图像边缘的感知编组研究
|
摘要
图像的主要特征包括了颜色、纹理和形状。其中“形状”这一特征对于物体识别的重要性可以鉴诸于人类视觉对物体的视觉感知,同时人类的视觉感知能力也为计算机视觉提供了参考模型。在实际应用当中,“形状”主要指的就是物体的边缘。图像边缘检测技术目前趋于成熟,厄待引入新的理论、方法,所以将新兴的感知编组技术与之相结合,无疑是一项很有意义的尝试。
感知编组理论源自于心理学的格式塔理论,其理论研究的主要对象是知觉和解决问题的过程。在计算机视觉领域里,感知编组理论就是利用人类认知事物的规律进行来推演的理论。换言之,通过感知编组可以把视觉系统获取的原始数据组织成为有意义的组合或结构。其中,感知编组的编组规则主要借鉴自格式塔理论,而目前为止这些规则只有很少的一部分被应用到计算机视觉领域当中。其原因一方面在于一些规则的抽象性,增大了实现的难度,另一方面在于感知编组技术仍然处于研究当中,尚待完善。
图像边缘检测技术旨在利用图像周围像素灰度变化来识别图像边缘。然而这样检测到的边缘本身只是像素点的集合,还无法直接应用感知编组原理。因此本文在对图像边缘的感知编组研究当中,于边缘检测之后加入了边缘连接与直线段检测,再进一步应用感知编组的原理对这些直线段进行编组得出物体的边缘轮廓。本文的主要内容如下:
1.对源图像进行边缘检测:鉴于传统边缘检测算法种类繁多,首先对经典的边缘检测算法(Roberts算子法、Sobel算子法、Prewitt算子法、LOG算子法、Canny算子法)进行对比测试,最后选择效果最佳的Canny算子法作为边缘提取的工具,得到图像的边缘集合。
2.边缘连接并组合成直线段:霍夫变换法是一种噪声敏感度低且很有效的边缘连接及直线检测方法。本文在分析霍夫变换原理之后,提出利用梯度阈值筛选有效像素点、用间隔阈值分割直线的改进方法。将第一步骤检测到的边缘集合利用改进后的霍夫变换进行直线段检测,得到感知编组的候选直线段组合。经实验证实,降低了计算复杂度并避免了小直线段的丢失。
3.感知编组:前面两个步骤得到的图像边缘仅仅是经过了像素级的处理,与我们视觉认知到的物体轮廓有相当的差距,并不是图像中物体的真正轮廓,所以需要模拟人的感知能力进行进一步的完善,这一过程主要是通过制定一些规则进行计算,诸如邻近性规则、相似性规则、平行性规则、封闭性规则等,从而得到最后的编组。本文就一些主要的规则提出几组概率模型作为编组规则,经过实验证实了其可行性与有效性。
The main features of image include color, texture and shape.“Shape”is very important in object recognition, which can be proved in human visual perception. In the mean time, the human visual perception ability can also provide reference models for the computer vision. By the way, the“shape”exactly means“edge”in practice. Nowadays, the Image Edge Detection technology has been developed, there are urgent requirements for introduction of new theories and methods. Thus, the combination of the Perceptual Organization technology and the Image Edge Detection technology will be a significant attempt undoubtedly.
Perceptual Organization Theory is derived from the Gestalt Theory in psychology, and its mainly research objects are processes of perception and problem-solving. In the computer vision field, this theory utilize the human’s perception laws, in other words, we can organize significant combinations or structures by the original data from vision system. Laws of the Perceptual Organization Theory mainly come from the Gestalt Theory. However, few of the laws can be applied in the computer vision field at present, reasons are complicated: first, it is too abstract to carry out for some laws; then second, Perceptual Organization Theory is a new developing technology.
