图像边缘检测及柔性形态学在其中的应用
|
摘要
图像边缘检测技术是图像分割、图像增强、图像复原、模式识别、图像压缩等图像分析和处理的基础,也是图像处理领域的研究热点问题。
数学形态学是建立在集合论的基础上,用于图像处理和模式识别领域的新的方法,它是用一定形态的结构元素去度量和提取图像中的对应形状以达到对图像分析和识别的目的。而柔性形态学是数学形态学发展的一个重要分支。它具有较强的抗噪声、抗干扰能力和鲁棒性。本文首先对传统和新兴的边缘检测理论和方法加以介绍,对不同算法进行了仿真,然后系统地介绍了数学形态学和柔性形态学的基本理论。本文基于柔性形态学做了以下研究。
将粒子群优化算法引入到柔性形态学边缘检测中,针对单一尺度的结构元素,寻求最优的柔性形态学边缘检测算子,经实验证明该方法简单适用,效果理想。但当原图像中所含噪声较严重时,单一元素的柔性形态学边缘检测算法抗噪性比多尺度结构元素的要差,所以针对多尺度结构元素的柔性形态学算子在图像边缘检测中的应用进行了研究。经过大量的实验证明,本文所提算法在含有大量噪声的干扰下,比传统的方法能更好的抑制噪声并提取有用的图像边缘信息。
最后,本文对柔性形态学边缘检测算法在实际中的广泛应用进行了总结。
Edge detection is the foundation of feature description, image segmentation, image enhancement, pattern recognition and image compression and so on. Also edge detection is a hot topic in digital image processing.
Mathematical morphology founded on set theory is applied in the field of image processing and pattern recognition. Its main principle is to use a certain structuring element to measure and extract the conesponding form in an image so that the image can be analyzed and recognized. Soft morphology is an important subclass of development of mathematical morphology. This makes more tolerant and robust to noise. In this paper, the traditional theory and new method are introduced at first, then many different algorithms are simulated.Then the basic theories of morphology and soft morphological are introduced systemically. Some researches are made based on the soft morphology theories as the follows.
This paper applies the PSO to the soft morphology to select soft morphology of the best image edge detection method aimed at a single size structure element. The method is approved to be simple, suitable and effective. But when a large number of noises are included in an original image, the restraining noises capability of soft morphology with single structure element is worse than that with multi-scale structure elements. So the applications of soft mathematical morphology with multi-scale structure elements are researched in the image edge detection. It was proved by large numbers of experiments that the algorithm proposed by the paper is better to restrain noises and extract useful image edge information than traditional methods under the interrupt of a great many of noises.
Finally, the extensive applications of soft morphology in practice are summarized.
数学形态学是建立在集合论的基础上,用于图像处理和模式识别领域的新的方法,它是用一定形态的结构元素去度量和提取图像中的对应形状以达到对图像分析和识别的目的。而柔性形态学是数学形态学发展的一个重要分支。它具有较强的抗噪声、抗干扰能力和鲁棒性。本文首先对传统和新兴的边缘检测理论和方法加以介绍,对不同算法进行了仿真,然后系统地介绍了数学形态学和柔性形态学的基本理论。本文基于柔性形态学做了以下研究。
将粒子群优化算法引入到柔性形态学边缘检测中,针对单一尺度的结构元素,寻求最优的柔性形态学边缘检测算子,经实验证明该方法简单适用,效果理想。但当原图像中所含噪声较严重时,单一元素的柔性形态学边缘检测算法抗噪性比多尺度结构元素的要差,所以针对多尺度结构元素的柔性形态学算子在图像边缘检测中的应用进行了研究。经过大量的实验证明,本文所提算法在含有大量噪声的干扰下,比传统的方法能更好的抑制噪声并提取有用的图像边缘信息。
最后,本文对柔性形态学边缘检测算法在实际中的广泛应用进行了总结。
Edge detection is the foundation of feature description, image segmentation, image enhancement, pattern recognition and image compression and so on. Also edge detection is a hot topic in digital image processing.
Mathematical morphology founded on set theory is applied in the field of image processing and pattern recognition. Its main principle is to use a certain structuring element to measure and extract the conesponding form in an image so that the image can be analyzed and recognized. Soft morphology is an important subclass of development of mathematical morphology. This makes more tolerant and robust to noise. In this paper, the traditional theory and new method are introduced at first, then many different algorithms are simulated.Then the basic theories of morphology and soft morphological are introduced systemically. Some researches are made based on the soft morphology theories as the follows.
