立体匹配技术的研究
|
摘要
人类感知的外部信息至少80%以上都是由视觉获得的，视觉是人类观察世界和认知世界的重要手段。随着科学技术的日益发展，人们尝试给计算机、机器人和其它智能机器赋予视觉功能，即像人一样用视觉系统来“看”这个世界，这样计算机视觉这门学科就产生了，并在无人机、机器人导航、三维重建等领域得到了广泛应用。
计算机视觉主要分为四个步骤：图像获取、图像校正、立体匹配和三维重建。其中，立体匹配的目的是在两个或多个对应同一场景的图像中找到匹配点，生成视差图。视差图可以通过一些简单的几何关系转换成深度图，用于三维重建。立体匹配是计算机视觉领域一个瓶颈问题，其结果的好坏直接影响着三维重建的效果。
近些年，许多学者都在研究立体匹配问题，提出了许多算法，主要分为特征匹配算法和区域匹配算法。特征匹配算法主要是提取图像特征进行匹配，生成视差图。由于只提取局部特征，因此特征匹配算法速度快，但是得到的都是稀疏的视差图，而稀疏的视差图在很多应用中都不适用。区域匹配算法依据是否使用全局搜索可以分为全局匹配算法和局部匹配算法。全局匹配算法是基于像素的，通常将匹配运算构建在一个能量最小化的框架下，然后使用优化算法来最小化或最大化能量函数，得到视差图。比较经典的全局匹配算法有：置信度传播算法（belief propagation）、图割算法（graph cut）等。全局匹配算法得到的视差图较为准确，但运行时间长。局部匹配算法将像素代价聚集在一个支持窗中，然后选择与其相匹配的支持窗。然而，在这个过程中，选择合适的支持窗是一个困难的问题。2006年Yoon提出了自适应权值算法，固定支持窗的大小，赋予支持窗中每个像素不同的权值。Yoon算法的性能超过了一般的局部算法，甚至可以和全局算法相媲美，因此得到了很多的关注。
本论文主要针对全局匹配算法中的置信度传播算法和局部匹配算法中的自适应权值算法做研究，目的是提高视差图的准确度、缩短算法时间及处理遮挡问题等。本论文的主要工作及创新点如下：
（1）基于置信度传播的立体匹配算法作为一个全局算法，能得到质量较好的视差图，但需要耗费大量的时间。为了提高置信度传播算法的效率，提出了一个基于运动估计的立体视频匹配算法。该算法首先通过传统的置信度传播算法获得I帧的视差图；然后，对于P帧，通过参考I帧的运动估计信息，得到重新排列的I帧视差值的传递信息，将其作为P帧置信度传播算法的初值进行迭代运算，从而大大减少了P帧置信度传播算法的迭代次数。实验结果表明，该算法能够大大提高置信度传播算法应用在立体视频上的匹配效率。
（2）置信度传播算法通常在灰度空间中进行计算，这会造成图像信息缺失，使得到的视差图不准确。为了提高视差图的性能，提出了RGB矢量空间，并在RGB矢量空间中用置信度传播算法对图像对进行立体匹配运算。实验证明了该方法的有效性。
（3）Yoon2006年提出了自适应权值立体匹配算法，这种算法作为一个局部算法，得到能和全局算法相媲美的视差图，因此近些年来受到了学者们的关注。但是自适应权值算法得到的视差图仍不是很完美，为了提高视差图的质量，本文在Yoon算法上增添了图像梯度信息，使得像素的特征更加明显，降低了匹配模糊性。然后，根据梯度信息，RGB颜色信息和距离来计算权值和匹配代价。实验结果表明，基于图像梯度信息的自适应权值算法比Yoon算法匹配的结果更加准确。
（4）遮挡是立体匹配中面临的大问题之一，遮挡是指匹配图像对中，一个图像的像素在另一个图像中找不到对应匹配点，导致最终出现匹配错误。为了提高视差图质量，减少遮挡区域，将双目立体视觉扩到三目立体视觉，提出了基于RGB矢量空间的三目立体匹配方法，并以自适应权值算法为例，将基于图像梯度信息的自适应权值算法扩展到三个视点进行实验。实验结果表明，三目算法不仅提高了遮挡区的匹配正确率，同时，由于匹配过程中增加了对偶运算，使得非遮挡区的匹配正确率也有所提高。
More than eighty percent of external information can be obtained by human vision.Vision is an important way for people to know something about the world. With the rapiddevelopment of technology, people try to give visual function to computers, robots, andother intelligent machines. Therefore, computer vision emerged as a new discipline.Computer vision is widely used in many fields such as drones, navigation andthree-dimensional measurement.