The Image Edge Detection method takes advantage of the gray scale alternation to identify edges. However, in spite of the noise sensitivity and the detection precision, these edges we got are only sets of the pixels, and can not be applied to the Perceptual Organization principle directly. Thus, in the research on perceptual grouping of image edge, we add a process of edge connection and straight line segments detection after the Edge Detection, and then organize and get the contour by the ways of the Perceptual Organization Principle. The main content of the paper is as follows.
1. Detect edges of the original image: For the amount of the classical edge detection algorithms, we have a contrast test on the classical algorithms at first, including the Roberts, Sobel, Prewitt, LOG, and the Canny algorithm. Finally we choose the most effective method-Canny algorithm to extract edges, and get the edge-sets.
2. Connect the edge-sets and change them into straight line segments: The Hough Transform is a low noise sensitivity and effective method for edge-connection and straight line detection. By analyzing the principle of the Hough transform, we propose an improved method, which makes a gradient threshold to select valid pixels and another gap threshold to segment straight lines. Thus, we can take advantage of the improved Hough Transform to receive candidate of straight line-sets. Experiment confirmed that the improved method reduces computational complexity and avoid the loss of small straight line segments.
3. Perceptual Grouping: Through the two steps above, we only get the pixel-level edges, but it is far away from the object contour obtained by human vision system, not exactly the real contour. So we need to simulate human vision system to constitute some laws, such as Proximity, Similarity, Parallelism, Closure laws, and etc, then we find the final groups. In this paper, we choose some of the major rules and propose several probability models as the grouping laws. Finally, experiments confirmed its feasibility and effectiveness.
感知编组理论源自于心理学的格式塔理论,其理论研究的主要对象是知觉和解决问题的过程。在计算机视觉领域里,感知编组理论就是利用人类认知事物的规律进行来推演的理论。换言之,通过感知编组可以把视觉系统获取的原始数据组织成为有意义的组合或结构。其中,感知编组的编组规则主要借鉴自格式塔理论,而目前为止这些规则只有很少的一部分被应用到计算机视觉领域当中。其原因一方面在于一些规则的抽象性,增大了实现的难度,另一方面在于感知编组技术仍然处于研究当中,尚待完善。
图像边缘检测技术旨在利用图像周围像素灰度变化来识别图像边缘。然而这样检测到的边缘本身只是像素点的集合,还无法直接应用感知编组原理。因此本文在对图像边缘的感知编组研究当中,于边缘检测之后加入了边缘连接与直线段检测,再进一步应用感知编组的原理对这些直线段进行编组得出物体的边缘轮廓。本文的主要内容如下:
1.对源图像进行边缘检测:鉴于传统边缘检测算法种类繁多,首先对经典的边缘检测算法(Roberts算子法、Sobel算子法、Prewitt算子法、LOG算子法、Canny算子法)进行对比测试,最后选择效果最佳的Canny算子法作为边缘提取的工具,得到图像的边缘集合。
2.边缘连接并组合成直线段:霍夫变换法是一种噪声敏感度低且很有效的边缘连接及直线检测方法。本文在分析霍夫变换原理之后,提出利用梯度阈值筛选有效像素点、用间隔阈值分割直线的改进方法。将第一步骤检测到的边缘集合利用改进后的霍夫变换进行直线段检测,得到感知编组的候选直线段组合。经实验证实,降低了计算复杂度并避免了小直线段的丢失。
3.感知编组:前面两个步骤得到的图像边缘仅仅是经过了像素级的处理,与我们视觉认知到的物体轮廓有相当的差距,并不是图像中物体的真正轮廓,所以需要模拟人的感知能力进行进一步的完善,这一过程主要是通过制定一些规则进行计算,诸如邻近性规则、相似性规则、平行性规则、封闭性规则等,从而得到最后的编组。本文就一些主要的规则提出几组概率模型作为编组规则,经过实验证实了其可行性与有效性。
The main features of image include color, texture and shape.“Shape”is very important in object recognition, which can be proved in human visual perception. In the mean time, the human visual perception ability can also provide reference models for the computer vision. By the way, the“shape”exactly means“edge”in practice. Nowadays, the Image Edge Detection technology has been developed, there are urgent requirements for introduction of new theories and methods. Thus, the combination of the Perceptual Organization technology and the Image Edge Detection technology will be a significant attempt undoubtedly.
