基于动态规划的鱼眼图像特征匹配方法研究
|
摘要
特征匹配是计算机视觉中的一个基本问题,也是一个非常困难的问题。与传统的透视图像相比,由于鱼眼图像的视场范围较大,导致图像的非线性畸变增大,同时图像中不同区域的光照变化也很复杂,在这种情况下,传统的基于纹理信息的特征匹配方法很难达到满意的匹配效果,相反,基于形状特征的匹配方法对这类问题相对不太敏感。基于此,本文主要就鱼眼图像的轮廓形状特征匹配问题,采用动态规划方法进行了系统研究,并取得了一定的研究成果。本文所完成的主要工作有：
1.给出了一种轮廓特征提取方法。该方法首先在图像中提取MSER仿射不变特征区域,然后计算区域轮廓中每点的弹性值。实验表明,轮廓弹性的极值点能较好的反映轮廓的主要特征,并且该方法提取的极值点稳定可靠。
2.采用典型数据对基于动态规划的轮廓弹性的匹配方法进行了性能测试,并研究了不同参数对最终性能的影响,同时验证了该方法对于图像的平移、旋转和尺度变化具有较好的稳定性。
3.利用轮廓的弹性序列,采用动态规划算法对轮廓匹配提出一种基于局部特征的匹配方案,该算法对于局部形变有较好的适应性。为了克服同级别轮廓受噪声影响的问题,还给出一种基于全局特征的匹配方法,该方法对于噪声具有较好的适应性。最后通过将局部特征和全局特征进行不同权值的融合,得到一种对于噪声和形变都比较稳定的鱼眼图像匹配方法。
实验结果证明,本文给出的方法是一种切实可行的方法,和传统的基于纹理信息的匹配方法具有互补性。
Feature matching is not only a fundamental but also a difficult problem in computer vision. Compared with the traditional perspective images, fish-eye images have more serious non-linear distortion and the illumination changing in different regions, which make feature extraction and matching more difficult. The traditional methods on image matching are most based on textural features which have a thin skin to varying illumination. Methods based on shape features are steady to varying illumination, so these means are hot spots of image matching. The research on feature correspondences for fisheye images is combined with dynamic programming and some fruit has been achived. The main points are as follows:
1. An novel shape descriptor of planar contours is proposed in this paper, called contour flexibility. First, maximally stable extremal regions was distilled, then a sequence of landmarks can be easily obtained from a simple closed plannar by uniform or nonuniform sampling. Then, a contour flexibility sequance can also be computed by the arithmetic which is expounded in this paper. Experiment results show that the contour flexibility has the capability to reflect the feature of the contour, and this method is reliable and stable.
2. Performance test with typical date is implemented, and different parameter has also been considered. Experiment results show that this method is stable to translation, rotation, and the size scale.
3. With the flexicibility sequences and dynamic programming, methods focusing on local features is implemented, which is adaptable to local distortion. However, the methods focusing on local features may fail to match objects of the same class when the contours have significant deformation or noise. We just develop a method in favour of the global shapes of objects with the contour flexicibility which is stable to noise. Our scheme for matching two shapes with the local and global features is combining the local and global method, which is stable to noise and distortion. Experiment results show that this method is feasible and complementary to methods based on textural features.
