彩色图像边缘检测算法研究
|
摘要
边缘是指其周围像素灰度有阶跃变化或屋顶变化的像素的集合。边缘是图像中的重要信息,也是视觉感知的重要线索,它不仅能够传递图像的大部分信息,更是图像分析和机器视觉的重要基础。
图像的边缘检测是图像处理与模式识别的主要内容之一,一幅图像就是一个信息系统,其大量信息是由它的轮廓边缘提供的。因此,边缘提取与检测在图像处理中占有很重要的地位。图像分析和理解的第一步通常是边缘检测,其结果的准确性和可靠性将直接影响到机器视觉系统对客观世界的理解。虽然长期以来对边缘检测算法的研究一直相当活跃,也相继提出了大量的边缘检测算子,然而大多数边缘检测的研究主要集中在灰度图像上,对彩色图像边缘检测的研究直到近年才活跃起来。
本文的主要内容及组织结构如下:
1.引言。开始部分主要介绍了图像边缘检测的背景和意义,简要说明了边缘检测的应用领域和目前国内外图像边缘检测的发展趋势。
2.灰度图像边缘检测。主要介绍了几种经典的灰度图像边缘检测算法,以及现如今在引入了新方法、新概念之后,出现的几类比较先进的灰度图像边缘检测算法。
3.彩色模型。本章在介绍彩色空间及其分类的基础上,又按照应用领域的不同,较为详细的介绍了几种典型的彩色空间。
4.彩色图像边缘检测。本章首先介绍了几种彩色图像边缘检测算法,并对它们的原理以及优缺点方面进行了简要的分析。本文在分析了原有彩色图像边缘检测方法不足的基础上,提出了基于三角形相似度的彩色图像边缘检测方法和基于HSV颜色信息量的彩色图像边缘检测方法。与以往将灰度图像边缘检测方法直接推广到彩色图像的检测方法不同的是,这两种方法分别将颜色的差别转化为三角形的差别和颜色信息量的差别,在一定程度上合理地考虑了各颜色分量之间的相关性,将有关颜色的向量问题转化成标量问题,使问题转入我们更熟悉,发展更完善的领域。实验结果表明,它们能够充分利用图像的颜色信息,更完整的保留原彩色图像的轮廓。
5.结论与展望。对全文进行了总结,并对进一步的研究提出了设想。
Edge is composed by a collection of its nearby pixels which has a step change or changes in roof.It's the important information to image and the significant clue to visual perception.It's not only to transmit the most information of image but also the important foundation for image analysis and machine vision.
Edge detection of the image is one of the most important issues to image processing and mode recognition.An image is an information system and most of its information comes from the edges.As a result,the edge extraction and detection play an important role in image processing.Edge detection is the first step of image analysis and understanding.The correctness and reliability of the results will have a direct effect on the understanding of the machine vision systems to the objective world.Although the research for edge detection has been very active for a long time and lots of practical edge detector has been put forwarded.However,most of the edge detection focus on gray image,the edge detection on color image has not been developed until recent years.
The paper's contents and the framework are as the following:
1.Introduction.At the beginning,we introduce the background and meaning of edge detection,descript its application areas and trends at home and abroad.
2.Edge detection of gray image.Introduce several edge detection algorithms of gray image as well as other developed method since there are so many new methods and new concepts come out.
3.Color space.In this chapter,we introduce the color space and its classification,then present several typical color spaces according to application area.
4.Edge detection of color image.In this chapter,we introduce several typical edge detection algorithms of color image then make analysis of there principle as well as advantages and disadvantages.Based on analysis of the defect of previous algorithms,we put forward two algorithms which were referred to an approach of color image edge detection based on triangle similarity and an approach of color image edge detection based on the quantity of color information in HSV.Is different from the past which extended the method used in gray image directly to color image. They all change the difference of color to triangle' and color information's form which makes the abstract problem become not only quantization but also a problem shift to us that developed more perfect field.The experiment indicated these methods can make the full use of the color image information and preserved the outline of original image more completely.
5.Conclusion and prospect.Make a summary of the full text and put forward envisage for further study.