This paper applies the PSO to the soft morphology to select soft morphology of the best image edge detection method aimed at a single size structure element. The method is approved to be simple, suitable and effective. But when a large number of noises are included in an original image, the restraining noises capability of soft morphology with single structure element is worse than that with multi-scale structure elements. So the applications of soft mathematical morphology with multi-scale structure elements are researched in the image edge detection. It was proved by large numbers of experiments that the algorithm proposed by the paper is better to restrain noises and extract useful image edge information than traditional methods under the interrupt of a great many of noises.
Finally, the extensive applications of soft morphology in practice are summarized.
引文
铩颷1]陈书海,傅录祥.实用数字图像处理.北京:科学出版社,2005
[2]姚敏等.数字图像处理.北京:机械工业出版社,2006.1
[3]章毓晋.图像处理和分析.北京:清华大学出版社,1999.2
[4]T.Hebert,D.Malagre.Edge detection using a priori model.Proceedings of IntelConference on Image Processing.Austin,Texas,USA,1994,94:303-307
[5]Geman,S.Geman,C.Graffigne,and P.Dong.Boundary detection constraintoptimization.IEEE.Pattern Analysis and Machine Intelligence,1990,12(7):609-628
[6]胡学龙,许开宇,章毓晋.数字图像处理.北京:电子工业出版社,2006
[7]Koskinen,Astola J,Neuvo Y.Soft morphological fillers.SPIE Svmp.In:ImageAlgebra and Morphological Image Processing Ⅱ,San Doego,USA,1991,262-270
[8]A.Gasteratos and I.Andreadis,Soft Mathematical Morphology.Adances inImaging and Electron Physics.2001.12
[9]Pavel A.Maltsev.Soft morphological image resizing.SPIE,97,176-183
[10]Lasse Koskinen,Heikki Huttunen,and Jako Astola.Tampere University ofTechnology.Document Recognition(1994),251
[11]马义德.一种彩色柔性形态模糊模型的设计及其实现.电路与系统学报.2006.6.3(11)
[12]张小琳.图像边缘检测技术综述.高能量密度物理.2007.3(1)
[13]Marr D,Hildreth E.Theory of Edge Detection.Proc.R.Soc.London,Ser.B.1980,207:187-17
[14]Marr D.Vision.San Francisco:Freeman,1982
[15]夏良正.数字图像处理.南京:东南大学出版社,1999
[16]邓鲁华,张延恒等.数字图像处理.北京:机械工业出版社,2004
[17]贾永红.数字图像处理.武汉:武汉大学出版社,2003
[18]崔旭东,邱春蓉,刘瑞根.用标记松弛方法检测闪光图像边缘.光电工程.2001.8.28(4)
[19]Rosenfeld A,Hummel R.A,Zuker SW.Scene labeling by relaxation operations.IEEE Trans.Syst.M an Cybern,1976,SMC26:420-453
[20]Hummel R.A,Zuker.SW.On the foundations of relaxation labeling processes.IEEE Trans.Patt.A nal.M ach.Intell.1983,125:267-287
铩颷21]周猛.图像边缘检测技术在车锁识别打码系统中的应用研究.合肥工业大学.2006,3
[22]冯会真,夏哲雷,林志一.基于神经网络的图像边缘检测方法.中国记量学院学报.2006.12.17(4)
[23]K Etemad,R Chellappa.A Neural Network Based Edge Detector.ProcIJCNN.1993,132-137
[24]Xiaofang Sun,D.C Lowe.Learning Optional Linear Filters for Early Vision.ProcIJCNN,1991,2570-2575
[25]ShiZhenghuo and FungKOB.A Comparison of Digital Speckle Filters.Geoscience and Remote Sensing Symposium.2004.6(29)41-44