1.给出了一种轮廓特征提取方法。该方法首先在图像中提取MSER仿射不变特征区域,然后计算区域轮廓中每点的弹性值。实验表明,轮廓弹性的极值点能较好的反映轮廓的主要特征,并且该方法提取的极值点稳定可靠。
2.采用典型数据对基于动态规划的轮廓弹性的匹配方法进行了性能测试,并研究了不同参数对最终性能的影响,同时验证了该方法对于图像的平移、旋转和尺度变化具有较好的稳定性。
3.利用轮廓的弹性序列,采用动态规划算法对轮廓匹配提出一种基于局部特征的匹配方案,该算法对于局部形变有较好的适应性。为了克服同级别轮廓受噪声影响的问题,还给出一种基于全局特征的匹配方法,该方法对于噪声具有较好的适应性。最后通过将局部特征和全局特征进行不同权值的融合,得到一种对于噪声和形变都比较稳定的鱼眼图像匹配方法。
实验结果证明,本文给出的方法是一种切实可行的方法,和传统的基于纹理信息的匹配方法具有互补性。
Feature matching is not only a fundamental but also a difficult problem in computer vision. Compared with the traditional perspective images, fish-eye images have more serious non-linear distortion and the illumination changing in different regions, which make feature extraction and matching more difficult. The traditional methods on image matching are most based on textural features which have a thin skin to varying illumination. Methods based on shape features are steady to varying illumination, so these means are hot spots of image matching. The research on feature correspondences for fisheye images is combined with dynamic programming and some fruit has been achived. The main points are as follows:
1. An novel shape descriptor of planar contours is proposed in this paper, called contour flexibility. First, maximally stable extremal regions was distilled, then a sequence of landmarks can be easily obtained from a simple closed plannar by uniform or nonuniform sampling. Then, a contour flexibility sequance can also be computed by the arithmetic which is expounded in this paper. Experiment results show that the contour flexibility has the capability to reflect the feature of the contour, and this method is reliable and stable.
2. Performance test with typical date is implemented, and different parameter has also been considered. Experiment results show that this method is stable to translation, rotation, and the size scale.
3. With the flexicibility sequences and dynamic programming, methods focusing on local features is implemented, which is adaptable to local distortion. However, the methods focusing on local features may fail to match objects of the same class when the contours have significant deformation or noise. We just develop a method in favour of the global shapes of objects with the contour flexicibility which is stable to noise. Our scheme for matching two shapes with the local and global features is combining the local and global method, which is stable to noise and distortion. Experiment results show that this method is feasible and complementary to methods based on textural features.
引文
[1]马颂德,张正友.计算机视觉—计算理论与算法基础.北京,科学出版社,1998.
[2]石正喜,多媒体信息检索技术的研究.f电信快报,2006,4,31-35.
[3]Thomas W Sederberg, Eugene Greenwood. A physically based approach to 2D shape blending[J]. Computer Graphics,1992,26(2),25-34.
[4]张鸿斌,陈豫.连接基于内容图像检索技术中的语义鸿沟.情报理论与实践,2004,2,43-45.
[5]A. Ooshtasby. Description and discrimination of planar shapes using shape matrices[J], IEEE Trans. Pattern Anal. Mach. Intell,1985,7,738-743.
[6]D. Scharstein and R. Szeliski. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. International Journal of Computer Vision,2002,47(1), 7-42.
[7]J.S.Beis, D.G.Lowe. Shape indexing using approximate nearest-neighbour search in high-dimensional spaces. In Proceedings of the Conference on Computer and Pattern Recognition,1997,1000-1006.
[8]G. Xu. and Z. Zhang. Epipolar geometry in stereo, motion and object recognition:A unified approach. Kluwer Academic Publishers,1996.
[9]R. Hartley. Multiple view geometry in computer vision.2th ed, cambridge, Cambridge University Press,2003.
[10]丁成刚,丁尚文.动态规划中的最优化原理及方法的进一步研究.工科数学,1997.4,43-45.
[11]McConnell Ross.Kwok Ronald.Curlander John C.Kober W,Pang Shirley S φ-s correlation and dynamic time warping:Two methods for tracking ice floes in sar image,1991.
[12]Lu Hanqing.Ma Songde Representation of feature point and error analysis,1993.
[13]Gorman J W.Mitchell O R.Kuhl F P Partial shape recognition using dynamic programming,1988.
[14]Wu W-Y.Wang M-J J Two dimensional object recognition through two stage string matching,1999.
[15]顾水清.试论城市的动态规划.城市规划汇刊,1994,1,40-43.
[16]常淑芬.动态规划及其在管理中的应用.华南金融研究,1997,2,71-75.
[17]赵奕,赵增墀,张隆昌,袁驰.动态规划在企业生产管理中的应用.纺织器材,1997.5,46-50.
[18]左志坚.动态规划在管理会计中的应用.西北大学学报(哲学社会科学版),1994.2,27-33.