图像的边缘检测是图像处理与模式识别的主要内容之一,一幅图像就是一个信息系统,其大量信息是由它的轮廓边缘提供的。因此,边缘提取与检测在图像处理中占有很重要的地位。图像分析和理解的第一步通常是边缘检测,其结果的准确性和可靠性将直接影响到机器视觉系统对客观世界的理解。虽然长期以来对边缘检测算法的研究一直相当活跃,也相继提出了大量的边缘检测算子,然而大多数边缘检测的研究主要集中在灰度图像上,对彩色图像边缘检测的研究直到近年才活跃起来。
本文的主要内容及组织结构如下:
1.引言。开始部分主要介绍了图像边缘检测的背景和意义,简要说明了边缘检测的应用领域和目前国内外图像边缘检测的发展趋势。
2.灰度图像边缘检测。主要介绍了几种经典的灰度图像边缘检测算法,以及现如今在引入了新方法、新概念之后,出现的几类比较先进的灰度图像边缘检测算法。
3.彩色模型。本章在介绍彩色空间及其分类的基础上,又按照应用领域的不同,较为详细的介绍了几种典型的彩色空间。
4.彩色图像边缘检测。本章首先介绍了几种彩色图像边缘检测算法,并对它们的原理以及优缺点方面进行了简要的分析。本文在分析了原有彩色图像边缘检测方法不足的基础上,提出了基于三角形相似度的彩色图像边缘检测方法和基于HSV颜色信息量的彩色图像边缘检测方法。与以往将灰度图像边缘检测方法直接推广到彩色图像的检测方法不同的是,这两种方法分别将颜色的差别转化为三角形的差别和颜色信息量的差别,在一定程度上合理地考虑了各颜色分量之间的相关性,将有关颜色的向量问题转化成标量问题,使问题转入我们更熟悉,发展更完善的领域。实验结果表明,它们能够充分利用图像的颜色信息,更完整的保留原彩色图像的轮廓。
5.结论与展望。对全文进行了总结,并对进一步的研究提出了设想。
Edge is composed by a collection of its nearby pixels which has a step change or changes in roof.It's the important information to image and the significant clue to visual perception.It's not only to transmit the most information of image but also the important foundation for image analysis and machine vision.
Edge detection of the image is one of the most important issues to image processing and mode recognition.An image is an information system and most of its information comes from the edges.As a result,the edge extraction and detection play an important role in image processing.Edge detection is the first step of image analysis and understanding.The correctness and reliability of the results will have a direct effect on the understanding of the machine vision systems to the objective world.Although the research for edge detection has been very active for a long time and lots of practical edge detector has been put forwarded.However,most of the edge detection focus on gray image,the edge detection on color image has not been developed until recent years.
The paper's contents and the framework are as the following:
1.Introduction.At the beginning,we introduce the background and meaning of edge detection,descript its application areas and trends at home and abroad.
2.Edge detection of gray image.Introduce several edge detection algorithms of gray image as well as other developed method since there are so many new methods and new concepts come out.
3.Color space.In this chapter,we introduce the color space and its classification,then present several typical color spaces according to application area.
4.Edge detection of color image.In this chapter,we introduce several typical edge detection algorithms of color image then make analysis of there principle as well as advantages and disadvantages.Based on analysis of the defect of previous algorithms,we put forward two algorithms which were referred to an approach of color image edge detection based on triangle similarity and an approach of color image edge detection based on the quantity of color information in HSV.Is different from the past which extended the method used in gray image directly to color image. They all change the difference of color to triangle' and color information's form which makes the abstract problem become not only quantization but also a problem shift to us that developed more perfect field.The experiment indicated these methods can make the full use of the color image information and preserved the outline of original image more completely.
5.Conclusion and prospect.Make a summary of the full text and put forward envisage for further study.
引文
[1]姚敏等.数字图像处理[M].北京:机械工业出版社,2006.
[2]余成波.数字图像处理及14ATLAB实现[M],重庆大学出版社,2003年.
[3]刘明艳.边缘的检测机细化研究[C].2007:7-8
[4]李炜,黄心汉.车辆自动识别系统中车牌分刻的研究[J].信号处理,2000:16:41-44.
[5]T.Kawaguchi,M.Rizon.Iris detection using intensity and edge information.Pattern Recognition[J].2003(36):549-562
[6]巧邵凌,张立明.一种基于肤色和模板的人脸检测方法明[J].红外与毫米波学报,2000,19(3):209-214.
[7]S.Konishi,A.Yuille and J.Coughlan,A statistical approach to crossing based multi-scale edge detection[J],Image and Vision Computing,2003(21):37-48.
[8]S.C.Douglas,T.H.Y.Meng.Design of Edge Detection Templates Using a Neural Network[J],Proc.International Joint Conference on Neural Networks,1990(2):331-334.
[9]D.S.Kim,W.H.Lee and I.S.Kweon.Automatic edge detection using 3x3 ideal binary pixel Patterns and fuzzy-based edge thresholding[J].Patern Recognition Letters,2004(25):101-106.
[10]Z.Lei,B.Paul.Edge detection by scale multiplication in wavelet domain[J].Pattern Recognition Letters,2002,23(14):1771-1784
[11]Q.Tian,X.Li and N.M.Bilgutay.Multiple Target Detection Using Split Spectrum Processing and Group Delay Moving Entropy[C].IEEE Trans on UFFC,1995,42(6):1075-1086.
