视频中运动目标阴影检测研究
|
摘要
随着社会经济和城市建设的快速发展，城市人口的急剧膨胀给社会治安带来了非常大的压力。近年来，随着数字图像处理技术的不断进步，智能视频监控系统越来越普及，它在建设智能城市、平安城市和智慧城市方面发挥着至关重要的作用。在计算机视觉领域，目标检测是智能视频监控系统的最核心部分，是视频场景分析、处理和行为理解等视频内容分析的基础。然而，由于视频监控场景的复杂性，使目标检测方法的研究仍然面临很多挑战。
阴影是自然界中普遍存在的一种光学现象，也是一种常见的图像降质现象，具有与目标相似的两种视觉特性。一种是阴影与背景具有明显的差异性，另一种是阴影与目标具有相同的运动特性，这两种特性使得它很容易被误检测为目标。阴影的存在可能会造成目标粘连、目标形状扭曲，甚至目标丢失，致使目标检测的准确率降低，严重影响后续的视频内容分析。因此，阴影检测成为智能视频监控系统中的一个关键问题，具有重要的理论意义和广泛的应用价值。
基于阴影的光学模型，本文围绕阴影检测问题展开深入研究，并提出了四种不同的运动阴影检测方法。通过在标准视频测试序列中的仿真实验以及与经典方法的对比，验证了本文方法的有效性和优越性。主要工作总结如下：
1.当目标与背景颜色相近时，目标常被误检测为阴影。为了克服单一类型颜色特征的不足，结合阴影与背景的纹理一致性，提出了基于颜色和纹理相结合的运动阴影检测方法。首先，利用HSV颜色空间中的亮度和色度对前景区域进行检测；同时，使用局部二值模式（Local Binary Patterns，LBP）和局部方差（Local Variance）分别描述前景区域的纹理，并通过计算与背景的纹理相似度进行检测。然后，采用逻辑操作融合这两种结果，得到最终的检测结果。该方法同时考虑了阴影的颜色恒常性和纹理一致性，使这两类特征能够实现优势互补，取得了比较好的检测结果。
2.传统的多特征阴影检测方法中，大多是在串行模式下独立使用每个特征并通过与/非逻辑机制判断前景像素是否属于阴影。与这些方法不同，提出了基于多特征融合的运动阴影检测方法，它同时考虑了亮度、颜色和纹理特征在并行模式下的使用。除亮度特征外，首先，从多个颜色空间和多尺度图像中分别提取颜色特征，并利用多通道的信息熵和LBP描述纹理特征；然后，对各个特征进行归一化，并采用不同的决策级融合策略得到最终的特征图；最后，通过阈值法对特征图二值化，实现阴影与目标的分类。通过在不同视频序列中的大量实验表明，线性加权的融合策略达到了较高的准确率，而且优于经典的阴影检测方法。
3.为了克服基于像素的方法对噪声或不确定因素敏感的不足，提出了基于自适应区域分割的运动阴影检测方法。该方法分别利用两种自适应分割算法划分前景区域，一种是吸引力传播（Affinity Propagation，AP）算法；另一种是分水岭算法。前者将前景图像分成不重叠的小块，提取其颜色特征并使用AP实现分割；后者利用分水岭算法对前景图像的梯度图像进行分割，通过像素的梯度变化得到若干个子区域。然后，通过比较子区域的梯度变化、相邻帧以及当前帧与背景之间的子区域纹理相似度进行分类。该方法同时考虑了帧内和帧间的子区域纹理变化，探索的实现了时空特征的结合。大量的实验结果表明，自适应区域分割方法与规则分块方法相比，更能保持区域的特征一致性；与基于像素的方法相比，具有更好的鲁棒性。
4.现有的阴影检测方法中，许多都依赖于一些限制性的假设条件和场景的光照变化，甚至要求每个视频都要有一组固定的参数，然而这在复杂的监控场景中并不适用。针对此问题，提出了基于统计判别模型的运动阴影检测方法。首先，从人工标记的像素中提取不同类型的特征并组成特征向量作为原始样本，特征个数即样本的维数；然后，采用偏最小二乘（Partial Least Squares，PLS）对原始样本进行维数约简，同时使用Logistic判别（Logistic Discrimination，LD）分类，构建统计判别模型PLS-LD；最后，对输入的待检测像素，利用PLS-LD对其分类。该方法不依赖阈值，能够自动判断像素的类别。通过对不同视频序列的测试以及交叉训练测试，验证了该方法的有效性和泛化能力。
With the rapid development of social economy and city construction, the sharpexpansion of urban population is putting great pressure to public security. In recent years,along with the advance of digital image processing techniques, intelligent video surveillancesystem becomes more and more popular, which plays a crucial role in the construction ofintelligent city, safe city and smart city. In the field of computer vision, the core of intelligentvideo surveillance system is object detection, which is the basis of video content analysisincluding scene analysis, processing and behavior understanding. However, the study ofobject detection still confronts many challenges due to the complexity of video scenes.
Shadow is one of the optical phenomena in nature as well as a kind of common imagedegradation phenomenon, which possesses two visual features similar to object. One is thatshadow and background have obvious differences; the other is that shadow and object havethe same motion characteristic. These two properties make it easy to be detected as object.Shadow reduces the accuracy of object detection, which may cause object merging, objectshape distortion and even object loss. It can severely impact the subsequent video contentanalysis. Therefore, shadow detection has become the key problem for video surveillancesystem, which has important theoretical significance and extensive application value.
On the basis of optical shadow model, the study on moving shadow detection is carriedout deeply and four different shadow detection methods are put forward. The effectivenessand superiority of the proposed methods are demonstrated by extensive experiments onseveral standard benchmarks and comparisons with several state-of-the-art methods. Themain work is summarized as follows:
1. When the color information of object is close to background, the object is usuallydetected as shadow mistakenly. To overcome the drawback of single color feature, we presenta moving shadow detection method by combining color with texture in terms of the textureconsistency. Firstly, we adopt the intensity and chromaticity in HSV color space to detectshadows in foreground image. Meanwhile, we utilize Local Binary Patterns (LBP) and localvariance to describe the texture for foreground image, calculate texture similarity betweenforeground and background for detection. Subsequently, logical operation is used to combinethe two results. The proposed method takes the color constancy and texture consistency ofshadow into account simultaneously, which makes the two properties complement with eachother. Experimental results indicate that our method achieves better detection accuracy.
2. Most of conventional shadow detection methods based on multiple features use eachfeature independently in serial mode and then adopt yes/no logical mechanism to determine whether a foreground pixel is shadow or not. Different from these methods, we address amultiple features fusion method by the utilization of intensity, color and texture in parallelmode. Besides the intensity, color features are firstly extracted from multiple color spaces andmulti-scale images while texture features are obtained by entropy and LBP from multiplechannels. Then, normalize each feature map and fuse these maps to generate the final map bydecision level fusion strategies. Finally, we employ appropriate threshold to final feature mapfor classification. Experiments on various benchmarks validate that the proposed method withlinear weighted fusion is superior to other fusion strategies and some representative methods.