[19]王孝尔.用动态规划来解背包问题.中小学电脑报NOI专刊,2007.1,33-34.
[20]钱杰军.例淡动态规划在竞赛中的应用.中小学电脑报NOI专刊,2007,5,26-28.
[21]杜彦娟.利用动态规划数学模型求最短路径.煤炭技术,2005,1,97-98.
[22]L. Rabiner and B.H. Juang, Fundamentals of speech recognition. prentice hall,1993.
[23]R. Basri, L. Costa, D. Geiger, and D. Jacobs, Determining the similarity of deformable shapes, Vision Research,1998,38,2365-2385.
[24]Roland T. Chin and Charles R. Dyer. Model-Based recognition in robot vision. ACM Computing Surveys,1986,18(1),67-108.
[25]Naonori Ueda and Satoshi Suzuki. Learning visual models from shape contours using multiscale convex/concave structure matching. IEEE Transactions on Pattern Analysis and Machine Intelligence,1993,15(4),337-352.
[26]Chang Fa-Shyang,Chen Shu—Yuan. Deformed shape retrieval based on markov model[J].Electronics Letters,2000,36(2),126-127.
[27]D.G. Kendall. Shape manifolds, procrustean metrics and complex projective spaces, Bull. London Math. Soc,1984,16,81-121.
[28]F.L. Bookstein. Size and shape spaces for landmark data in two dimensions, Statistical Science,1986,2,181-222.
[29]K.V. Mardia and I.L. Dryden. Shape sistributions for Landmark Data, Advances in Applied Probability,1989,21,742-755.
[30]E. Klassen, A. Srivastava, W. Mio, and S.H. Joshi. Analysis of planar shapes using geodesic paths on shape spaces, IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26,372-383.
[31]D.D. Hoffman and W.A. Richards. Parts of recognition. Cognition,1984,18,65-96.
[32]K. Siddiqi and B.B. Kimia. Parts of visual form:computational aspects. IEEE Conf. Computer Vision and Pattern Recognition,1993,1,75-81.
[33]G. McNeill and S. Vijayakumar. Hierarchical procrustes matching for shape retrieval. IEEE Conf. Computer Vision and Pattern Recognition,2006,1,885-894.
[34]I. Biederman. Recognition-by-components:A theory of human image understanding. Psychological Rev,1987,84,115-147.
[35]F. Attneave. Dimensions of similarity. Am. J. Psychology,1950,63(4),516-556.
[36]T.B. Sebastian, P.N. Klein, and B.B. Kimia. On aligning curves. IEEE Trans. Pattern Analysis and Machine Intelligence,2003,25(1),116-125.
[37]H. Ling and D.W. Jacobs. Using the inner-distance for classification of articulated shapes. IEEE Conf. Computer Vision and Pattern Recognition,2005,2,719-726.
[38]T. Adamek and N.E. O'Connor. A multiscale representation method for nonrigid shapes with a single closed contour. IEEE Trans. Circuits and Systems for Video Technology,2004,14,742-753.
[39]B.W. Hong, E. Prados, S. Soatto, and L. Vese. Shape representation based on integral kernels:Application to image matching and segmentation. IEEE Conf. Computer Vision and Pattern Recognition,2006,1,833-840.
[40]I. Kunttu, L. Lepisto, J. Rauhamaa, and A. Visa. Multiscale fourier descriptor for shape classification.12th Int'l Conf. Image Analysis and Processing,2003,1, 536-541.
[41]H. Liu, L. J. Latecki, W. Liu. A unified curvature definition for regular Polygonal, and digital planar curves. IJCV,2008,80,104-124.
[42]J. Matas, O. Chum, U. Martin and T. Pajdla. Robust wide baseline stereo from maximally stable extremal regions. In Proceedings of the British Machine Vision Conference,2002,1,384-393.
[43]D. G. Lowe. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision,2004,60(2),91-110.
[44]J. Matas, O. Chum, U. Martin and T. Pajdla. Robust wide baseline stereo from maximally stable extremal regions. In Proceedings of the British Machine Vision Conference,2002,1,384-393.