[12]M.Brejl,M.Sonka.Directional 3D Edge Detection in Anisotropic Data.Detector Design and Performance Assessment[J].Computer Vision and ImageUnderstanding,2000(77):84-110.
[13]J.M.Geusebroek,A.Dev,R.Boomgaard,A.W.M.Smeulders,F.Cornelissen and H.Geerts.Color invariant edge detection.Scale space theories in computer vision[J],Springer Verlag GmbH,Berlin,1999,1682:459-464.
[14]D.Borghys,VLacroix and C.Perneel.Edge and Line Detection in Polarimetric SAR images[J],Proc.Int.Conf.on Pattern Recognition,Quebec,Canada,2002,2:921-924.
[15]魏伟波,芮筱亭.图像边缘检测方法研究[J].计算机工程与应用,2006.30:88-91
[16]胡守仁.神经网络应用技术[M].北京:国防工业出版社,1993.
[17]基于数学形态学的边缘检测和图像分割方法[J].http://www.jicheng.net.cn
[18]S K Pal,R A King.On edge detection of x-ray images using fuzzy sets[J].IEEE Trans on PAMI,1983,5(1):69-77
[19]E H Marmdani.Advances in the linguistic synthesis of fuzzy controller[J].Int'l J Man-machine studies,1976,8(6):669-679
[20]周德龙,潘泉.图像模糊边缘检测的改进算法[J].中国图像图形学报,2001,6(4):353-358
[21]王倩,阮海波.快速模糊边缘检测算法[J].中国图像图形学报,2001,6(1):92-95
[22]L O Chua,L Yang.Cellular neural networks.Theory[J].IEEE Trans Circuits Syst, 1988,35(10):1257-1272
[23]Koschan A.A Comparative Study On Color Edge Detection[A].Proceedings 2nd Asian Conference on Computer Vision ACCV'95[C].Singapore,5-8 December 1995,Vol.Ⅲ:574-578.
[24]朱宏擎,经致远等.医学彩色图像边缘检测方法研究[J].中国医疗器械杂志.1999.23(2).
[25]TrahaniasPE,VenetsanopoulosAN.Vector order statistics Operators as Color Edge Detectors[J].IEEE Trans on Sys,Man and Cyb,1996,26(1):135-143.
[26]ScharcanskiJ,Venetsanopoulous AN.Edge Detection of Color Image Using Directional Operators[J].IEEE Trans Circuits and Systems for Vediou Technology,1997,7(2):397-401.
[27]DonyRD,WesolkowskiS.Edge Detection on color Images Using RGB Vector Angle [A].Proceedings of the 1999 IEEE Canadian Conference on Electrical and Computer Engineering Shaw Conference Center[C].Edmonton,Alberta,Canada,9-12 May 1999:687-692.
[28]WesolkowskiSB.Color Image Edge Detection and Segmentation:A Comparison of the vector Angle and the Euclidean Distance Color Similarity Measures[J].A thesis presented to the University of Waterloo in fulfillment of the Applied Science in Systems Design Engineering.Waterloo,Ontario,Canada,1999:52-54.
[29]赵景秀,王刚,张静远,赵家刚,杨秀国.基于彩色信息区分度的彩色边缘检测[J].计算机应用,2001,21(8):29-31
[30]王静,李利军.彩色图像边缘检测方法的对比与研究[J].计算机技术与自动化,2003.22(4):55-58
[31]郎方年,周激流,闫斌,宋恩彬,钟凡.四元数与彩色图像边缘检测[J].计算机科学技术2007Vol.34No.11:212-216
[32]樊娜,李晋惠.图像边缘检测的Prewitt算子的改进算法[J].西安工业学院学报,2005,25(1):37-44
[33]姚敏等.数字图像处理[M].北京:机械工业出版社,2006
[34]刘明艳,赵景秀,孙宁.用Prewitt算子细化边缘[J].光电子技术,2006,26(4):259-263
[35]http://wz.csdn.net/url/1313942/
[2]余成波.数字图像处理及14ATLAB实现[M],重庆大学出版社,2003年.
[3]刘明艳.边缘的检测机细化研究[C].2007:7-8
[4]李炜,黄心汉.车辆自动识别系统中车牌分刻的研究[J].信号处理,2000:16:41-44.
[5]T.Kawaguchi,M.Rizon.Iris detection using intensity and edge information.Pattern Recognition[J].2003(36):549-562
[6]巧邵凌,张立明.一种基于肤色和模板的人脸检测方法明[J].红外与毫米波学报,2000,19(3):209-214.
[7]S.Konishi,A.Yuille and J.Coughlan,A statistical approach to crossing based multi-scale edge detection[J],Image and Vision Computing,2003(21):37-48.