Perceptual Organization Theory is derived from the Gestalt Theory in psychology, and its mainly research objects are processes of perception and problem-solving. In the computer vision field, this theory utilize the human’s perception laws, in other words, we can organize significant combinations or structures by the original data from vision system. Laws of the Perceptual Organization Theory mainly come from the Gestalt Theory. However, few of the laws can be applied in the computer vision field at present, reasons are complicated: first, it is too abstract to carry out for some laws; then second, Perceptual Organization Theory is a new developing technology.
The Image Edge Detection method takes advantage of the gray scale alternation to identify edges. However, in spite of the noise sensitivity and the detection precision, these edges we got are only sets of the pixels, and can not be applied to the Perceptual Organization principle directly. Thus, in the research on perceptual grouping of image edge, we add a process of edge connection and straight line segments detection after the Edge Detection, and then organize and get the contour by the ways of the Perceptual Organization Principle. The main content of the paper is as follows.
1. Detect edges of the original image: For the amount of the classical edge detection algorithms, we have a contrast test on the classical algorithms at first, including the Roberts, Sobel, Prewitt, LOG, and the Canny algorithm. Finally we choose the most effective method-Canny algorithm to extract edges, and get the edge-sets.
2. Connect the edge-sets and change them into straight line segments: The Hough Transform is a low noise sensitivity and effective method for edge-connection and straight line detection. By analyzing the principle of the Hough transform, we propose an improved method, which makes a gradient threshold to select valid pixels and another gap threshold to segment straight lines. Thus, we can take advantage of the improved Hough Transform to receive candidate of straight line-sets. Experiment confirmed that the improved method reduces computational complexity and avoid the loss of small straight line segments.
3. Perceptual Grouping: Through the two steps above, we only get the pixel-level edges, but it is far away from the object contour obtained by human vision system, not exactly the real contour. So we need to simulate human vision system to constitute some laws, such as Proximity, Similarity, Parallelism, Closure laws, and etc, then we find the final groups. In this paper, we choose some of the major rules and propose several probability models as the grouping laws. Finally, experiments confirmed its feasibility and effectiveness.
引文
[1]雷英杰,郑全第,张刚等.图像工程的基本概念和理论基础[J].空军工程大学学报(自然科学版), 2003, 4(4): 61-64.
[2] Peter Bergstrom. Eye-movement-controlled transform image coders[J]. Signal Processing: Image Communication, 2003, 18: 115-125.
[3]张小琳.图像边缘检测技术综述[J].高能量密度物理, 2007, 1:38-40.
[4]徐建华.图像处理与分析[M].北京:科学出版社,1992.
[5] L.G.Roberts. Machine Perception of Three-Dimensional Solids[D]. Optical and Electro-Optical Information Processing, 1965, 55: 159-197.
[6] J.Prewitt. Object Enhancement and Extraction[D]. New York: Picture Processing and Psychopictorics, 1970.
[7] L.S.Davis. A Survey of Edge Detection Techniques[J]. CGIP, 1975, 4: 448-270.
[8] R.A.Kirsch.Computer Determination of the Constituent Structure of Biological Images[J]. Computers in Biomedical Research, 1971, 4: 315-328.
[9] R.C.Gonzales, R.E.Woods.Digital Image Processing. Beijing:Publishing House of Electronics Industry, 2003.
[10]李小红.基于LOG滤波器的图像边缘检测算法的研究[J].计算机应用与软件, 2005, 22(5): 107-108.
[11] D.Marr, E.Hildreth. Theory of edge detection[J]. Computer Vision, 1991: 77-107.