3. To address the sensitivity against noise or uncertain factors for pixel-based shadowdetection methods, we suggest an adaptive region segmentation method. Two differentmethods are exploited to segment foreground. One is Affinity Propagation (AP), and the otheris watershed algorithm. The former is to partition the foreground into blocks withoutoverlapping, extract color feature from blocks and then use AP to cluster for pre-segmentation;the latter perform watershed algorithm to gradient image of the foreground and several subareas are obtained according to gradient changes. Then, we compare gradient changes in eachsub area, compute texture similarity between sub areas in consecutive frames and in currentframe and background for classification. Obviously, the method utilizes the texture similarityin both of intra-frame and interframe simultaneously, meanwhile carries out the combinationof spatial and temporal features. Compared with the method of fixed block, experimentalresults demonstrate that the adaptive region segmentation method can retain the consistencyproperty in sub area well. Yet, it also has better robustness than pixels-based methods.
4. The effectiveness of most existing shadow detection methods relies on somerestrictive assumptions and illumination changes in scenes, and even requires each video tohave a fixed parameter set. However, it is not adaptive for complex scenes. Consequently, wepropose a moving shadow detection method based on statistical discriminant model. First,different types of features are extracted from man-made labeled pixels and formed a featurevector as original samples. The number of features is the sample dimension. Second, partialleast square (PLS) is applied to dimensional reduction and logistic discrimination (LD) isadopted for classification. Meanwhile, the statistical discrimination model PLS-LD isestablished. Finally, PLS-LD is used to classify new input pixel. The method does not dependon any threshold and can automatically judge the class of pixels. Extensive experimentalresults and cross training-test on various videos justify the effectiveness and generalizationability of proposed method.
阴影是自然界中普遍存在的一种光学现象，也是一种常见的图像降质现象，具有与目标相似的两种视觉特性。一种是阴影与背景具有明显的差异性，另一种是阴影与目标具有相同的运动特性，这两种特性使得它很容易被误检测为目标。阴影的存在可能会造成目标粘连、目标形状扭曲，甚至目标丢失，致使目标检测的准确率降低，严重影响后续的视频内容分析。因此，阴影检测成为智能视频监控系统中的一个关键问题，具有重要的理论意义和广泛的应用价值。
基于阴影的光学模型，本文围绕阴影检测问题展开深入研究，并提出了四种不同的运动阴影检测方法。通过在标准视频测试序列中的仿真实验以及与经典方法的对比，验证了本文方法的有效性和优越性。主要工作总结如下：
1.当目标与背景颜色相近时，目标常被误检测为阴影。为了克服单一类型颜色特征的不足，结合阴影与背景的纹理一致性，提出了基于颜色和纹理相结合的运动阴影检测方法。首先，利用HSV颜色空间中的亮度和色度对前景区域进行检测；同时，使用局部二值模式（Local Binary Patterns，LBP）和局部方差（Local Variance）分别描述前景区域的纹理，并通过计算与背景的纹理相似度进行检测。然后，采用逻辑操作融合这两种结果，得到最终的检测结果。该方法同时考虑了阴影的颜色恒常性和纹理一致性，使这两类特征能够实现优势互补，取得了比较好的检测结果。
2.传统的多特征阴影检测方法中，大多是在串行模式下独立使用每个特征并通过与/非逻辑机制判断前景像素是否属于阴影。与这些方法不同，提出了基于多特征融合的运动阴影检测方法，它同时考虑了亮度、颜色和纹理特征在并行模式下的使用。除亮度特征外，首先，从多个颜色空间和多尺度图像中分别提取颜色特征，并利用多通道的信息熵和LBP描述纹理特征；然后，对各个特征进行归一化，并采用不同的决策级融合策略得到最终的特征图；最后，通过阈值法对特征图二值化，实现阴影与目标的分类。通过在不同视频序列中的大量实验表明，线性加权的融合策略达到了较高的准确率，而且优于经典的阴影检测方法。
3.为了克服基于像素的方法对噪声或不确定因素敏感的不足，提出了基于自适应区域分割的运动阴影检测方法。该方法分别利用两种自适应分割算法划分前景区域，一种是吸引力传播（Affinity Propagation，AP）算法；另一种是分水岭算法。前者将前景图像分成不重叠的小块，提取其颜色特征并使用AP实现分割；后者利用分水岭算法对前景图像的梯度图像进行分割，通过像素的梯度变化得到若干个子区域。然后，通过比较子区域的梯度变化、相邻帧以及当前帧与背景之间的子区域纹理相似度进行分类。该方法同时考虑了帧内和帧间的子区域纹理变化，探索的实现了时空特征的结合。大量的实验结果表明，自适应区域分割方法与规则分块方法相比，更能保持区域的特征一致性；与基于像素的方法相比，具有更好的鲁棒性。
4.现有的阴影检测方法中，许多都依赖于一些限制性的假设条件和场景的光照变化，甚至要求每个视频都要有一组固定的参数，然而这在复杂的监控场景中并不适用。针对此问题，提出了基于统计判别模型的运动阴影检测方法。首先，从人工标记的像素中提取不同类型的特征并组成特征向量作为原始样本，特征个数即样本的维数；然后，采用偏最小二乘（Partial Least Squares，PLS）对原始样本进行维数约简，同时使用Logistic判别（Logistic Discrimination，LD）分类，构建统计判别模型PLS-LD；最后，对输入的待检测像素，利用PLS-LD对其分类。该方法不依赖阈值，能够自动判断像素的类别。通过对不同视频序列的测试以及交叉训练测试，验证了该方法的有效性和泛化能力。
With the rapid development of social economy and city construction, the sharpexpansion of urban population is putting great pressure to public security. In recent years,along with the advance of digital image processing techniques, intelligent video surveillancesystem becomes more and more popular, which plays a crucial role in the construction ofintelligent city, safe city and smart city. In the field of computer vision, the core of intelligentvideo surveillance system is object detection, which is the basis of video content analysisincluding scene analysis, processing and behavior understanding. However, the study ofobject detection still confronts many challenges due to the complexity of video scenes.
Shadow is one of the optical phenomena in nature as well as a kind of common imagedegradation phenomenon, which possesses two visual features similar to object. One is thatshadow and background have obvious differences; the other is that shadow and object havethe same motion characteristic. These two properties make it easy to be detected as object.Shadow reduces the accuracy of object detection, which may cause object merging, objectshape distortion and even object loss. It can severely impact the subsequent video contentanalysis. Therefore, shadow detection has become the key problem for video surveillancesystem, which has important theoretical significance and extensive application value.
On the basis of optical shadow model, the study on moving shadow detection is carriedout deeply and four different shadow detection methods are put forward. The effectivenessand superiority of the proposed methods are demonstrated by extensive experiments onseveral standard benchmarks and comparisons with several state-of-the-art methods. Themain work is summarized as follows:
1. When the color information of object is close to background, the object is usuallydetected as shadow mistakenly. To overcome the drawback of single color feature, we presenta moving shadow detection method by combining color with texture in terms of the textureconsistency. Firstly, we adopt the intensity and chromaticity in HSV color space to detectshadows in foreground image. Meanwhile, we utilize Local Binary Patterns (LBP) and localvariance to describe the texture for foreground image, calculate texture similarity betweenforeground and background for detection. Subsequently, logical operation is used to combinethe two results. The proposed method takes the color constancy and texture consistency ofshadow into account simultaneously, which makes the two properties complement with eachother. Experimental results indicate that our method achieves better detection accuracy.