Computer vision includes four steps: image acquisition, image calibration, stereomatching and three-dimensional reconstruction. Stereo matching aims to find thecorresponding points in two or more images taken from the same scene. The relative shiftof position of corresponding points is called disparity. With an accurate disparity map, thedepths of the points in space can be yielded via simple geometric computation, which isuseful in three-dimensional reconstruction. Stereo matching is an important problem, andthe quality of disparity map will directly affect the reconstruction results.
In recent years, many researchers have been working on stereo matching. A lot ofalgorithms including feature-based algorithms and area-based algorithms are proposed.Feature-based algorithms extract local features from a stereo image pair and obtaindisparity maps by finding the corresponding features. The full range pixel correspondenceestimation is reduced to a sparse set of pixel correspondence estimation due to extractingonly local features. Therefore, the feature-based algorithms are time-saving, but usuallygenerate sparse disparity maps which are not appropriate for some applications.Area-based algorithms are classified into the local algorithms and the global algorithmsdepending on whether using global reasoning. The global algorithms have some classicalmethods, such as belief propagation and graph cut. The global algorithms are generallysummarized into a frame based on minimizing a global energy function, and the finaldisparity maps are iteratively obtained by minimizing or maximizing the energy functionusing some optimizing algorithms. The global algorithms usually can generate accurateand dense disparity maps, but are very time-consuming. The local algorithms aggregatesimilarities in a support window around each pixel. However, the process of finding theoptimal window for each pixel is a challenge. Adaptive support-weight methods proposedby Yoon in2006try to assign different support-weights to the pixels in a fixed-sizesupport window. Yoon’s algorithm outperforms other local methods and can becomparable to some global methods, which obtains much attention.
Belief propagation and adaptive support-weight method are discussed in this paper.The aim is to improve some problems, such as the accuracy of depth map, the efficiencyof the algorithm, occlusions and so on. The main work and innovation are as follows:
(1) Belief propagation as one of the global algorithms can obtain more accuratedisparity map than local algorithms, but is very time-consuming. To improve theperformance of the algorithm, this paper presents a stereo video matching algorithm basedon motion estimation. Firstly, the traditional BP algorithm is used to get the disparity mapof I frame. Then, for P frame, the rearranged propagating messages through referring to motion estimation information from I frame are used as the initial values for iteration ofbelief propagation algorithm. Thus this reduced the number of iterative times. Experimentresults show that the proposed stereo video matching algorithm based on motionestimation can dramatically enhance the efficiency of stereo video matching.
(2) The appearances of the pixels are ambiguous because of only making use of grayinformation of the pixels in the stereo image, which causes the poor performance. Toimprove the performance, pixels are considered in the RGB vector space. Then disparitymaps can be obtained by belief propagation method in the RGB vector space.Experimental results show that the proposed method is very effective.
(3) Yoon proposed the adaptive support-weight algorithm in2006. Yoon’s algorithmoutperforms other local methods and can be comparable to some global methods, whichobtains much attention in recent years. Gradient similarity is a simple, yet powerful, datadescriptor which shows robustness in stereo matching. Based on the adaptivesupport-weight approach, a matching algorithm, which uses the pixel gradient similarity,color similarity, and proximity in the RGB vector space to compute the correspondingsupport-weights and dissimilarity measurements, is proposed. The experimental results areevaluated on the Middlebury stereo benchmark, showing that our algorithm outperformsother stereo matching algorithms and the algorithm with gradient similarity can obtainbetter results in stereo matching.
(4) Occlusion is one of the most challenge problems in stereo matching. In a stereoimage pair, a pixel is called occluded pixel when it only can be seen in one image. Theoccluded pixel can not obtain an accurate disparity. In order to improve the performanceof stereo matching in some regions such as occlusion region, an improved method oftrinocular stereo matching is proposed. The experimental results are evaluated on theMiddlebury stereo benchmark, showing that our algorithm outperforms binocular stereomatching and obtains an accurate disparity map.