[12] Canny J. A Computational approach to edge detection[J]. IEEE trans on PAMI, 1986, 8(6): 679-698.
[13]程正兴.小波分析与应用实例[M].西安:西安交通大学出版社.2006.
[14]马启新,王文涛,杜鹏飞.图像边缘检测技术[J].多媒体技术及其应用,2007:129-133.
[15]姚敏.数字图像处理[M].北京:机械工业出版社, 2007.
[16] Stephane Mallat著,杨力华,等译.信号处理的小波导论[M].北京:机械工业出版社, 2002.
[17]柳薇,马争鸣.基于边缘检测的图像小波阈值去噪方法[J].中国图像图形学报, 2002, 8(7): 788-793.
[18]张书玲,张小华.基于小波变换的边缘检测[J].西北大学学报(自然科学版), 2000, 4(30): 93-97.
[19] Pati YC,Krishnaprasad PS. Analysis And Syntaesis Of Feenforward Neural Network Using discrete Affine Wavelet.IEEE Trans. On NN, 1993, 4(1) :73-75.
[20]王新春.基于小波神经网络的人脸识别[J].微机发展, 2003, 13(6): 27-28.
[21] Zhang Qinghua, Bmvenlste. A Wavelet Network[J].IEEE Trans.On Neural Networks, 1992, 3(6): 889-898.
[22]杨华千,张伟,韦鹏程等.基于小波多尺度变换和模糊聚类的图像边缘检测研究[J].计算机科学, 2006, 33(1) :274-276.
[23] Vetterli M,Herley C. Wavelets and Filter Banks:theory and Design. IEEE Transactions on Signal Processing. 1992, 40: 2207-2232.
[24]冯俊萍,赵转萍.基于数学形态学的图像边缘检测技术[J].航空计算技术, 2004, 34(3): 53-56.
[25]王树文,闫成新,张天序.数学形态学在图像处理中的应用[J].计算机工程与应用, 2004, 3: 89-92.
[26]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[J].工程图学学报, 2004, 2: 112-116.
[27]崔屹.数字图像处理技术与应用[M].北京:电子工业出版社.1997.
[28] Pal S K, and King R A. Image enhancement using fuzzy sets[J]. Electron. Lett, 1980, 16: 376-378.
[29]施成湘,杨丹,尚晋等.扩展的多尺度模糊边缘检测[J].计算机工程与应用, 2006, 7: 65-69.
[30]张世华,宋振明.一种基于模糊增强的图像边缘提取改进算法[J].湖南工程学院学报, 2005, 15(3): 49-53.
[31]肖锋.基于BP神经网络的数字图像边缘检测算法的研究[J].西安科技大学学报, 2005, 25(3): 372-377.
[32] Marr.D. Vision: a computational investigation into the human representation of processing of visual information[M]. San Francisco: Freeman, 1981.
[33] Witkin A,Tenenbuam J. On the role of structure in vision, in Human and machine Vision[M]. Beck J, Hope B, Rosenfeld New York: Academic, 1983: 453-481.
[34]董鸿燕,沈振康,罗军等.感知编组综述[J].计算机工程与应用, 2007, 43(14): 9-14.
[35] Medioni G, Lee MS, Tang CK. A computational framework for feature extraction and segmentation[M]. New York: Elsevier, 2000.
[36] Tan M, Gao Q. Feature grid neural networks for curve partitioning[C]. Proceedings of the 2000 IEEE Signal Processing Society Workshop on Neural Networks for Signal Processing X, Halifax, 2000, 2: 642-651.
[37]朱娟,刘艳滢,王延杰.一种基于Hough变换的新直线段检测算法[J].微电子学与计算机, 2008, 25(12): 60-63.
[38]滕今朝,邱杰.利用Hough变换实现直线的快速精确检测[J].中国图像图形学报, 2008, 13(2): 234-237.
