用户名: 密码: 验证码:
数字图像处理中分割方法的研究与实现
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
数字图像处理是对在空间上离散、幅度上量化分层的图像进行某些特定数理模式加工处理的过程。而图像分割是把图像分成各具特性的区域并提出感兴趣目标的技术和过程,图像分割在图像工程中占有重要的位置。
     本文以湖北省科技厅重点科技发展计划项目资助课题——“智能运输系统的视频信号采集及识别算法研究”为背景,以课题中数字图像处理部分主要环节之一——车牌图像的分割问题为主要研究目标,在参阅大量文献资料的基础上,对数字图像分割方法进行了学习和研究,并结合特定的数学理论,如数学形态学、小波分析等,对车牌图像分割方法进行了着重地探讨与实现。图像处理过程在去噪的同时,尽量保留边缘特征是图像分割首要顾及的问题。本文采用小波多分辨分析和阈值去噪的处理方法,取得了较好的效果。为了将车牌区域从复杂背景中分割出来,需要在一定程度模仿人类视觉系统在分割与识别图像时具有的对多种信息综合运用的能力。小波多分辨率分析方法的主要思路是在不同的分辨率层次上对图像进行分割,粗分割结果对精分割具有一定的指导作用。数学形态学是一种非线性滤波方法,其特点是能将复杂的形状进行分解,并将有意义的形状分量从无用的信息中提取出来。利用数学形态学这一特性来对小波分解出来的图像的细节分量进行处理,在计算开销和定位精度这一对矛盾上可达到较好的平衡。针对智能运输系统视频信号采集识别中车牌图像的分割问题,本文提出了结合运用小波多分辨分析和数学形态学来处理图像分割问题的实验方案。实验结果证明该方案是可行和有效的,具有一定的实用价值。
     本文主要分六部分。第一部分介绍图像处理的分类、特点和相关技术,图像分割在图像工程中的位置;第二部分以图像变换的有关数学理论为基础分析了数字图像处理的特点与优势;第三部分论述了典型的数字图像预处理的方法;第四部分阐述了各种图像分割的主要算法原理,并对它们进行了比较、归纳、分类,着重研究了小波分析和数学形态学等在图像分割中的应用;第五部分简要介绍了数字图像分割的程序设计常规工具,提出了综合运用多种图像信息的图像分割实验方案;最后在第六部分对所做的主要工作进行了总结与展望。
Digital image processing is a mathematical procedure which calculated by computer. Image segmentation is a technology that divide image into characteristic areas and put forward the interested target. Image segmentation is a key step from digital image processing and analysis, so it has an important position in the technology of the image processing.
    The background of this paper is "the research on the recognition algorithm of video signal sampling in ITS", which is a scientific & technical development plan of HuBei province. Image segmentation mainly has following several treatments. Firstly, proceeding from the whole graph. Judge each pixel's classification based on collected groups of attribute spaces, finally get the areas. Secondly, setting out from pixels. Form an area by collecting near pixels with unanimous attribute. Thirdly, division-collection method. Divide the image into several regular blocks, don't separate the blocks without consistent attribute and unite the blocks with the same attribute, until forming a district map. The ability that man can distinguish different goals from the complicated scenery quickly, at least partly benefit from many kinds of information in the image, such as the grey level, border, texture, etc. It illumine people to create methods on how to use mis information to segmentation. Edge information is the most important high frequency information of an image. Therefore we should try to maintain more edge information in the process of denoising. We present a new image denoising and segmentation method based on wavelet analysis and mathematical morphology. In some extent, It simulate the ability of human vision system by this means. The experiment results show that the segmentation method is effective, it has some practicality.
    This thesis can divide into six parts. The first part is to introduce the classification, characteristic, and relevant technology of image processing, and segmentation's position in the image engineering. The second part will analyse the characteristic and advantage of digital image transformation, basing on the relevant mathematics theories. The third part is to introduce several typical digital pretreatment of the image. The fourth part will put an emphasis on various kinds of main algorithm of the image segmentation, followed by the comparison, the conclusion, and classification. The fifth part will introduce some general tools of program designs for image processing, and put forward a segmentation experiment method on vehicle license plates. The last part is a summary and prospect.