2. Most of conventional shadow detection methods based on multiple features use eachfeature independently in serial mode and then adopt yes/no logical mechanism to determine whether a foreground pixel is shadow or not. Different from these methods, we address amultiple features fusion method by the utilization of intensity, color and texture in parallelmode. Besides the intensity, color features are firstly extracted from multiple color spaces andmulti-scale images while texture features are obtained by entropy and LBP from multiplechannels. Then, normalize each feature map and fuse these maps to generate the final map bydecision level fusion strategies. Finally, we employ appropriate threshold to final feature mapfor classification. Experiments on various benchmarks validate that the proposed method withlinear weighted fusion is superior to other fusion strategies and some representative methods.
3. To address the sensitivity against noise or uncertain factors for pixel-based shadowdetection methods, we suggest an adaptive region segmentation method. Two differentmethods are exploited to segment foreground. One is Affinity Propagation (AP), and the otheris watershed algorithm. The former is to partition the foreground into blocks withoutoverlapping, extract color feature from blocks and then use AP to cluster for pre-segmentation;the latter perform watershed algorithm to gradient image of the foreground and several subareas are obtained according to gradient changes. Then, we compare gradient changes in eachsub area, compute texture similarity between sub areas in consecutive frames and in currentframe and background for classification. Obviously, the method utilizes the texture similarityin both of intra-frame and interframe simultaneously, meanwhile carries out the combinationof spatial and temporal features. Compared with the method of fixed block, experimentalresults demonstrate that the adaptive region segmentation method can retain the consistencyproperty in sub area well. Yet, it also has better robustness than pixels-based methods.
4. The effectiveness of most existing shadow detection methods relies on somerestrictive assumptions and illumination changes in scenes, and even requires each video tohave a fixed parameter set. However, it is not adaptive for complex scenes. Consequently, wepropose a moving shadow detection method based on statistical discriminant model. First,different types of features are extracted from man-made labeled pixels and formed a featurevector as original samples. The number of features is the sample dimension. Second, partialleast square (PLS) is applied to dimensional reduction and logistic discrimination (LD) isadopted for classification. Meanwhile, the statistical discrimination model PLS-LD isestablished. Finally, PLS-LD is used to classify new input pixel. The method does not dependon any threshold and can automatically judge the class of pixels. Extensive experimentalresults and cross training-test on various videos justify the effectiveness and generalizationability of proposed method.
引文
[1] Held C, Krumm J, Markel P, et al. Intelligent video surveillance[J]. Computer,2012:83-84．
[2] http://enterprise.huawei.com/cn/．
[3] Sonka M, Hlavac V, Boyle R, et al.分析与机器视觉[M]．北京：人民邮电出版社，2003．
[4] David A F, Jean P.计算机视觉——一种现代方法[M]．北京：电子工业出版社，2004．
[5] Shapiro L G，夏皮罗，Stockman G C，等．计算机视觉[M]．北京：机械工业出版社，2005．
[6]万卫兵，霍宏，赵宇明．智能视频监控中目标检测与识别[M]．上海交通大学出版社，2010．
[7] Anderson C H, Burt P J, Van Der Wal G S. Change detection and tracking using pyramid transformtechniques[C]. In:1985Cambridge Symposium. International Society for Optics and Photonics,1985:72-78.
[8] Tsai D M, Lai S C. Independent component analysis-based background subtraction for indoorsurveillance[J]. IEEE Transactions on Image Processing,2009,18(1):158-167.
[9] McHugh J M, Konrad J, Saligrama V, et al. Foreground-adaptive background subtraction[J]. IEEESignal Processing Letters,2009,16(5):390-393.
[10] Barron J L, Fleet D J, Beauchemin S S. Performance of optical flow techniques[J]. InternationalJournal of Computer Vision,1994,12(1):43-77.
[11] Stauffer C, Grimson W E L. Adaptive background mixture models for real-time tracking[C]. In: IEEEComputer Society Conference on Computer Vision and Pattern Recognition. IEEE,1999,2.
[12] Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2003,25(5):564-577.
[13] Sheikh Y, Shah M. Bayesian modeling of dynamic scenes for object detection[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2005,27(11):1778-1792.
[14] Elgammal A, Harwood D, Davis L. Non-parametric model for background subtraction[C]. ComputerVision-ECCV2000, Springer Berlin Heidelberg,2000:751-767.
[15] Movshovitz-Attias Y, Peleg S. Bacteria-Filters: Persistent particle filters for backgroundsubtraction[C]. In:201017th IEEE International Conference on Image Processing (ICIP). IEEE,2010:677-680.
[16] Mason M, Duric Z. Using histograms to detect and track objects in color video[C]. In: AppliedImagery Pattern Recognition Workshop, AIPR200130th. IEEE,2001:154-159.
[17] Oliver N M, Rosario B, Pentland A P. A Bayesian computer vision system for modeling humaninteractions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):831-843.
[18] Farcas D, Marghes C, Bouwmans T. Background subtraction via incremental maximum margincriterion: a discriminative subspace approach[J]. Machine Vision and Applications,2012,23(6):1083-1101.
[19] Tsai V J D. A comparative study on shadow compensation of color aerial images in invariant colormodels[J]. IEEE Transactions on Geoscience and Remote Sensing,2006,44(6):1661-1671.
[20] Liu J, Li D, Fang T. Shadow detection for color remotely sensed images based on multi-featureintegration[J]. Journal of Applied Remote Sensing,2012,6(1):063521-1-063521-12.
[21] Tian J, Sun J, Tang Y. Tricolor attenuation model for shadow detection[J]. IEEE Transactions onImage Processing,2009,18(10):2355-2363.
[22] Philipson W. The manual of photographic interpretation[C]. American Society for Photogrammetryand Remote Sensing,1997.
[23] Lillesand T M, Kiefer R W, Chipman J W. Remote sensing and image interpretation[M]. John Wiley&Sons Ltd,2004.
[24] Irvin R B, McKeown D M. Methods for exploiting the relationship between buildings and theirshadows in aerial imagery[C]. In: OE/LASE'89,15-20Jan., Los Angeles. CA. International Society forOptics and Photonics,1989:156-164.
[25] Watanabe S, Miyajima K, Mukawa N. Detecting changes of buildings from aerial images usingshadow and shading model[C]. In: Proceedings of Fourteenth International Conference on PatternRecognition. IEEE,1998,2:1408-1412.
[26] Rao K. Curved3-D object description from single aerial images using shadows[P]. U. S. Patent5,943,164.1999-8-24.
[27] Wang C, Huang L, Rosenfeld A. Detecting clouds and cloud shadows on aerial photographs[J].Pattern Recognition Letters,1991,12(1):55-64.
[28] Shafer S A, Kanade T. Using shadows in finding surface orientations[J]. Computer Vision, Graphics,and Image Processing,1983,22(1):145-176.
[29] Jiang C, Ward M O. Shadow identification[C]. In:1992IEEE Computer Society Conference onComputer Vision and Pattern Recognition. IEEE,1992:606-612.
[30] Prati A, Mikic I, Trivedi M M, et al. Detecting moving shadows: algorithms and evaluation[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2003,25(7):918-923.
[31] Al-Najdawi N, Bez H E, Singhai J, et al. A survey of cast shadow detection algorithms[J]. PatternRecognition Letters,2012,33(6):752-764.
[32] Sanin A, Sanderson C, Lovell B C. Shadow detection: A survey and comparative evaluation of recentmethods[J]. Pattern recognition,2012,45(4):1684-1695.
[33] Stauder J, Mech R, Ostermann J. Detection of moving cast shadows for object segmentation[J]. IEEETransactions on Multimedia,1999,1(1):65-76.