[39]李燕,邵作叶,余旭初.基于感知编组的道路网自动提取研究[J].遥感信息, 2005, 1: 11-15.
[40]王小鹏,王紫婷.基于视觉感知的双层次阈值边缘连接方法[J].计算机应用, 2006, 26(8): 1845-1847.
[41]魏丽,吴中福,李云等.感知归类在目标识别中的应用研究[J].计算机科学, 2006, 33(5): 238-240.
[42]邹琪,罗四维,钟晶晶.全局显著结构主导下的知觉编组算法[J].计算机学报, 2007, 30(11): 2008-2016.
[43]席学强,翟为刚,卢利斌.基于感知组织的直线段编组方法[J].计算机应用研究, 2002, (6):64-67.
[44] S Sarkar, K L Boyer. Perceptual Organization in Computer Vision: A Review and a Proposal for a Classificatory Structure [J]. IEEE Trans. Systems, Man, and Cybernetics, 1993, 23(2): 3822399.
[45] A L Ralescu, J G Shanahan. Perceptual Organization for Inferring Object Boundaries in an Image [J]. Pattern Recognition, 1999, 32(11): 1923-1933.
[46] P Vasseur. Perceptual Organization Approach Based on Dempster Shafer Theory [J]. Pattern Recognition, 1999, 32: 1449-1462.
[2] Peter Bergstrom. Eye-movement-controlled transform image coders[J]. Signal Processing: Image Communication, 2003, 18: 115-125.
[3]张小琳.图像边缘检测技术综述[J].高能量密度物理, 2007, 1:38-40.
[4]徐建华.图像处理与分析[M].北京:科学出版社,1992.
[5] L.G.Roberts. Machine Perception of Three-Dimensional Solids[D]. Optical and Electro-Optical Information Processing, 1965, 55: 159-197.
[6] J.Prewitt. Object Enhancement and Extraction[D]. New York: Picture Processing and Psychopictorics, 1970.
[7] L.S.Davis. A Survey of Edge Detection Techniques[J]. CGIP, 1975, 4: 448-270.
[8] R.A.Kirsch.Computer Determination of the Constituent Structure of Biological Images[J]. Computers in Biomedical Research, 1971, 4: 315-328.
[9] R.C.Gonzales, R.E.Woods.Digital Image Processing. Beijing:Publishing House of Electronics Industry, 2003.
[10]李小红.基于LOG滤波器的图像边缘检测算法的研究[J].计算机应用与软件, 2005, 22(5): 107-108.
[11] D.Marr, E.Hildreth. Theory of edge detection[J]. Computer Vision, 1991: 77-107.
[12] Canny J. A Computational approach to edge detection[J]. IEEE trans on PAMI, 1986, 8(6): 679-698.
[13]程正兴.小波分析与应用实例[M].西安:西安交通大学出版社.2006.
[14]马启新,王文涛,杜鹏飞.图像边缘检测技术[J].多媒体技术及其应用,2007:129-133.
[15]姚敏.数字图像处理[M].北京:机械工业出版社, 2007.
[16] Stephane Mallat著,杨力华,等译.信号处理的小波导论[M].北京:机械工业出版社, 2002.
[17]柳薇,马争鸣.基于边缘检测的图像小波阈值去噪方法[J].中国图像图形学报, 2002, 8(7): 788-793.
[18]张书玲,张小华.基于小波变换的边缘检测[J].西北大学学报(自然科学版), 2000, 4(30): 93-97.
[19] Pati YC,Krishnaprasad PS. Analysis And Syntaesis Of Feenforward Neural Network Using discrete Affine Wavelet.IEEE Trans. On NN, 1993, 4(1) :73-75.
[20]王新春.基于小波神经网络的人脸识别[J].微机发展, 2003, 13(6): 27-28.
[21] Zhang Qinghua, Bmvenlste. A Wavelet Network[J].IEEE Trans.On Neural Networks, 1992, 3(6): 889-898.