[8]S.C.Douglas,T.H.Y.Meng.Design of Edge Detection Templates Using a Neural Network[J],Proc.International Joint Conference on Neural Networks,1990(2):331-334.
[9]D.S.Kim,W.H.Lee and I.S.Kweon.Automatic edge detection using 3x3 ideal binary pixel Patterns and fuzzy-based edge thresholding[J].Patern Recognition Letters,2004(25):101-106.
[10]Z.Lei,B.Paul.Edge detection by scale multiplication in wavelet domain[J].Pattern Recognition Letters,2002,23(14):1771-1784
[11]Q.Tian,X.Li and N.M.Bilgutay.Multiple Target Detection Using Split Spectrum Processing and Group Delay Moving Entropy[C].IEEE Trans on UFFC,1995,42(6):1075-1086.
[12]M.Brejl,M.Sonka.Directional 3D Edge Detection in Anisotropic Data.Detector Design and Performance Assessment[J].Computer Vision and ImageUnderstanding,2000(77):84-110.
[13]J.M.Geusebroek,A.Dev,R.Boomgaard,A.W.M.Smeulders,F.Cornelissen and H.Geerts.Color invariant edge detection.Scale space theories in computer vision[J],Springer Verlag GmbH,Berlin,1999,1682:459-464.
[14]D.Borghys,VLacroix and C.Perneel.Edge and Line Detection in Polarimetric SAR images[J],Proc.Int.Conf.on Pattern Recognition,Quebec,Canada,2002,2:921-924.
[15]魏伟波,芮筱亭.图像边缘检测方法研究[J].计算机工程与应用,2006.30:88-91
[16]胡守仁.神经网络应用技术[M].北京:国防工业出版社,1993.
[17]基于数学形态学的边缘检测和图像分割方法[J].http://www.jicheng.net.cn
[18]S K Pal,R A King.On edge detection of x-ray images using fuzzy sets[J].IEEE Trans on PAMI,1983,5(1):69-77
[19]E H Marmdani.Advances in the linguistic synthesis of fuzzy controller[J].Int'l J Man-machine studies,1976,8(6):669-679
[20]周德龙,潘泉.图像模糊边缘检测的改进算法[J].中国图像图形学报,2001,6(4):353-358
[21]王倩,阮海波.快速模糊边缘检测算法[J].中国图像图形学报,2001,6(1):92-95
[22]L O Chua,L Yang.Cellular neural networks.Theory[J].IEEE Trans Circuits Syst, 1988,35(10):1257-1272
[23]Koschan A.A Comparative Study On Color Edge Detection[A].Proceedings 2nd Asian Conference on Computer Vision ACCV'95[C].Singapore,5-8 December 1995,Vol.Ⅲ:574-578.
[24]朱宏擎,经致远等.医学彩色图像边缘检测方法研究[J].中国医疗器械杂志.1999.23(2).
[25]TrahaniasPE,VenetsanopoulosAN.Vector order statistics Operators as Color Edge Detectors[J].IEEE Trans on Sys,Man and Cyb,1996,26(1):135-143.
[26]ScharcanskiJ,Venetsanopoulous AN.Edge Detection of Color Image Using Directional Operators[J].IEEE Trans Circuits and Systems for Vediou Technology,1997,7(2):397-401.
[27]DonyRD,WesolkowskiS.Edge Detection on color Images Using RGB Vector Angle [A].Proceedings of the 1999 IEEE Canadian Conference on Electrical and Computer Engineering Shaw Conference Center[C].Edmonton,Alberta,Canada,9-12 May 1999:687-692.
[28]WesolkowskiSB.Color Image Edge Detection and Segmentation:A Comparison of the vector Angle and the Euclidean Distance Color Similarity Measures[J].A thesis presented to the University of Waterloo in fulfillment of the Applied Science in Systems Design Engineering.Waterloo,Ontario,Canada,1999:52-54.
[29]赵景秀,王刚,张静远,赵家刚,杨秀国.基于彩色信息区分度的彩色边缘检测[J].计算机应用,2001,21(8):29-31
[30]王静,李利军.彩色图像边缘检测方法的对比与研究[J].计算机技术与自动化,2003.22(4):55-58
[31]郎方年,周激流,闫斌,宋恩彬,钟凡.四元数与彩色图像边缘检测[J].计算机科学技术2007Vol.34No.11:212-216
[32]樊娜,李晋惠.图像边缘检测的Prewitt算子的改进算法[J].西安工业学院学报,2005,25(1):37-44
[33]姚敏等.数字图像处理[M].北京:机械工业出版社,2006
[34]刘明艳,赵景秀,孙宁.用Prewitt算子细化边缘[J].光电子技术,2006,26(4):259-263
[35]http://wz.csdn.net/url/1313942/