[34] Cucchiara R, Grana C, Piccardi M, et al. Improving shadow suppression in moving object detectionwith HSV color information[C]. In: Proceedings of2001IEEE Intelligent Transportation Systems. IEEE,2001:334-339.
[35] Cucchiara R, Grana C, Piccardi M, et al. Detecting moving objects, ghosts, and shadows in videostreams[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10):1337-1342.
[36] Guan Y P. Spatio-temporal motion-based foreground segmentation and shadow suppression[J]. IETComputer Vision,2010,4(1):50-60.
[37] Sanin A, Sanderson C, Lovell B C. Improved shadow removal for robust person tracking insurveillance scenarios[C]. In:201020th International Conference on Pattern Recognition (ICPR). IEEE,2010:141-144.
[38] Leone A, Distante C. Shadow detection for moving objects based on texture analysis[J]. PatternRecognition,2007,40(4):1222-1233.
[39] Qin R, Liao S, Lei Z, et al. Moving cast shadow removal based on local descriptors[C]. In:201020thInternational Conference on Pattern Recognition (ICPR). IEEE,2010:1377-1380.
[40] Yang M T, Lo K H, Chiang C C, et al. Moving cast shadow detection by exploiting multiple cues[J].IET Image Processing,2008,2(2):95-104.
[41]曹健，陈红倩，张凯，等．结合区域颜色和纹理的运动阴影检测方法[J]．机器人，2011，33(5)：628-633．
[42]李宗民，公绪超，刘玉杰．多特征联合建模的视频对象分割技术研究[J]．计算机学报，2013，36(11)．
[43] Sun B, Li S. Moving cast shadow detection of vehicle using combined color models[C]. In:2010Chinese Conference on Pattern Recognition (CCPR). IEEE,2010:1-5.
[44] Tang C, Ahmad M O, Wang C. An efficient method of cast shadow removal using multiple features[J].Signal, Image and Video Processing,2013:1-9.
[45] McFeely R, Glavin M, Jones E. Shadow identification for digital imagery using colour and texturecues[J]. IET Image Processing,2012,6(2):148-159.
[46] Koller D, Daniilidis K, Nagel H H. Model-based object tracking in monocular image sequences ofroad traffic scenes[J]. International Journal of Computer Vision,1993,10(3):257-281.
[47] Xu D, Li X, Liu Z, et al. Cast shadow detection in video segmentation[J]. Pattern Recognition Letters,2005,26(1):91-99.
[48] Hsieh J W, Hu W F, Chang C J, et al. Shadow elimination for effective moving object detection byGaussian shadow modeling[J]. Image and Vision Computing,2003,21(6):505-516.
[49] Yoneyama A, Yeh C H, Kuo C C J. Moving cast shadow elimination for robust vehicle extractionbased on2D joint vehicle/shadow models[C]. In: Proceedings of IEEE Conference on Advanced Video andSignal Based Surveillance,2003. IEEE,2003:229-236.
[50] Fang L Z, Qiong W Y, Sheng Y Z. A method to segment moving vehicle cast shadow based onwavelet transform[J]. Pattern Recognition Letters,2008,29(16):2182-2188.
[51] Mikic I, Cosman P C, Kogut G T, et al. Moving shadow and object detection in traffic scenes[C]. In:Proceedings of15th International Conference on Pattern Recognition. IEEE,2000,1:321-324.
[52] Salvador E, Cavallaro A, Ebrahimi T. Cast shadow segmentation using invariant color features[J].Computer Vision and Image Understanding,2004,95(2):238-259.
[53] Joshi A J, Atev S, Masoud O, et al. Moving shadow detection with low-and mid-level reasoning[C]. In:2007IEEE International Conference on Robotics and Automation. IEEE,2007:4827-4832.
[54] Zhang W, Fang X Z, Yang X K, et al. Moving cast shadows detection using ratio edge[J]. IEEETransactions on Multimedia,2007,9(6):1202-1214.
[55] Martel-Brisson N, Zaccarin A. Learning and removing cast shadows through a multidistributionapproach[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(7):1133-1146.
[56] Jung C R. Efficient background subtraction and shadow removal for monochromatic videosequences[J]. IEEE Transactions on Multimedia,2009,11(3):571-577.
[57]陈功，周荷琴，严捷丰．采用UKF建模的实时背景提取和运动阴影检测[J]．中国图象图形学报，2009，14(5)：931-937．
[58] Liu Z, Huang K, Tan T. Cast shadow removal in a hierarchical manner using MRF[J]. IEEETransactions on Circuits and Systems for Video Technology,2012,22(1):56-66.
[59]查宇飞，楚瀛，王勋，等．一种基于Boosting判别模型的运动阴影检测方法[J]．计算机学报，2007，30(8)：1295-1301．
[60] Meher S K, Murty M N. Efficient method of moving shadow detection and vehicle classification[J].AEU-International Journal of Electronics and Communications,2013,67(8):665-670.
[61] Horprasert T, Harwood D, Davis L S. A statistical approach for real-time robust backgroundsubtraction and shadow detection[C]. In: IEEE ICCV.1999,99:1-19.
[62] Siala K, Chakchouk M, Chaieb F, et al. Moving shadow detection with support vector domaindescription in the color ratios space[C]. In: Proceedings of the17th International Conference on PatternRecognition. IEEE,2004,4:384-387.
[63] Song K T, Tai J C. Image-based traffic monitoring with shadow suppression[J]. In: Proceedings of theIEEE,2007,95(2):413-426.
[64] Choi J M, Yoo Y J, Choi J Y. Adaptive shadow estimator for removing shadow of moving object[J].Computer Vision and Image Understanding,2010,114(9):1017-1029.
[65] Joshi A J, Papanikolopoulos N P. Learning to detect moving shadows in dynamic environments[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(11):2055-2063.
[66] Benedek C, Sziranyi T. Bayesian foreground and shadow detection in uncertain frame ratesurveillance videos[J]. IEEE Transactions on Image Processing,2008,17(4):608-621.
[67]鲍海英，李艳，尹永宜．城市航空影像的阴影检测和阴影消除方法研究[J]．遥感信息，2010，1：010．
[68]董士海．人机交互的进展及面临的挑战[J]．计算机辅助设计与图形学学报，2004，16(1)：1-13．
[69]赵永锋，王毅刚．采用手指的三维人机交互方法[J]．微电子学与计算机，2009，26(6)：21-24．
[70] Szeliski R. Computer vision: algorithms and applications[M]. Springer,2011.
[71] Phong B T. Illumination for computer generated pictures[J]. Communications of the ACM,1975,18(6):311-317.
[72] Skifstad K, Jain R. Illumination independent change detection for real world image sequences[J].Computer Vision, Graphics, and Image Processing,1989,46(3):387-399.
[73] Alexander S T. Adaptive signal processing: theory and applications[M]. Springer-Verlag New York,Inc.,1986.
[74] Nadimi S, Bhanu B. Moving shadow detection using a physics-based approach[C]. In: Proceedings of16th International Conference on Pattern Recognition. IEEE,2002,2:701-704.
[75] Nadimi S, Bhanu B. Physical models for moving shadow and object detection in video[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(8):1079-1087.
[76] Mamassian P, Knill D C, Kersten D. The perception of cast shadows[J]. Trends in cognitive sciences,1998,2(8):288-295.
[77] Nishita T, Okamura I, Nakamae E. Shading models for point and linear sources[J]. ACM Transactionson Graphics (TOG),1985,4(2):124-146.