[45]P.F. Felzenszwalb and J. Schwartz. Hierarchical matching of deformable shapes. IEEE Conf. Computer Vision and Pattern Recognition,2007,1,1-8.
[46]H. Ling and D.W. Jacobs. Using the inner-distance for classification of articulated shapes. IEEE Conf. Computer Vision and Pattern Recognition,2005.2,719-726.
[47]F. Mokhtarian. Curvature scale space representation:Theory, Applications, and MPEG-7 Standardization. Springer,2003.
[48]Z. Tu and A.L. Yuille. Shape matching and recognition—using generative models and informative features. Eighth European Conf. Computer Vision,2004,195-209.
[49]S. Belongie, J. Malik, and J. Puzicha. Shape matching and object recognition using shape contexts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24,509-522.
[50]Chunjing Xu,Jianzhuang Liu, and Xiaoou Tang.2D shape matching by contour flexibily. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009, 31(1),180-186.
[2]石正喜,多媒体信息检索技术的研究.f电信快报,2006,4,31-35.
[3]Thomas W Sederberg, Eugene Greenwood. A physically based approach to 2D shape blending[J]. Computer Graphics,1992,26(2),25-34.
[4]张鸿斌,陈豫.连接基于内容图像检索技术中的语义鸿沟.情报理论与实践,2004,2,43-45.
[5]A. Ooshtasby. Description and discrimination of planar shapes using shape matrices[J], IEEE Trans. Pattern Anal. Mach. Intell,1985,7,738-743.
[6]D. Scharstein and R. Szeliski. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. International Journal of Computer Vision,2002,47(1), 7-42.
[7]J.S.Beis, D.G.Lowe. Shape indexing using approximate nearest-neighbour search in high-dimensional spaces. In Proceedings of the Conference on Computer and Pattern Recognition,1997,1000-1006.
[8]G. Xu. and Z. Zhang. Epipolar geometry in stereo, motion and object recognition:A unified approach. Kluwer Academic Publishers,1996.
[9]R. Hartley. Multiple view geometry in computer vision.2th ed, cambridge, Cambridge University Press,2003.
[10]丁成刚,丁尚文.动态规划中的最优化原理及方法的进一步研究.工科数学,1997.4,43-45.
[11]McConnell Ross.Kwok Ronald.Curlander John C.Kober W,Pang Shirley S φ-s correlation and dynamic time warping:Two methods for tracking ice floes in sar image,1991.
[12]Lu Hanqing.Ma Songde Representation of feature point and error analysis,1993.
[13]Gorman J W.Mitchell O R.Kuhl F P Partial shape recognition using dynamic programming,1988.
[14]Wu W-Y.Wang M-J J Two dimensional object recognition through two stage string matching,1999.
[15]顾水清.试论城市的动态规划.城市规划汇刊,1994,1,40-43.
[16]常淑芬.动态规划及其在管理中的应用.华南金融研究,1997,2,71-75.
[17]赵奕,赵增墀,张隆昌,袁驰.动态规划在企业生产管理中的应用.纺织器材,1997.5,46-50.
[18]左志坚.动态规划在管理会计中的应用.西北大学学报(哲学社会科学版),1994.2,27-33.
[19]王孝尔.用动态规划来解背包问题.中小学电脑报NOI专刊,2007.1,33-34.
[20]钱杰军.例淡动态规划在竞赛中的应用.中小学电脑报NOI专刊,2007,5,26-28.
[21]杜彦娟.利用动态规划数学模型求最短路径.煤炭技术,2005,1,97-98.
[22]L. Rabiner and B.H. Juang, Fundamentals of speech recognition. prentice hall,1993.
[23]R. Basri, L. Costa, D. Geiger, and D. Jacobs, Determining the similarity of deformable shapes, Vision Research,1998,38,2365-2385.
[24]Roland T. Chin and Charles R. Dyer. Model-Based recognition in robot vision. ACM Computing Surveys,1986,18(1),67-108.
[25]Naonori Ueda and Satoshi Suzuki. Learning visual models from shape contours using multiscale convex/concave structure matching. IEEE Transactions on Pattern Analysis and Machine Intelligence,1993,15(4),337-352.