计算机视觉主要分为四个步骤：图像获取、图像校正、立体匹配和三维重建。其中，立体匹配的目的是在两个或多个对应同一场景的图像中找到匹配点，生成视差图。视差图可以通过一些简单的几何关系转换成深度图，用于三维重建。立体匹配是计算机视觉领域一个瓶颈问题，其结果的好坏直接影响着三维重建的效果。
近些年，许多学者都在研究立体匹配问题，提出了许多算法，主要分为特征匹配算法和区域匹配算法。特征匹配算法主要是提取图像特征进行匹配，生成视差图。由于只提取局部特征，因此特征匹配算法速度快，但是得到的都是稀疏的视差图，而稀疏的视差图在很多应用中都不适用。区域匹配算法依据是否使用全局搜索可以分为全局匹配算法和局部匹配算法。全局匹配算法是基于像素的，通常将匹配运算构建在一个能量最小化的框架下，然后使用优化算法来最小化或最大化能量函数，得到视差图。比较经典的全局匹配算法有：置信度传播算法（belief propagation）、图割算法（graph cut）等。全局匹配算法得到的视差图较为准确，但运行时间长。局部匹配算法将像素代价聚集在一个支持窗中，然后选择与其相匹配的支持窗。然而，在这个过程中，选择合适的支持窗是一个困难的问题。2006年Yoon提出了自适应权值算法，固定支持窗的大小，赋予支持窗中每个像素不同的权值。Yoon算法的性能超过了一般的局部算法，甚至可以和全局算法相媲美，因此得到了很多的关注。
本论文主要针对全局匹配算法中的置信度传播算法和局部匹配算法中的自适应权值算法做研究，目的是提高视差图的准确度、缩短算法时间及处理遮挡问题等。本论文的主要工作及创新点如下：
（1）基于置信度传播的立体匹配算法作为一个全局算法，能得到质量较好的视差图，但需要耗费大量的时间。为了提高置信度传播算法的效率，提出了一个基于运动估计的立体视频匹配算法。该算法首先通过传统的置信度传播算法获得I帧的视差图；然后，对于P帧，通过参考I帧的运动估计信息，得到重新排列的I帧视差值的传递信息，将其作为P帧置信度传播算法的初值进行迭代运算，从而大大减少了P帧置信度传播算法的迭代次数。实验结果表明，该算法能够大大提高置信度传播算法应用在立体视频上的匹配效率。
（2）置信度传播算法通常在灰度空间中进行计算，这会造成图像信息缺失，使得到的视差图不准确。为了提高视差图的性能，提出了RGB矢量空间，并在RGB矢量空间中用置信度传播算法对图像对进行立体匹配运算。实验证明了该方法的有效性。
（3）Yoon2006年提出了自适应权值立体匹配算法，这种算法作为一个局部算法，得到能和全局算法相媲美的视差图，因此近些年来受到了学者们的关注。但是自适应权值算法得到的视差图仍不是很完美，为了提高视差图的质量，本文在Yoon算法上增添了图像梯度信息，使得像素的特征更加明显，降低了匹配模糊性。然后，根据梯度信息，RGB颜色信息和距离来计算权值和匹配代价。实验结果表明，基于图像梯度信息的自适应权值算法比Yoon算法匹配的结果更加准确。
（4）遮挡是立体匹配中面临的大问题之一，遮挡是指匹配图像对中，一个图像的像素在另一个图像中找不到对应匹配点，导致最终出现匹配错误。为了提高视差图质量，减少遮挡区域，将双目立体视觉扩到三目立体视觉，提出了基于RGB矢量空间的三目立体匹配方法，并以自适应权值算法为例，将基于图像梯度信息的自适应权值算法扩展到三个视点进行实验。实验结果表明，三目算法不仅提高了遮挡区的匹配正确率，同时，由于匹配过程中增加了对偶运算，使得非遮挡区的匹配正确率也有所提高。
More than eighty percent of external information can be obtained by human vision.Vision is an important way for people to know something about the world. With the rapiddevelopment of technology, people try to give visual function to computers, robots, andother intelligent machines. Therefore, computer vision emerged as a new discipline.Computer vision is widely used in many fields such as drones, navigation andthree-dimensional measurement.
Computer vision includes four steps: image acquisition, image calibration, stereomatching and three-dimensional reconstruction. Stereo matching aims to find thecorresponding points in two or more images taken from the same scene. The relative shiftof position of corresponding points is called disparity. With an accurate disparity map, thedepths of the points in space can be yielded via simple geometric computation, which isuseful in three-dimensional reconstruction. Stereo matching is an important problem, andthe quality of disparity map will directly affect the reconstruction results.