[78] Baba M, Mukunoki M, Asada N. Shadow removal from a real image based on shadow density[C]. In:ACM SIGGRAPH2004Posters. ACM,2004:60.
[79] Prati A, Mikic I, Grana C, et al. Shadow detection algorithms for traffic flow analysis: a comparativestudy[C]. In: Proceedings of2001IEEE Intelligent Transportation Systems. IEEE,2001:340-345.
[80] Weale R A. Colour vision in the peripheral retina[J]. British Medical Bulletin,1953,9(1):55-60.
[81]胡维生．颜色物理和CIE标准色度系统[J]．物理，1982，11(4)：202-208．
[82] Lee H C. Color Imaging Science[M]. Cambridge: Cambridge University Press,2005.
[83] Wang Y, Loe K F, Wu J K. A dynamic conditional random field model for foreground and shadowsegmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(2):279-289.
[84] Land E H, McCann J J. Lightness and retinex theory[J]. Journal of the Optical Society of America,1971,61(1):1-11.
[85]解兰昌，黄峰，郭铭，等．颜色在视觉的心理学认知结构[J]．光学仪器，1995，17(4)：85-86．
[86] Brainard D H, Wandell B A. Analysis of the retinex theory of color vision[J]. JOSA A,1986,3(10):1651-1661.
[87] Maloney L T, Wandell B A. Color constancy: a method for recovering surface spectral reflectance[J].JOSA A,1986,3(1):29-33.
[88] Meyer G W, Greenberg D P. Perceptual color spaces for computer graphics[C]. In: ACM SIGGRAPHComputer Graphics. ACM,1980,14(3):254-261.
[89] Hinchey M, Pagnoni A, Rammig F J, et al. Color theory and its application in art and design[M].Springer Publishing Company, Incorporated,2008.
[90] Smith A R. Color gamut transform pairs[C]. In: ACM Siggraph Computer Graphics. ACM,1978,12(3):12-19.
[91] Gevers T, Smeulders A W M. Color-based object recognition[J]. Pattern Recognition,1999,32(3):453-464.
[92] Lin C T, Yang C T, Shou Y W, et al. An efficient and robust moving shadow removal algorithm and itsapplications in ITS[J]. EURASIP Journal on Advances in Signal Processing,2010,2010:39.
[93] Xiao M, Han C Z, Zhang L. Moving shadow detection and removal for traffic sequences[J].International Journal of Automation and Computing,2007,4(1):38-46.
[94] Jacques J C S, Jung C R, Musse S R. Background subtraction and shadow detection in grayscale videosequences[C]. In:18th Brazilian Symposium on Computer Graphics and Image Processing,2005.SIBGRAPI2005. IEEE,2005:189-196.
[95] Gonzalez R C, Woods R E, Eddins S L．冈萨雷斯数字图像处理(第二版)[M]．北京：电子工业出版社，2006．
[96]冈萨雷斯，伍兹，埃丁斯，等．数字图像处理：MATLAB版[M]．北京：电子工业出版社，2005．
[97]傅祖芸．信息论：基础理论与应用[M]．北京：电子工业出版社，2001．
[98] Ojala T, Pietik inen M, Harwood D. A comparative study of texture measures with classificationbased on featured distributions[J]. Pattern Recognition,1996,29(1):51-59.
[99] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotation invariant textureclassification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,2002,24(7):971-987.
[100] Heikkila M, Pietikainen M. A texture-based method for modeling the background and detectingmoving objects[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(4):657-662.
[101] Daugman J G. Two-dimensional spectral analysis of cortical receptive field profiles[J]. VisionResearch,1980,20(10):847-856.
[102] Daugman J G. Uncertainty relation for resolution in space, spatial frequency, and orientationoptimized by two-dimensional visual cortical filters[J]. Optical Society of America, Journal, A: Optics andImage Science,1985,2(7):1160-1169.
[103] Xu D, Liu J, Li X, et al. Insignificant shadow detection for video segmentation[J]. IEEE Transactionson Circuits and Systems for Video Technology,2005,15(8):1058-1064.
[104] Guo R, Dai Q, Hoiem D. Single-image shadow detection and removal using paired regions[C]. In:2011IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE,2011:2033-2040.
[105] Yao J, Odobez J M. Multi-layer background subtraction based on color and texture[C]. In: IEEEConference on Computer Vision and Pattern Recognition,2007. IEEE,2007:1-8.
[106]刘宏，李锦涛，刘群，等．融合颜色和梯度特征的运动阴影消除方法[J]．计算机辅助设计与图形学学报，2007，19(10)：1279-1285．
[107] Hamad A M, Tsumura N. Background updating and shadow detection based on spatial, color, andtexture information of detected objects[J]. Optical Review,2012,19(3):182-197.
[108] Boroujeni H S, Charkari N M. Robust moving shadow detection with hierarchical mixture of MLPexperts[J]. Signal, Image and Video Processing,2012:1-15.
[109] Liu J, Li D, Fang T. Shadow detection for color remotely sensed images based on multi-featureintegration[J]. Journal of Applied Remote Sensing,2012,6(1):063521-1-063521-12.
[110] Bullkich E, Ilan I, Moshe Y, et al. Moving shadow detection by nonlinear tone-mapping[C]. In:201219th International Conference on Systems, Signals and Image Processing (IWSSIP). IEEE,2012:146-149.
[111] Herodotou N, Plataniotis K N, Venetsanopoulos A N. A color segmentation scheme for object-basedvideo coding[C]. In:1998IEEE Symposium on Advances in Digital Filtering and Signal Processing. IEEE,1998:25-29.
[112] Hel-Or Y, Hel-Or H, David E. Fast template matching in non-linear tone-mapped images[C]. In:2011IEEE International Conference on Computer Vision (ICCV). IEEE,2011:1355-1362.
[113] Otsu N. A threshold selection method from gray-level histograms[J]. Automatica,1975,11(285-296):23-27.
[114] http://cvrr.ucsd.edu/aton/shadow/.
[115] http://vision.gel.ulaval.ca/~CastShadows/.
[116] Shadow ground-truth data: http://itee.uq.edu.au/~uqasanin/.
[117] CAVIAR dataset: http://homepages.inf.ed.ac.uk/rbf/CAVIAR/.
[118] Soh Y S, Lee H, Wang Y. Invariant color model-based shadow removal in traffic image and a newmetric for evaluating the performance of shadow removal methods[M]. PRICAI2006: Trends in ArtificialIntelligence. Springer Berlin Heidelberg,2006:544-552.
[119] Hahn G J, Shapiro S S. Statistical models in engineering[M]. Statistical models in engineering. JohnWiley&Sons,1968.
[120] Rosin P L, Ellis T J. Image difference threshold strategies and shadow detection[C]: In: Proceedingsof the Sixth British Machine Vision Conference, Birmingham, UK, July1995:347-356.
[121] Mitra B K, Young R, Chatwin C. On shadow elimination after moving region segmentation based ondifferent threshold selection strategies[J]. Optics and Lasers in Engineering,2007,45(11):1088-1093.
[122] Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(11):1254-1259.
[123] Greenspan H, Belongie S, Goodman R, et al. Overcomplete steerable pyramid filters and rotationinvariance[C]. In: Proceedings of1994IEEE Computer Society Conference on Computer Vision andPattern Recognition. IEEE,1994:222-228.