引文
[1] 田村秀行(日),计算机图像处理技术。赫荣威等译,北京:北京师范大学出版社,1988
    [2] 李介谷等,图像处理技术。上海:上海交通大学出版社,1988
    [3] 吴健康,数字图像分析。北京:人民邮电出版社,1989
    [4] 荆仁杰等,计算机图像处理。杭州:浙江大学出版社,1990
    [5] 徐建华,图像处理与分析。北京:科学出版社,1992
    [6] 章毓晋,图像工程(上册)——图像处理和分析。北京:清华大学出版社,1999
    [7] 章毓晋,图像工程(下册)——图像理解与计算机视觉。北京:清华大学出版社,1999
    [8] 周凌翔,顾伟康,最佳边缘检测的准则与算子。模式识别与人工智能,1998,11(1)
    [9] 傅卓,章毓晋,一种新的亚象素边缘检测方法及其性能研究。电脑应用技术,1995,35:1-5
    [10] 狄宇春,邓雁萍,关于图像分割性能评估的评述。中国图像图形学报,1999,1(2):183-187
    [11] 吴敏金,图像形态学。上海科技文献出版社,1991。
    [12] 余松煜,数字图像处理。电子工业出版社,1989。
    [13] 唐常青等,数学形态学的方法及应用。科学出版社,1990。
    [14] 崔屹,数字图像处理技术及应用。电子工业出版社,1997。
    [15] 朱秀昌等,数字图像通信。北京:人们邮电出版社,1994。
    [16] 吴成柯等,图像通信。西安:西安电子科技大学出版社,1994。
    [17] 郭德方,遥感图像的计算机处理和模式识别。北京:电子工业出版社,1987。
    [18] 李月景,图像识别技术及应用。北京:机械工业出版社,1985。
    [19] 王润生,图像理解。长沙:国防科技大学出版社,1995。
    
    
    [20]徐建华,图像处理与分析。北京:科学出版社,1992。
    [21]何兵,马天予,王运坚,朱红莲。Visual C~(++)数字图像处理。北京人民邮电出版社,2001。
    [22]阮秋琦。数字图像处理学。北京:电子工业出版社,2001。
    [23]吴一全,朱兆达,图像处理中阈值选取方法30年(1962-1992)的进展(一),数据采集与处理,1993,8(3):193-201
    [24]吴一全,朱兆达,图像处理中阈值选取方法30年(1962-1992)的进展(二),数据采集与处理,1993,8(4):268-281
    [25]赵荣椿,迟耀斌,朱重光,图像分割技术进展,中国体视学和图像分析,1998,3(2)
    [26]罗希平等,图像分割方法综述,模式识别与人工智能,1999,12(3):300-312
    [27]章毓晋,罗惠韬,基于评价知识的图像分割算法优选系统,高技术通讯,1998,8
    [28]张贤达.现代信号处理.北京:清华大学出版社.1995.156~198
    [29]陈桂明,张明照,成红雨.应用MATLAB语言处理数字信号与图像.北京:科学技术出版社.2000.273~307
    [30]Duane Hanselman, Bruce Littlefield. 精通MATLAB(,李人厚,张平安等).西安:西安交通大学出版社,1998.155~166
    [31]姚天任,江太辉.数字信号处理.武汉:华中理工大学出版社,1998.172-199
    [32]冯宗哲,程相君.模式识别原理.西安:西安电子科技大学出版社,1993.