[26]陆雨花,章勇.图像去噪技术探讨.科技经济市场.2007.10
[27]杨冬云,李国辉.一种适合于边缘保存的混合噪声抑制方法.哈尔滨:哈尔滨理工大学学报.2005.4.10(2)
[28]胡永刚.基于数学形态学的遥感图像边缘检测算法.河北:燕山大学硕士学位论文.2005.9
[29]刘先锋,沈胜宏,郑明洁.边缘保留的图像噪声滤除方法.计算机应用.2000.11
[30]韩晓微等.一种基于脉冲噪声检测的图像均值滤波方法.计算机工程与应用.2004,27:102-104
[31]刘积平,余英林.一种简易的图像去噪方法.广东:华南理工大学学报.2000,28(2):60-63
[32]唐常青,吕宏伯,黄铮等.数学形态学方法及其应用.北京:科学出版社,1990
[33]华续钊,彩色图像边缘检测相关算法研究,扬州大学硕士学位论文,2004.4
[34]赵于前,陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[35]L.Koskinen,J.Astola,Y Neuvo.Soft Morphological Filters.Proc.SPIE.Symp.Image Algebra and Morphological Image Processing Ⅱ,1991.1.262-270
[36]于子翊,李晓娟,柏玲.柔性形态边缘检测算法在菌类显微图像中的应用.北京:首都师范大学学报.2006.4
[37]Berhart,R.C,and Kennedy,J.A new optimizer using particles swarm theory,Proc.SixthInternational Symposium on Micro Machine and Human Science,Nagoya,Japan,IEEE ervice Center,Piscataway,1995,39-43
[38]陈虎,王守尊,周朝辉.基于数学形态学的图像边缘检测方法研究.工程图学学报.2004,2
铩颷39]Koschan A.A comparative study on color edge detection.Proceedings and AsianConference on Computer Vision ACCV'95.1995.574-578
[40]吴观峰.基于仿生算法的柔性形态滤波器优化研究.工学硕士学位论文.哈尔滨工程大学.2004,12
[41]马慧民,叶春明,张爽.二进制改进粒子群算法在背包问题中的应用.上海理工大学学报.28(1)
[42]徐小慧,张安.基于粒子群优化算法的最佳炳闽值图像分制.计算机工程与应用.2006,10
[43]范立南,韩晓微,徐心和.基于HSI空间彩色图像多结构元形态边缘检测.工程图学报.2005.2,155-156
[44]黄凤岗,杨国,宋克欧.柔性(soft)形态学在图像边缘检测中的应用.中国图像图形学报.2000,4(5)
[45]Dan Schonfeld.Optimal structuring elements for the morphological patternrestoration of binary images,IEEE Trans on PAMI,1994:16(6):589-601
[46]文华.基于数学形态学的图像处理算法的研究.哈尔滨:哈尔滨工程大学学位论文.2006.12.30
[47]姚进.基于数学形态学的图像边缘检测研究.山东:山东师范大学学位论文.2005.5.17
[48]樊尚春,周游.医学图像处理技术概述.计算机科学与技术.2007.4.12
[49]袁华,藤奇志等.医学图像处理中的边缘检测方法进展(综述),暨南大学学报.2002.2(21.1)
[50]Haralicrm,Sternbergsr.Zhuang Xin-hua.Image analysis using mathematicalmorphology.IEEE ransonPAMI,1987,9(4):532
[51]Mallats,Wangwl.Ingularity detection and processing with wavelets.IEEE Transon Information Theory,1992,38(2):679
[52]胡守仁.神经网络应用技术.北京:国防科技出版社,1993
[53]ChenWun-fan,LuXia.A new algorithm of edge detection for colorimage:generalized fuzzy operator.Science in China,1995,38(10)
[54]赵春晖,张乾等.基于数学形态滤波算子的医学图像边缘检测.信息技术.2002(11):49-50
[55]赵于前.陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[56]颜七笙,丁木华等.软形念学算子在医学图像边缘检测中的应用研究.计算机
铩颷54]赵春晖,张乾等.基于数学形态滤波算子的医学图像边缘检测.信息技术.2002(11):49-50
[55]赵于前.陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[56]颜七笙,丁木华等.软形态学算子在医学图像边缘检测中的应用研究.计算机与现代化.2005.11
[57]陆斌,何友金.制作天空背景动态红外图像.微计算机信息.2003.8.19(8)
[58]陈仁,许开宇.红外图像的人像边缘检测技术研究.激光与红外,2005.9
[59]刘曙,罗予频,杨士元.基于多尺度形态学的红外图像边缘检测方法.计算机应用.2007.4
[60]R.Gonzalez,R.Wood,Digital Image Processing.Addison-Wesley,1992
[61]友金,李楠.船红外图像边缘检测方法对比研究.计算机仿真.2006.4
[62]惠阿丽,林辉.基于小波的红外图像边缘检测.红外技术.2007.1.29(1)
[63]叶斌,彭嘉雄.基于顺序形态滤波的运动点目标检测.红外与激光工程.2001,30,25-29