[124] Engel S, Zhang X, Wandell B. Colour tuning in human visual cortex measured with functionalmagnetic resonance imaging[J]. Nature,1997,388(6637):68-71.
[125] Jolliffe I. Principal component analysis[M]. John Wiley&Sons, Ltd,2005.
[126] Fukunaga K. Introduction to statistical pattern recognition[M]. Academic press,1990.
[127] Nguyen D V, Rocke D M. Tumor classification by partial least squares using microarray geneexpression data[J]. Bioinformatics,2002,18(1):39-50.
[128] Vapnik V. Statistical learning theory.1998[J].1998.
[129] Frey B J, Dueck D. Clustering by passing messages between data points[J]. Science,2007,315(5814):972-976.
[130] Arivazhagan S, Ganesan L, Priyal S P. Texture classification using Gabor wavelets based rotationinvariant features[J]. Pattern Recognition Letters,2006,27(16):1976-1982.
[131] Li M, Staunton R C. Optimum Gabor filter design and local binary patterns for texturesegmentation[J]. Pattern Recognition Letters,2008,29(5):664-672.
[132] Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based on immersionsimulations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583-598.
[2] http://enterprise.huawei.com/cn/．
[3] Sonka M, Hlavac V, Boyle R, et al.分析与机器视觉[M]．北京：人民邮电出版社，2003．
[4] David A F, Jean P.计算机视觉——一种现代方法[M]．北京：电子工业出版社，2004．
[5] Shapiro L G，夏皮罗，Stockman G C，等．计算机视觉[M]．北京：机械工业出版社，2005．
[6]万卫兵，霍宏，赵宇明．智能视频监控中目标检测与识别[M]．上海交通大学出版社，2010．
[7] Anderson C H, Burt P J, Van Der Wal G S. Change detection and tracking using pyramid transformtechniques[C]. In:1985Cambridge Symposium. International Society for Optics and Photonics,1985:72-78.
[8] Tsai D M, Lai S C. Independent component analysis-based background subtraction for indoorsurveillance[J]. IEEE Transactions on Image Processing,2009,18(1):158-167.
[9] McHugh J M, Konrad J, Saligrama V, et al. Foreground-adaptive background subtraction[J]. IEEESignal Processing Letters,2009,16(5):390-393.
[10] Barron J L, Fleet D J, Beauchemin S S. Performance of optical flow techniques[J]. InternationalJournal of Computer Vision,1994,12(1):43-77.
[11] Stauffer C, Grimson W E L. Adaptive background mixture models for real-time tracking[C]. In: IEEEComputer Society Conference on Computer Vision and Pattern Recognition. IEEE,1999,2.
[12] Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2003,25(5):564-577.
[13] Sheikh Y, Shah M. Bayesian modeling of dynamic scenes for object detection[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2005,27(11):1778-1792.
[14] Elgammal A, Harwood D, Davis L. Non-parametric model for background subtraction[C]. ComputerVision-ECCV2000, Springer Berlin Heidelberg,2000:751-767.
[15] Movshovitz-Attias Y, Peleg S. Bacteria-Filters: Persistent particle filters for backgroundsubtraction[C]. In:201017th IEEE International Conference on Image Processing (ICIP). IEEE,2010:677-680.
[16] Mason M, Duric Z. Using histograms to detect and track objects in color video[C]. In: AppliedImagery Pattern Recognition Workshop, AIPR200130th. IEEE,2001:154-159.
[17] Oliver N M, Rosario B, Pentland A P. A Bayesian computer vision system for modeling humaninteractions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):831-843.
[18] Farcas D, Marghes C, Bouwmans T. Background subtraction via incremental maximum margincriterion: a discriminative subspace approach[J]. Machine Vision and Applications,2012,23(6):1083-1101.
[19] Tsai V J D. A comparative study on shadow compensation of color aerial images in invariant colormodels[J]. IEEE Transactions on Geoscience and Remote Sensing,2006,44(6):1661-1671.
[20] Liu J, Li D, Fang T. Shadow detection for color remotely sensed images based on multi-featureintegration[J]. Journal of Applied Remote Sensing,2012,6(1):063521-1-063521-12.
[21] Tian J, Sun J, Tang Y. Tricolor attenuation model for shadow detection[J]. IEEE Transactions onImage Processing,2009,18(10):2355-2363.
[22] Philipson W. The manual of photographic interpretation[C]. American Society for Photogrammetryand Remote Sensing,1997.
[23] Lillesand T M, Kiefer R W, Chipman J W. Remote sensing and image interpretation[M]. John Wiley&Sons Ltd,2004.
[24] Irvin R B, McKeown D M. Methods for exploiting the relationship between buildings and theirshadows in aerial imagery[C]. In: OE/LASE'89,15-20Jan., Los Angeles. CA. International Society forOptics and Photonics,1989:156-164.
[25] Watanabe S, Miyajima K, Mukawa N. Detecting changes of buildings from aerial images usingshadow and shading model[C]. In: Proceedings of Fourteenth International Conference on PatternRecognition. IEEE,1998,2:1408-1412.
[26] Rao K. Curved3-D object description from single aerial images using shadows[P]. U. S. Patent5,943,164.1999-8-24.
[27] Wang C, Huang L, Rosenfeld A. Detecting clouds and cloud shadows on aerial photographs[J].Pattern Recognition Letters,1991,12(1):55-64.
[28] Shafer S A, Kanade T. Using shadows in finding surface orientations[J]. Computer Vision, Graphics,and Image Processing,1983,22(1):145-176.
[29] Jiang C, Ward M O. Shadow identification[C]. In:1992IEEE Computer Society Conference onComputer Vision and Pattern Recognition. IEEE,1992:606-612.
[30] Prati A, Mikic I, Trivedi M M, et al. Detecting moving shadows: algorithms and evaluation[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2003,25(7):918-923.
[31] Al-Najdawi N, Bez H E, Singhai J, et al. A survey of cast shadow detection algorithms[J]. PatternRecognition Letters,2012,33(6):752-764.
[32] Sanin A, Sanderson C, Lovell B C. Shadow detection: A survey and comparative evaluation of recentmethods[J]. Pattern recognition,2012,45(4):1684-1695.
[33] Stauder J, Mech R, Ostermann J. Detection of moving cast shadows for object segmentation[J]. IEEETransactions on Multimedia,1999,1(1):65-76.
[34] Cucchiara R, Grana C, Piccardi M, et al. Improving shadow suppression in moving object detectionwith HSV color information[C]. In: Proceedings of2001IEEE Intelligent Transportation Systems. IEEE,2001:334-339.
[35] Cucchiara R, Grana C, Piccardi M, et al. Detecting moving objects, ghosts, and shadows in videostreams[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10):1337-1342.
[36] Guan Y P. Spatio-temporal motion-based foreground segmentation and shadow suppression[J]. IETComputer Vision,2010,4(1):50-60.
[37] Sanin A, Sanderson C, Lovell B C. Improved shadow removal for robust person tracking insurveillance scenarios[C]. In:201020th International Conference on Pattern Recognition (ICPR). IEEE,2010:141-144.
[38] Leone A, Distante C. Shadow detection for moving objects based on texture analysis[J]. PatternRecognition,2007,40(4):1222-1233.
[39] Qin R, Liao S, Lei Z, et al. Moving cast shadow removal based on local descriptors[C]. In:201020thInternational Conference on Pattern Recognition (ICPR). IEEE,2010:1377-1380.