[26]Chang Fa-Shyang,Chen Shu—Yuan. Deformed shape retrieval based on markov model[J].Electronics Letters,2000,36(2),126-127.
[27]D.G. Kendall. Shape manifolds, procrustean metrics and complex projective spaces, Bull. London Math. Soc,1984,16,81-121.
[28]F.L. Bookstein. Size and shape spaces for landmark data in two dimensions, Statistical Science,1986,2,181-222.
[29]K.V. Mardia and I.L. Dryden. Shape sistributions for Landmark Data, Advances in Applied Probability,1989,21,742-755.
[30]E. Klassen, A. Srivastava, W. Mio, and S.H. Joshi. Analysis of planar shapes using geodesic paths on shape spaces, IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26,372-383.
[31]D.D. Hoffman and W.A. Richards. Parts of recognition. Cognition,1984,18,65-96.
[32]K. Siddiqi and B.B. Kimia. Parts of visual form:computational aspects. IEEE Conf. Computer Vision and Pattern Recognition,1993,1,75-81.
[33]G. McNeill and S. Vijayakumar. Hierarchical procrustes matching for shape retrieval. IEEE Conf. Computer Vision and Pattern Recognition,2006,1,885-894.
[34]I. Biederman. Recognition-by-components:A theory of human image understanding. Psychological Rev,1987,84,115-147.
[35]F. Attneave. Dimensions of similarity. Am. J. Psychology,1950,63(4),516-556.
[36]T.B. Sebastian, P.N. Klein, and B.B. Kimia. On aligning curves. IEEE Trans. Pattern Analysis and Machine Intelligence,2003,25(1),116-125.
[37]H. Ling and D.W. Jacobs. Using the inner-distance for classification of articulated shapes. IEEE Conf. Computer Vision and Pattern Recognition,2005,2,719-726.
[38]T. Adamek and N.E. O'Connor. A multiscale representation method for nonrigid shapes with a single closed contour. IEEE Trans. Circuits and Systems for Video Technology,2004,14,742-753.
[39]B.W. Hong, E. Prados, S. Soatto, and L. Vese. Shape representation based on integral kernels:Application to image matching and segmentation. IEEE Conf. Computer Vision and Pattern Recognition,2006,1,833-840.
[40]I. Kunttu, L. Lepisto, J. Rauhamaa, and A. Visa. Multiscale fourier descriptor for shape classification.12th Int'l Conf. Image Analysis and Processing,2003,1, 536-541.
[41]H. Liu, L. J. Latecki, W. Liu. A unified curvature definition for regular Polygonal, and digital planar curves. IJCV,2008,80,104-124.
[42]J. Matas, O. Chum, U. Martin and T. Pajdla. Robust wide baseline stereo from maximally stable extremal regions. In Proceedings of the British Machine Vision Conference,2002,1,384-393.
[43]D. G. Lowe. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision,2004,60(2),91-110.
[44]J. Matas, O. Chum, U. Martin and T. Pajdla. Robust wide baseline stereo from maximally stable extremal regions. In Proceedings of the British Machine Vision Conference,2002,1,384-393.
[45]P.F. Felzenszwalb and J. Schwartz. Hierarchical matching of deformable shapes. IEEE Conf. Computer Vision and Pattern Recognition,2007,1,1-8.
[46]H. Ling and D.W. Jacobs. Using the inner-distance for classification of articulated shapes. IEEE Conf. Computer Vision and Pattern Recognition,2005.2,719-726.
[47]F. Mokhtarian. Curvature scale space representation:Theory, Applications, and MPEG-7 Standardization. Springer,2003.
[48]Z. Tu and A.L. Yuille. Shape matching and recognition—using generative models and informative features. Eighth European Conf. Computer Vision,2004,195-209.
[49]S. Belongie, J. Malik, and J. Puzicha. Shape matching and object recognition using shape contexts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24,509-522.
[50]Chunjing Xu,Jianzhuang Liu, and Xiaoou Tang.2D shape matching by contour flexibily. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009, 31(1),180-186.