[64]董言治,周晓东,娄树理.柔性数学形态学在红外图像边缘检测中的MATLAB实现.电光与控制,2004,11.11(4)
[65]Franck J.J.Manual of Remote Sensing.American Society of Photo grammetry,1975,(12):237-240
[66]Harper D.D.巡天千里眼-遥感入门(周诗健译).北京:科学出版社,1980:2-5
[67]Rioul,Olivier,Vetterli M.Wavelets and Signal Processing.IEEE SignalProcessingMagazine,1991:14-38
[68]KUOSMANENP,ASTOLA J.Soft morphological filtering.Journal ofMathematical Imaging and Vision archive,1995,5(3):231-262
[69]刘明嫒,赵春晖.适于遥感噪声图像恢复的柔性形态滤波器优化设计.应用科技.2004.7.31(7)
[2]姚敏等.数字图像处理.北京:机械工业出版社,2006.1
[3]章毓晋.图像处理和分析.北京:清华大学出版社,1999.2
[4]T.Hebert,D.Malagre.Edge detection using a priori model.Proceedings of IntelConference on Image Processing.Austin,Texas,USA,1994,94:303-307
[5]Geman,S.Geman,C.Graffigne,and P.Dong.Boundary detection constraintoptimization.IEEE.Pattern Analysis and Machine Intelligence,1990,12(7):609-628
[6]胡学龙,许开宇,章毓晋.数字图像处理.北京:电子工业出版社,2006
[7]Koskinen,Astola J,Neuvo Y.Soft morphological fillers.SPIE Svmp.In:ImageAlgebra and Morphological Image Processing Ⅱ,San Doego,USA,1991,262-270
[8]A.Gasteratos and I.Andreadis,Soft Mathematical Morphology.Adances inImaging and Electron Physics.2001.12
[9]Pavel A.Maltsev.Soft morphological image resizing.SPIE,97,176-183
[10]Lasse Koskinen,Heikki Huttunen,and Jako Astola.Tampere University ofTechnology.Document Recognition(1994),251
[11]马义德.一种彩色柔性形态模糊模型的设计及其实现.电路与系统学报.2006.6.3(11)
[12]张小琳.图像边缘检测技术综述.高能量密度物理.2007.3(1)
[13]Marr D,Hildreth E.Theory of Edge Detection.Proc.R.Soc.London,Ser.B.1980,207:187-17
[14]Marr D.Vision.San Francisco:Freeman,1982
[15]夏良正.数字图像处理.南京:东南大学出版社,1999
[16]邓鲁华,张延恒等.数字图像处理.北京:机械工业出版社,2004
[17]贾永红.数字图像处理.武汉:武汉大学出版社,2003
[18]崔旭东,邱春蓉,刘瑞根.用标记松弛方法检测闪光图像边缘.光电工程.2001.8.28(4)
[19]Rosenfeld A,Hummel R.A,Zuker SW.Scene labeling by relaxation operations.IEEE Trans.Syst.M an Cybern,1976,SMC26:420-453
[20]Hummel R.A,Zuker.SW.On the foundations of relaxation labeling processes.IEEE Trans.Patt.A nal.M ach.Intell.1983,125:267-287
铩颷21]周猛.图像边缘检测技术在车锁识别打码系统中的应用研究.合肥工业大学.2006,3
[22]冯会真,夏哲雷,林志一.基于神经网络的图像边缘检测方法.中国记量学院学报.2006.12.17(4)
[23]K Etemad,R Chellappa.A Neural Network Based Edge Detector.ProcIJCNN.1993,132-137
[24]Xiaofang Sun,D.C Lowe.Learning Optional Linear Filters for Early Vision.ProcIJCNN,1991,2570-2575
[25]ShiZhenghuo and FungKOB.A Comparison of Digital Speckle Filters.Geoscience and Remote Sensing Symposium.2004.6(29)41-44
[26]陆雨花,章勇.图像去噪技术探讨.科技经济市场.2007.10
[27]杨冬云,李国辉.一种适合于边缘保存的混合噪声抑制方法.哈尔滨:哈尔滨理工大学学报.2005.4.10(2)
[28]胡永刚.基于数学形态学的遥感图像边缘检测算法.河北:燕山大学硕士学位论文.2005.9
[29]刘先锋,沈胜宏,郑明洁.边缘保留的图像噪声滤除方法.计算机应用.2000.11
[30]韩晓微等.一种基于脉冲噪声检测的图像均值滤波方法.计算机工程与应用.2004,27:102-104
[31]刘积平,余英林.一种简易的图像去噪方法.广东:华南理工大学学报.2000,28(2):60-63
[32]唐常青,吕宏伯,黄铮等.数学形态学方法及其应用.北京:科学出版社,1990
[33]华续钊,彩色图像边缘检测相关算法研究,扬州大学硕士学位论文,2004.4