[40] Yang M T, Lo K H, Chiang C C, et al. Moving cast shadow detection by exploiting multiple cues[J].IET Image Processing,2008,2(2):95-104.
[41]曹健，陈红倩，张凯，等．结合区域颜色和纹理的运动阴影检测方法[J]．机器人，2011，33(5)：628-633．
[42]李宗民，公绪超，刘玉杰．多特征联合建模的视频对象分割技术研究[J]．计算机学报，2013，36(11)．
[43] Sun B, Li S. Moving cast shadow detection of vehicle using combined color models[C]. In:2010Chinese Conference on Pattern Recognition (CCPR). IEEE,2010:1-5.
[44] Tang C, Ahmad M O, Wang C. An efficient method of cast shadow removal using multiple features[J].Signal, Image and Video Processing,2013:1-9.
[45] McFeely R, Glavin M, Jones E. Shadow identification for digital imagery using colour and texturecues[J]. IET Image Processing,2012,6(2):148-159.
[46] Koller D, Daniilidis K, Nagel H H. Model-based object tracking in monocular image sequences ofroad traffic scenes[J]. International Journal of Computer Vision,1993,10(3):257-281.
[47] Xu D, Li X, Liu Z, et al. Cast shadow detection in video segmentation[J]. Pattern Recognition Letters,2005,26(1):91-99.
[48] Hsieh J W, Hu W F, Chang C J, et al. Shadow elimination for effective moving object detection byGaussian shadow modeling[J]. Image and Vision Computing,2003,21(6):505-516.
[49] Yoneyama A, Yeh C H, Kuo C C J. Moving cast shadow elimination for robust vehicle extractionbased on2D joint vehicle/shadow models[C]. In: Proceedings of IEEE Conference on Advanced Video andSignal Based Surveillance,2003. IEEE,2003:229-236.
[50] Fang L Z, Qiong W Y, Sheng Y Z. A method to segment moving vehicle cast shadow based onwavelet transform[J]. Pattern Recognition Letters,2008,29(16):2182-2188.
[51] Mikic I, Cosman P C, Kogut G T, et al. Moving shadow and object detection in traffic scenes[C]. In:Proceedings of15th International Conference on Pattern Recognition. IEEE,2000,1:321-324.
[52] Salvador E, Cavallaro A, Ebrahimi T. Cast shadow segmentation using invariant color features[J].Computer Vision and Image Understanding,2004,95(2):238-259.
[53] Joshi A J, Atev S, Masoud O, et al. Moving shadow detection with low-and mid-level reasoning[C]. In:2007IEEE International Conference on Robotics and Automation. IEEE,2007:4827-4832.
[54] Zhang W, Fang X Z, Yang X K, et al. Moving cast shadows detection using ratio edge[J]. IEEETransactions on Multimedia,2007,9(6):1202-1214.
[55] Martel-Brisson N, Zaccarin A. Learning and removing cast shadows through a multidistributionapproach[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(7):1133-1146.
[56] Jung C R. Efficient background subtraction and shadow removal for monochromatic videosequences[J]. IEEE Transactions on Multimedia,2009,11(3):571-577.
[57]陈功，周荷琴，严捷丰．采用UKF建模的实时背景提取和运动阴影检测[J]．中国图象图形学报，2009，14(5)：931-937．
[58] Liu Z, Huang K, Tan T. Cast shadow removal in a hierarchical manner using MRF[J]. IEEETransactions on Circuits and Systems for Video Technology,2012,22(1):56-66.
[59]查宇飞，楚瀛，王勋，等．一种基于Boosting判别模型的运动阴影检测方法[J]．计算机学报，2007，30(8)：1295-1301．
[60] Meher S K, Murty M N. Efficient method of moving shadow detection and vehicle classification[J].AEU-International Journal of Electronics and Communications,2013,67(8):665-670.
[61] Horprasert T, Harwood D, Davis L S. A statistical approach for real-time robust backgroundsubtraction and shadow detection[C]. In: IEEE ICCV.1999,99:1-19.
[62] Siala K, Chakchouk M, Chaieb F, et al. Moving shadow detection with support vector domaindescription in the color ratios space[C]. In: Proceedings of the17th International Conference on PatternRecognition. IEEE,2004,4:384-387.
[63] Song K T, Tai J C. Image-based traffic monitoring with shadow suppression[J]. In: Proceedings of theIEEE,2007,95(2):413-426.
[64] Choi J M, Yoo Y J, Choi J Y. Adaptive shadow estimator for removing shadow of moving object[J].Computer Vision and Image Understanding,2010,114(9):1017-1029.
[65] Joshi A J, Papanikolopoulos N P. Learning to detect moving shadows in dynamic environments[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(11):2055-2063.
[66] Benedek C, Sziranyi T. Bayesian foreground and shadow detection in uncertain frame ratesurveillance videos[J]. IEEE Transactions on Image Processing,2008,17(4):608-621.
[67]鲍海英，李艳，尹永宜．城市航空影像的阴影检测和阴影消除方法研究[J]．遥感信息，2010，1：010．
[68]董士海．人机交互的进展及面临的挑战[J]．计算机辅助设计与图形学学报，2004，16(1)：1-13．
[69]赵永锋，王毅刚．采用手指的三维人机交互方法[J]．微电子学与计算机，2009，26(6)：21-24．
[70] Szeliski R. Computer vision: algorithms and applications[M]. Springer,2011.
[71] Phong B T. Illumination for computer generated pictures[J]. Communications of the ACM,1975,18(6):311-317.
[72] Skifstad K, Jain R. Illumination independent change detection for real world image sequences[J].Computer Vision, Graphics, and Image Processing,1989,46(3):387-399.
[73] Alexander S T. Adaptive signal processing: theory and applications[M]. Springer-Verlag New York,Inc.,1986.
[74] Nadimi S, Bhanu B. Moving shadow detection using a physics-based approach[C]. In: Proceedings of16th International Conference on Pattern Recognition. IEEE,2002,2:701-704.
[75] Nadimi S, Bhanu B. Physical models for moving shadow and object detection in video[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(8):1079-1087.
[76] Mamassian P, Knill D C, Kersten D. The perception of cast shadows[J]. Trends in cognitive sciences,1998,2(8):288-295.
[77] Nishita T, Okamura I, Nakamae E. Shading models for point and linear sources[J]. ACM Transactionson Graphics (TOG),1985,4(2):124-146.
[78] Baba M, Mukunoki M, Asada N. Shadow removal from a real image based on shadow density[C]. In:ACM SIGGRAPH2004Posters. ACM,2004:60.
[79] Prati A, Mikic I, Grana C, et al. Shadow detection algorithms for traffic flow analysis: a comparativestudy[C]. In: Proceedings of2001IEEE Intelligent Transportation Systems. IEEE,2001:340-345.
[80] Weale R A. Colour vision in the peripheral retina[J]. British Medical Bulletin,1953,9(1):55-60.
[81]胡维生．颜色物理和CIE标准色度系统[J]．物理，1982，11(4)：202-208．
[82] Lee H C. Color Imaging Science[M]. Cambridge: Cambridge University Press,2005.
[83] Wang Y, Loe K F, Wu J K. A dynamic conditional random field model for foreground and shadowsegmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(2):279-289.
[84] Land E H, McCann J J. Lightness and retinex theory[J]. Journal of the Optical Society of America,1971,61(1):1-11.
