[M].北京:科学出版社,2001.
- [2]吴高洪.纹理分析的若干多尺度方法研究:博士论文[D].北京:清华大学,1999.
- [3]王润生.图像理解[M].长沙:国防科技大学出版社,1995.
- [4]Sklansky J.Image segmentation and feature extraction[J].IEEE Trans.on Systems,Man and Cybernetics,1978.
- [5]郑晓曦,严俊龙.图像分割新方法综述[J].计算机与数字工程.2007,35(8).
- [6]朱述龙,钱曾波.基于小波包特征的纹理影像分割[J].中国图象图形学报,1998,3(8).
- [7]Haralick R M,Shanmugam K,Dinstein I.Textural fea-tures for image classification[J].IEEE Trans.on Sys-tems,Man,Cybernetics,1973,SMC-3,610-621.
- [8]Shapiro L G and Stockman G C.Computer Vision[M].Upper Saddle River,NJ:Prentice Hall,2001:233-234.
- [9]J torwaki,Y Yashima,S yokoi.Adjacency Graphs on adigitized figure set and their application to texture analy-sis[C]//Proc.7th Int.Conf.on Patt.Rec.Montre-al,Canada,July30-Aug.2,1984:1216-1218.
- [10]郝文化.MATLAB图形图像处理应用教程[M].北京:中国水利水电出版社,2003.
- [11]Xavier Munoz Pujol,Image segmentation integrating col-or,texture and boundary information[D].PhD The-sis,2002.
- [12]Geman S,Geman D.Stochastic Relaxation,Gibbs Dis-tributions,and the Bayesian Restoration of images[J].IEEE Trans.on Pattern Analysis and Machine Intelli-gence.1984,6(6).
- [13]Chellappa R,Chatterjee S.Classification of textures usingGaussian Markov random fields[J].IEEE Trans.OnAcoust.Speech and Signal Proc.ASSP,1985.33(4).
- [14]Cohen F S,Fan Z,Patel M A.Classification of rotatedand scaled textured images using Gaussian Markov ran-dom field models[J].IEEE Trans.on Pattern Analysisand Machine Intelligence,1991,13(2).
- [15]Philippe Andrey,Philippe Tarroux.Unsupervised Seg-mentation of Markov random Field Modeled Textured Ima-ges Using Selection Relaxation[J].IEEE Trans.on Pat-tern Analysis and Machine Intelligence,1998,20(3).
- [16]Zhang Y,Brady M.Segmentation of Brain ImagesThrough a Hidden Markov Random Field Model and theExpectation-Maximization Algorithm[J].IEEETrans.Med.Imag,2001,20(1).
- [17]Forbes F,Peyrard N.Hidden Markov Random FieldModel Selection Criteria Based on Mean Field-Like Ap-proximations[J].IEEE Trans.On Pattern Analysis andMachine Intelligence,2003,(25).
- [18]Fabien S,Wojciech P.Parameter Estimation in HiddenFuzzy Markov RandomFields and Image Segmentation[J].Graphical Models and Image Processing,1997,59(4).
- [19]Pentland A P.Fractal-based description of natural scene[J].IEEE-PAMI,1984,6(6).
- [20]Benoit B Mandelbrot.The fractal geometry of nature[M].San Francesco Freeman W H.1982.
- [21]杨绍国,尹忠科,罗炳伟.分形多尺度纹理图像分割[J].模式识别与人工智能,1998,11(3).
- [22]吴更石,梁德群,田原.基于分形维数的纹理图像分割[J].计算机学报,1999,(10).
- [23]吴高洪,章毓晋,林行刚.基于分形的自然纹理自相关描述和分类[J].清华大学学报,2000,40(3).
- [24]Beck J,Butter A,,Ivry R.Spatial frequency channelsand perceptual grouping in texture segregation[J].Comput.Vision.Graphics Image Process,1987,37.
- [25]Mallat S G.Multifrequency channel decompositions of im-ages and wavelet models[J].IEEE Trans.on Acous-tics,Speech,and Signal Processing,1989a,37(12).
- [26]朱超波,陈康力,郑小松.基于小波包框架的纹理分割[J].模式识别与人工智能,2000,13(4).
- [27]吴高洪,章毓晋,林行刚.利用小波变换和特征加权进行纹理分割[J].中国图象图形学报,2001,6A(1).
- [28]Acharyya M,De R K,Kundu MK.Segmentation of re-motely sensed images using wavelet features and their eval-uation in soft computing framework[J].IEEE Trans.on Geoscience and Remote Sensing,2003,41(12).
- [29]Bovik A C,Clark M L and Geisler W S.Multichanneltexture analysis using localized spatial filters[J].IEEETrans.Pattern Anal.Mach.Intel.1990-01,12:55-73.
- [30]李厚强,王超,等.一种改进的基于Gabor滤波器的纹理分割方法[J].电路与系统学报.2003,(6).