[22]杨华千,张伟,韦鹏程等.基于小波多尺度变换和模糊聚类的图像边缘检测研究[J].计算机科学, 2006, 33(1) :274-276.
[23] Vetterli M,Herley C. Wavelets and Filter Banks:theory and Design. IEEE Transactions on Signal Processing. 1992, 40: 2207-2232.
[24]冯俊萍,赵转萍.基于数学形态学的图像边缘检测技术[J].航空计算技术, 2004, 34(3): 53-56.
[25]王树文,闫成新,张天序.数学形态学在图像处理中的应用[J].计算机工程与应用, 2004, 3: 89-92.
[26]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[J].工程图学学报, 2004, 2: 112-116.
[27]崔屹.数字图像处理技术与应用[M].北京:电子工业出版社.1997.
[28] Pal S K, and King R A. Image enhancement using fuzzy sets[J]. Electron. Lett, 1980, 16: 376-378.
[29]施成湘,杨丹,尚晋等.扩展的多尺度模糊边缘检测[J].计算机工程与应用, 2006, 7: 65-69.
[30]张世华,宋振明.一种基于模糊增强的图像边缘提取改进算法[J].湖南工程学院学报, 2005, 15(3): 49-53.
[31]肖锋.基于BP神经网络的数字图像边缘检测算法的研究[J].西安科技大学学报, 2005, 25(3): 372-377.
[32] Marr.D. Vision: a computational investigation into the human representation of processing of visual information[M]. San Francisco: Freeman, 1981.
[33] Witkin A,Tenenbuam J. On the role of structure in vision, in Human and machine Vision[M]. Beck J, Hope B, Rosenfeld New York: Academic, 1983: 453-481.
[34]董鸿燕,沈振康,罗军等.感知编组综述[J].计算机工程与应用, 2007, 43(14): 9-14.
[35] Medioni G, Lee MS, Tang CK. A computational framework for feature extraction and segmentation[M]. New York: Elsevier, 2000.
[36] Tan M, Gao Q. Feature grid neural networks for curve partitioning[C]. Proceedings of the 2000 IEEE Signal Processing Society Workshop on Neural Networks for Signal Processing X, Halifax, 2000, 2: 642-651.
[37]朱娟,刘艳滢,王延杰.一种基于Hough变换的新直线段检测算法[J].微电子学与计算机, 2008, 25(12): 60-63.
[38]滕今朝,邱杰.利用Hough变换实现直线的快速精确检测[J].中国图像图形学报, 2008, 13(2): 234-237.
[39]李燕,邵作叶,余旭初.基于感知编组的道路网自动提取研究[J].遥感信息, 2005, 1: 11-15.
[40]王小鹏,王紫婷.基于视觉感知的双层次阈值边缘连接方法[J].计算机应用, 2006, 26(8): 1845-1847.
[41]魏丽,吴中福,李云等.感知归类在目标识别中的应用研究[J].计算机科学, 2006, 33(5): 238-240.
[42]邹琪,罗四维,钟晶晶.全局显著结构主导下的知觉编组算法[J].计算机学报, 2007, 30(11): 2008-2016.
[43]席学强,翟为刚,卢利斌.基于感知组织的直线段编组方法[J].计算机应用研究, 2002, (6):64-67.
[44] S Sarkar, K L Boyer. Perceptual Organization in Computer Vision: A Review and a Proposal for a Classificatory Structure [J]. IEEE Trans. Systems, Man, and Cybernetics, 1993, 23(2): 3822399.
[45] A L Ralescu, J G Shanahan. Perceptual Organization for Inferring Object Boundaries in an Image [J]. Pattern Recognition, 1999, 32(11): 1923-1933.
[46] P Vasseur. Perceptual Organization Approach Based on Dempster Shafer Theory [J]. Pattern Recognition, 1999, 32: 1449-1462.