In recent years, many researchers have been working on stereo matching. A lot ofalgorithms including feature-based algorithms and area-based algorithms are proposed.Feature-based algorithms extract local features from a stereo image pair and obtaindisparity maps by finding the corresponding features. The full range pixel correspondenceestimation is reduced to a sparse set of pixel correspondence estimation due to extractingonly local features. Therefore, the feature-based algorithms are time-saving, but usuallygenerate sparse disparity maps which are not appropriate for some applications.Area-based algorithms are classified into the local algorithms and the global algorithmsdepending on whether using global reasoning. The global algorithms have some classicalmethods, such as belief propagation and graph cut. The global algorithms are generallysummarized into a frame based on minimizing a global energy function, and the finaldisparity maps are iteratively obtained by minimizing or maximizing the energy functionusing some optimizing algorithms. The global algorithms usually can generate accurateand dense disparity maps, but are very time-consuming. The local algorithms aggregatesimilarities in a support window around each pixel. However, the process of finding theoptimal window for each pixel is a challenge. Adaptive support-weight methods proposedby Yoon in2006try to assign different support-weights to the pixels in a fixed-sizesupport window. Yoon’s algorithm outperforms other local methods and can becomparable to some global methods, which obtains much attention.
Belief propagation and adaptive support-weight method are discussed in this paper.The aim is to improve some problems, such as the accuracy of depth map, the efficiencyof the algorithm, occlusions and so on. The main work and innovation are as follows:
(1) Belief propagation as one of the global algorithms can obtain more accuratedisparity map than local algorithms, but is very time-consuming. To improve theperformance of the algorithm, this paper presents a stereo video matching algorithm basedon motion estimation. Firstly, the traditional BP algorithm is used to get the disparity mapof I frame. Then, for P frame, the rearranged propagating messages through referring to motion estimation information from I frame are used as the initial values for iteration ofbelief propagation algorithm. Thus this reduced the number of iterative times. Experimentresults show that the proposed stereo video matching algorithm based on motionestimation can dramatically enhance the efficiency of stereo video matching.
(2) The appearances of the pixels are ambiguous because of only making use of grayinformation of the pixels in the stereo image, which causes the poor performance. Toimprove the performance, pixels are considered in the RGB vector space. Then disparitymaps can be obtained by belief propagation method in the RGB vector space.Experimental results show that the proposed method is very effective.
(3) Yoon proposed the adaptive support-weight algorithm in2006. Yoon’s algorithmoutperforms other local methods and can be comparable to some global methods, whichobtains much attention in recent years. Gradient similarity is a simple, yet powerful, datadescriptor which shows robustness in stereo matching. Based on the adaptivesupport-weight approach, a matching algorithm, which uses the pixel gradient similarity,color similarity, and proximity in the RGB vector space to compute the correspondingsupport-weights and dissimilarity measurements, is proposed. The experimental results areevaluated on the Middlebury stereo benchmark, showing that our algorithm outperformsother stereo matching algorithms and the algorithm with gradient similarity can obtainbetter results in stereo matching.
(4) Occlusion is one of the most challenge problems in stereo matching. In a stereoimage pair, a pixel is called occluded pixel when it only can be seen in one image. Theoccluded pixel can not obtain an accurate disparity. In order to improve the performanceof stereo matching in some regions such as occlusion region, an improved method oftrinocular stereo matching is proposed. The experimental results are evaluated on theMiddlebury stereo benchmark, showing that our algorithm outperforms binocular stereomatching and obtains an accurate disparity map.
引文
[1]张广军.机器视觉[M].第二版.北京：机械工业出版社，2005.
[2]池凌鸿.立体匹配算法的研究与应用[D].中国科技大学，2011.
[3] L G，Roberts. Machine perception of three-dimensional solids, In Optical andElectroptical Information Processing,1965[C]. MIT Press, Cambridge.
[4] D.Marr. Vision, W.H.Freeman and Company [M]. San Francisco,1982.
[5] Xing Yinglin, Fang Lin. A Stereo Matching Method Based on Chain Code Vector [J].Computer Science and Engineering,2009,(1):372-375.
[6]伊世明.立体匹配中若干问题的研究[D].国防科学技术大学,2006.