[85]解兰昌，黄峰，郭铭，等．颜色在视觉的心理学认知结构[J]．光学仪器，1995，17(4)：85-86．
[86] Brainard D H, Wandell B A. Analysis of the retinex theory of color vision[J]. JOSA A,1986,3(10):1651-1661.
[87] Maloney L T, Wandell B A. Color constancy: a method for recovering surface spectral reflectance[J].JOSA A,1986,3(1):29-33.
[88] Meyer G W, Greenberg D P. Perceptual color spaces for computer graphics[C]. In: ACM SIGGRAPHComputer Graphics. ACM,1980,14(3):254-261.
[89] Hinchey M, Pagnoni A, Rammig F J, et al. Color theory and its application in art and design[M].Springer Publishing Company, Incorporated,2008.
[90] Smith A R. Color gamut transform pairs[C]. In: ACM Siggraph Computer Graphics. ACM,1978,12(3):12-19.
[91] Gevers T, Smeulders A W M. Color-based object recognition[J]. Pattern Recognition,1999,32(3):453-464.
[92] Lin C T, Yang C T, Shou Y W, et al. An efficient and robust moving shadow removal algorithm and itsapplications in ITS[J]. EURASIP Journal on Advances in Signal Processing,2010,2010:39.
[93] Xiao M, Han C Z, Zhang L. Moving shadow detection and removal for traffic sequences[J].International Journal of Automation and Computing,2007,4(1):38-46.
[94] Jacques J C S, Jung C R, Musse S R. Background subtraction and shadow detection in grayscale videosequences[C]. In:18th Brazilian Symposium on Computer Graphics and Image Processing,2005.SIBGRAPI2005. IEEE,2005:189-196.
[95] Gonzalez R C, Woods R E, Eddins S L．冈萨雷斯数字图像处理(第二版)[M]．北京：电子工业出版社，2006．
[96]冈萨雷斯，伍兹，埃丁斯，等．数字图像处理：MATLAB版[M]．北京：电子工业出版社，2005．
[97]傅祖芸．信息论：基础理论与应用[M]．北京：电子工业出版社，2001．
[98] Ojala T, Pietik inen M, Harwood D. A comparative study of texture measures with classificationbased on featured distributions[J]. Pattern Recognition,1996,29(1):51-59.
[99] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotation invariant textureclassification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,2002,24(7):971-987.
[100] Heikkila M, Pietikainen M. A texture-based method for modeling the background and detectingmoving objects[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(4):657-662.
[101] Daugman J G. Two-dimensional spectral analysis of cortical receptive field profiles[J]. VisionResearch,1980,20(10):847-856.
[102] Daugman J G. Uncertainty relation for resolution in space, spatial frequency, and orientationoptimized by two-dimensional visual cortical filters[J]. Optical Society of America, Journal, A: Optics andImage Science,1985,2(7):1160-1169.
[103] Xu D, Liu J, Li X, et al. Insignificant shadow detection for video segmentation[J]. IEEE Transactionson Circuits and Systems for Video Technology,2005,15(8):1058-1064.
[104] Guo R, Dai Q, Hoiem D. Single-image shadow detection and removal using paired regions[C]. In:2011IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE,2011:2033-2040.
[105] Yao J, Odobez J M. Multi-layer background subtraction based on color and texture[C]. In: IEEEConference on Computer Vision and Pattern Recognition,2007. IEEE,2007:1-8.
[106]刘宏，李锦涛，刘群，等．融合颜色和梯度特征的运动阴影消除方法[J]．计算机辅助设计与图形学学报，2007，19(10)：1279-1285．
[107] Hamad A M, Tsumura N. Background updating and shadow detection based on spatial, color, andtexture information of detected objects[J]. Optical Review,2012,19(3):182-197.
[108] Boroujeni H S, Charkari N M. Robust moving shadow detection with hierarchical mixture of MLPexperts[J]. Signal, Image and Video Processing,2012:1-15.
[109] Liu J, Li D, Fang T. Shadow detection for color remotely sensed images based on multi-featureintegration[J]. Journal of Applied Remote Sensing,2012,6(1):063521-1-063521-12.
[110] Bullkich E, Ilan I, Moshe Y, et al. Moving shadow detection by nonlinear tone-mapping[C]. In:201219th International Conference on Systems, Signals and Image Processing (IWSSIP). IEEE,2012:146-149.
[111] Herodotou N, Plataniotis K N, Venetsanopoulos A N. A color segmentation scheme for object-basedvideo coding[C]. In:1998IEEE Symposium on Advances in Digital Filtering and Signal Processing. IEEE,1998:25-29.
[112] Hel-Or Y, Hel-Or H, David E. Fast template matching in non-linear tone-mapped images[C]. In:2011IEEE International Conference on Computer Vision (ICCV). IEEE,2011:1355-1362.
[113] Otsu N. A threshold selection method from gray-level histograms[J]. Automatica,1975,11(285-296):23-27.
[114] http://cvrr.ucsd.edu/aton/shadow/.
[115] http://vision.gel.ulaval.ca/~CastShadows/.
[116] Shadow ground-truth data: http://itee.uq.edu.au/~uqasanin/.
[117] CAVIAR dataset: http://homepages.inf.ed.ac.uk/rbf/CAVIAR/.
[118] Soh Y S, Lee H, Wang Y. Invariant color model-based shadow removal in traffic image and a newmetric for evaluating the performance of shadow removal methods[M]. PRICAI2006: Trends in ArtificialIntelligence. Springer Berlin Heidelberg,2006:544-552.
[119] Hahn G J, Shapiro S S. Statistical models in engineering[M]. Statistical models in engineering. JohnWiley&Sons,1968.
[120] Rosin P L, Ellis T J. Image difference threshold strategies and shadow detection[C]: In: Proceedingsof the Sixth British Machine Vision Conference, Birmingham, UK, July1995:347-356.
[121] Mitra B K, Young R, Chatwin C. On shadow elimination after moving region segmentation based ondifferent threshold selection strategies[J]. Optics and Lasers in Engineering,2007,45(11):1088-1093.
[122] Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(11):1254-1259.
[123] Greenspan H, Belongie S, Goodman R, et al. Overcomplete steerable pyramid filters and rotationinvariance[C]. In: Proceedings of1994IEEE Computer Society Conference on Computer Vision andPattern Recognition. IEEE,1994:222-228.
[124] Engel S, Zhang X, Wandell B. Colour tuning in human visual cortex measured with functionalmagnetic resonance imaging[J]. Nature,1997,388(6637):68-71.
[125] Jolliffe I. Principal component analysis[M]. John Wiley&Sons, Ltd,2005.
[126] Fukunaga K. Introduction to statistical pattern recognition[M]. Academic press,1990.
[127] Nguyen D V, Rocke D M. Tumor classification by partial least squares using microarray geneexpression data[J]. Bioinformatics,2002,18(1):39-50.
[128] Vapnik V. Statistical learning theory.1998[J].1998.
[129] Frey B J, Dueck D. Clustering by passing messages between data points[J]. Science,2007,315(5814):972-976.
[130] Arivazhagan S, Ganesan L, Priyal S P. Texture classification using Gabor wavelets based rotationinvariant features[J]. Pattern Recognition Letters,2006,27(16):1976-1982.
[131] Li M, Staunton R C. Optimum Gabor filter design and local binary patterns for texturesegmentation[J]. Pattern Recognition Letters,2008,29(5):664-672.
[132] Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based on immersionsimulations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583-598.