    [33]张静远,张冰.基于小波变换的特征提取方法分析.信号处理,2000,16(2)
    [34]姚玉荣,章毓晋,利用小波和矩进行基于形状的图象检索,中国图象图形学报,5A(3):206~210
    [35]尹平,王润生,基于边像信息的分开合并图象分割方法,中国图象图形学报,1998,3(6):450~454
    [36]丁震等,一种适用于灰图象分割的快速FCM算法。摸式识别与人工智能,1997,10(2):132~151
    
    
    [37]姬光荣,王国宇,王宁,基于小波交换的多尺度边缘检测,中国图象图形学报,1997,2(10):717~720
    [38]薛景浩,章毓晋,林行刚,图像分割中的交叉熵和模糊散度算法。电子学报,1999,27(10):131~134
    [39]Guy Eddon, Henry Eddon.组件编程技术(希望图书创作室).北京:北京希望电子出版社,1998
    [40]常炯.信息理论基础.北京:清华大学出版社,1993.32~46
    [41]许传祥,石青云,程民德,零对称和反对称二进小波及其在边缘检测中的应用,中国图象图形学报,1996,1(1):4~11
    [42] Zhang Y J,Gerbrands J J,Objective and quantiative and quantitative segmentation evaluation and comparison. Signal Processing, 1994,39:43-84
    [43] Davis L S.A survey of edge detection techniques. CGIP, 1975,4:248-270
    [44] Riseman E M,Arbib M A. Computational techniques in the visual segmentation of static scenes. CGIP, 1977,6:221-276
    [45] Riseman E M,Arbib M A. Computational techniques in the visual segmentation of static scenes. CGIP, 1977,6:221-276
    [46] Wesaka J S. A survey of threshold selection techniques CGIP, 1978,7:259-265
    [47] Fu K S, Mui J K. A survey on image segmentation .Pattern Recognition, 1981,13
    [48] Peli T, Malah D. A study of edge determination algorithms. CGIP, 1982,20:1-20 (36).
    [49] Haralick RM,Shapiro LG.Image segmentation techniques. CVGIP, 1985,29:100-132 (37).
    [50] Pavlidis, T. Critical survey of image analysis methods. Proceedings of CPR, 1986
    [51] Borisenko V I,Zlatotol A A, Muchnik I B. Image segmentation(state of the art survey). Automatic Remote Control,1987,48:837-879
    [52] Sahoo P K,et al. A survey of thresholding techniques. CVGIP, 1988, 41: 233-260
    
    
    [53] Buf J M H,Campbell T G. A quantitative comparison of edge-preserving smoothing techniques. Signal Processing, 1990,21:289-301
    [54] Sarkar S, Boyer K L.Perceptual organization in computer vision:a review and a proposal for a classificatory structure,IEEE SMC, 1993,23:382-399
    [55] Pal N R,Pal S K. A review on image segmentation techniques. Pattern Recognition, 1993,26:1277-1294
    [56] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation. IEEE-PAMI, 1989,11(7):647~693
    [57] Hong TH. Rosenfeld A. Compact region extraction using weighted pixel linking in a pyramind. IEEE-PAMI, 1984,6(2):222~237
    [58] Wu GH, Zhang YJ, Lin XG. Wavlet transform-based texture classification using feature weighting. Proceeding of ICIP'99, 1999, 28ap5b.3
    [59] Javier LN, Mario HT. Image segmentation using a modified split and merge technique. Cybernetics and Systems, 1994, 25(1):137~162
    [60] M. Mrak and N.Sprljan, Disrete cosine transform in image coding applications, Dept. Radiocommun. Microwave Eng., Univ. Zagreb, Internal Pub., 1999.
    [61] S. Grgic, M. Grgic, and B. Zovko-Cihlar, Picture quality measurements in wavelet compression system, in Int.broadcasting Convention Conf. Pub. IBC99, Amsterdam, The Netherlands,1999, pp. 554-559
    [62] K. Muramoto, T. Yamauchi: Classification of polar satellite data using image features and decision tree classifier, Proc. NIPR Symp. Polar Meteorol. Glaciol., 10, p.127-137, 1996

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700