[7] Z Zhang, R Deriche, O D Faugeras, Q.T. Luong. A robust technique for matchingtwo uncalibrated images through the recovery of the unknown epipolar geometry [J].Artificial Intelligence,1994,78(1):87–119.
[8] J H McIntosh, K M Mutch, Matching straight lines Computer Vision [J]. Graphicsand Image Processing,1998,43(3).
[9]王继阳，文贡坚，李德仁.直线特征立体匹配中的不确定性问题[J].信号处理，2010,26(5).
[10]高峰,文贡坚.利用仿射几何的仿射不变特征提取方法[J].中国图象图形学报,2011,16(3):389~397.
[11]施陈博.快速图像配准和高精度立体匹配算法研究[D].清华大学博士论文,2011.
[12] Maxime Lhuillier, Long Quan. Match Propagation for Image-Based Modeling andRendering [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(8):1140-1146.
[13] Bin Fan, Chunlei Huo, Chunhong Pan, Qingqun Kong. Registration of Optical andSAR Satellite Images by Exploring the Spatial Relationship of the Improved SIFT[J]. IEEE Geoscience and Remote Sensing Letters,2013,10(4):657-661.
[14] Bin Fan, Fuchao Wu, Zhanyi Hu. Robust Line Matching through Line-pointInvariants [J]. Pattern Recognition,2012,45(2):794-805.
[15] Bin Fan, Fuchao Wu, Zhanyi Hu. Towards Reliable Matching of ImagesContaining Repetitive Patterns [J]. Pattern Recognition Letters,2011,32(14):1851-1859.
[16] Zhenhua Wang, Bin Fan, Fuchao Wu. Local Intensity Order Pattern for FeatureDescription, In Proceedings of International Conference on Computer Vision, Nov.6-13,2011[C]. Barcelona, Spain.
[17] Bin Fan, Fuchao Wu, Zhanyi Hu, Aggregating Gradient Distributions into IntensityOrders: A Novel Local Image Descriptor, In Proceedings of IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, June20-25,2011[C]. Colorado Springs, USA.
[18] A. Klaus, M. Sormann, K. Karner. Segment-based stereo matching using beliefpropagation and a self-adapting dissimilarity measure, In18th InternationalConference Pattern Recognition, August20-24,2006[C]. pp.15-18, Hong Kong,China.
[19] Quanquan Gu, Jie Zhou. Belief propagation on Riemannian manifold for stereomatching [J]. Image Processing,2008:1788-1791.
[20] J Sun, N N Zheng, H Y Shum. Stereo matching using belief propagation [J].Transactions on Pattern Analysis and Machine Intelligence,2003,25(7):787-800.
[21] F Z Wang, D G Huang, G Sen. Belief propagation stereo matching based ondifferential geometry constraint of disparity, in Proceedings of the InternationalConference on Digital Manufacturing and Automation, December18-20,2010[C].vol.1, pp.324-327. Chendu, China.
[22] Yedidia, J S, Freeman. Understanding Belief Propagation and Its Generalizations[J]. Exploring Artificial Intelligence in the New Millennium,2003:239-236.
[23] L Hong, G Chen. Segment-based stereo matching using graph cuts, in Proceedingsof IEEE Conference on Computer Vision and Pattern Recognition, June2-27,2004[C], vol.1, pp.74-81.
[24] V Kolmogorov, R Zabih. Multi-camera scene reconstruction via graph cuts,European Conference on Computer Vision, May28-31,2002[C], vol.3, pp.82-96.Berlin: Springer.
[25] V Kolmogorov, R Zabih. What Energy Funetions can be minimized via Graph Cuts?[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2004,26(2):147-159.
[26] Y Boykov, O Veksler, R.Zabih. Fast Approximate Energy Minimization via GraphCuts [J]. Pattern Analysis and Machine Intelligence,2001,23(11):1222-1239.
[27] Cheng Lei, Jason Selzer, Yee-Hong Yang. Region-Tree Based Stereo Using DynamicProgramming Optimization, IEEE Conference on Computer Vision and PatternRecognition, June17-22,2006[C], vol.2, pp.2378-2385.
[28] Won Pyo Dong, Yun Seok Lee, Chang Sung Jeong. A Stereo Matching UsingVariable Windows and Dynamic Programming,18th Australian Joint Conference onArtificial Intelligence, December5-9,2005[C]. Sydney, Australia.