[34]赵于前,陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[35]L.Koskinen,J.Astola,Y Neuvo.Soft Morphological Filters.Proc.SPIE.Symp.Image Algebra and Morphological Image Processing Ⅱ,1991.1.262-270
[36]于子翊,李晓娟,柏玲.柔性形态边缘检测算法在菌类显微图像中的应用.北京:首都师范大学学报.2006.4
[37]Berhart,R.C,and Kennedy,J.A new optimizer using particles swarm theory,Proc.SixthInternational Symposium on Micro Machine and Human Science,Nagoya,Japan,IEEE ervice Center,Piscataway,1995,39-43
[38]陈虎,王守尊,周朝辉.基于数学形态学的图像边缘检测方法研究.工程图学学报.2004,2
铩颷39]Koschan A.A comparative study on color edge detection.Proceedings and AsianConference on Computer Vision ACCV'95.1995.574-578
[40]吴观峰.基于仿生算法的柔性形态滤波器优化研究.工学硕士学位论文.哈尔滨工程大学.2004,12
[41]马慧民,叶春明,张爽.二进制改进粒子群算法在背包问题中的应用.上海理工大学学报.28(1)
[42]徐小慧,张安.基于粒子群优化算法的最佳炳闽值图像分制.计算机工程与应用.2006,10
[43]范立南,韩晓微,徐心和.基于HSI空间彩色图像多结构元形态边缘检测.工程图学报.2005.2,155-156
[44]黄凤岗,杨国,宋克欧.柔性(soft)形态学在图像边缘检测中的应用.中国图像图形学报.2000,4(5)
[45]Dan Schonfeld.Optimal structuring elements for the morphological patternrestoration of binary images,IEEE Trans on PAMI,1994:16(6):589-601
[46]文华.基于数学形态学的图像处理算法的研究.哈尔滨:哈尔滨工程大学学位论文.2006.12.30
[47]姚进.基于数学形态学的图像边缘检测研究.山东:山东师范大学学位论文.2005.5.17
[48]樊尚春,周游.医学图像处理技术概述.计算机科学与技术.2007.4.12
[49]袁华,藤奇志等.医学图像处理中的边缘检测方法进展(综述),暨南大学学报.2002.2(21.1)
[50]Haralicrm,Sternbergsr.Zhuang Xin-hua.Image analysis using mathematicalmorphology.IEEE ransonPAMI,1987,9(4):532
[51]Mallats,Wangwl.Ingularity detection and processing with wavelets.IEEE Transon Information Theory,1992,38(2):679
[52]胡守仁.神经网络应用技术.北京:国防科技出版社,1993
[53]ChenWun-fan,LuXia.A new algorithm of edge detection for colorimage:generalized fuzzy operator.Science in China,1995,38(10)
[54]赵春晖,张乾等.基于数学形态滤波算子的医学图像边缘检测.信息技术.2002(11):49-50
[55]赵于前.陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[56]颜七笙,丁木华等.软形念学算子在医学图像边缘检测中的应用研究.计算机
铩颷54]赵春晖,张乾等.基于数学形态滤波算子的医学图像边缘检测.信息技术.2002(11):49-50
[55]赵于前.陈真诚,李凌云等.基于柔性数学形态学的医学图像边缘提取.计算机工程与应用.2005.2
[56]颜七笙,丁木华等.软形态学算子在医学图像边缘检测中的应用研究.计算机与现代化.2005.11
[57]陆斌,何友金.制作天空背景动态红外图像.微计算机信息.2003.8.19(8)
[58]陈仁,许开宇.红外图像的人像边缘检测技术研究.激光与红外,2005.9
[59]刘曙,罗予频,杨士元.基于多尺度形态学的红外图像边缘检测方法.计算机应用.2007.4
[60]R.Gonzalez,R.Wood,Digital Image Processing.Addison-Wesley,1992
[61]友金,李楠.船红外图像边缘检测方法对比研究.计算机仿真.2006.4
[62]惠阿丽,林辉.基于小波的红外图像边缘检测.红外技术.2007.1.29(1)
[63]叶斌,彭嘉雄.基于顺序形态滤波的运动点目标检测.红外与激光工程.2001,30,25-29
[64]董言治,周晓东,娄树理.柔性数学形态学在红外图像边缘检测中的MATLAB实现.电光与控制,2004,11.11(4)
[65]Franck J.J.Manual of Remote Sensing.American Society of Photo grammetry,1975,(12):237-240
[66]Harper D.D.巡天千里眼-遥感入门(周诗健译).北京:科学出版社,1980:2-5
[67]Rioul,Olivier,Vetterli M.Wavelets and Signal Processing.IEEE SignalProcessingMagazine,1991:14-38
[68]KUOSMANENP,ASTOLA J.Soft morphological filtering.Journal ofMathematical Imaging and Vision archive,1995,5(3):231-262
[69]刘明嫒,赵春晖.适于遥感噪声图像恢复的柔性形态滤波器优化设计.应用科技.2004.7.31(7)