[29] Yuichi Ohta, Takehiko Yamamoto, Katsuo Ikeda. Collinear Trinocular Stereo UsingTwo Level Dynamic Programming,9th International Conference on PatternRecognition, Nov14-17,1988[C], vol.2, pp.658-662.
[30] Yi Deng, Xueyin Lin. A Fast Line Segment Based Dense Stereo Algorithm UsingTree Dynamic Programming,9th European Conference on Computer Vision, May7-13,2006[C]. Graz, Austria.
[31] A Criminisi, A Blake, C Rother, J Shotton. Effcient Dense Stereo with Occlusionsfor New View-Synthesis by Four-State Dynamic Programming [J]. InternationalJournal of Computer Vision,2007,71(1):89-110.
[32] Minglun Gong, Yee Hong Yang, Real-Time Stereo Matching Using OrthogonalReliability-Based Dynamic Programming [J]. IEEE Transactions on ImageProcessing,2007,16(3):879-884.
[33] Lei Yang, Chao Lu, Jian Sun, Approximate dynamic programming for continuousstate and control problems [J]. Control and Automation,2009:1415-1420.
[34] T Kanada, M Okutomi. A stereo matching algorithm with an adaptive window:theory and experiment [J]. IEEE Transaction Pattern Analysis and MachineIntelligence,1994,16(9):920-932.
[35] K Zhang, J Lu, G Lafruit. Scalable stereo matching with locally adaptive polygonapproximation, in Proceedings of the15th IEEE International Conference ImageProcessing, Oct.12-15,2008[C], pp.313-316. San Diego, CA.
[36] G. Luo, X Yang, Q Xu. Fast stereo matching algorithm using adaptive window [J].International Symposiums Information Processing,2008:25-30.
[37] X Su, T M Khoshgoftaar. Arbitrarily-shaped window based stereo matching usingthe go-light optimization algorithm, in IEEE International Conference ImageProcessing, Sept.16-Oct.19,2007[C]. pp. VI-556-559. San Antonio, TX.
[38] M Okutomi, T Kanade. A locally adaptive window for signal matching, ThirdInternational Conference on Computer Vision, Dec.4-7,1990[C]. pp.190-199.Osaka.
[39] Y Boykov, O Veksler, R Zabih. A variable window approach to early vision [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(12):1283–1294.
[40] O Veksler. Stereo matching by compact windows via minimum ratio cycle. EighthIEEE International Conference on Computer Vision, Jul.7-14,2001[C]. vol.1, pp.540-547.
[41] L De Maeztu, A Villanueva, R Cabgeza. Stereo matching using gradient similarityand locally adaptive supportweight [J]. Pattern Recognition Letter,2011,32:1643-1651.
[42] J Lu, K Zhang, G Lafruit, F Catthoor. Real-time stereo matching: a cross-based localapproach, in IEEE International Conference on Acoustics, Speech and SignalProcessing, April19-24,2009[C]. pp.733-736. Taipei.
[43] A Hosni, M Bleyer, M Gelautz. Local stereo matching using geodesic supportweights,16th IEEE International Conference Image Processing, Nov.7-10,2009[C].pp.2093-2096. Cairo.
[44] Z Gu, X Su, Y Liu, Q Zhang. Local stereo matching with adaptive support-weight,rank transform and disparity calibration [J]. Pattern Recognition Letter,2008,29:1230-1235.
[45] Zhengang Zhai, Yao Lu, Hong Zhao. Stereo Matching with Adaptive ArbitrarySupport-Pixel Set and Adaptive Support-Weight，International Conference onComputational Intelligence and Security, Dec.11-14,2009[C]. vol.1, pp.598-602.Beijing.
[46] Y Xu, D Wang, T Feng, H Y Shum. Stereo Computation using Radial AdaptiveWindows,16th International Conference on Pattern Recognition,2002[C]. vol.3, pp.595-598.
[47] K J Yoon, I S Kweon. Adaptive support-weight approach for correspondence search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2006,28:650-656.
[48] F Tombari, S Mattoccia, L Di Stefano. Segmentation based adaptive support foraccurate stereo correspondence, in Proceedings of the Pacific-Rim Symposium onImage and Video Technology, Decemeber17-19,2007[C]. pp.427-428. Santiago.
[49] Dibyendu Mukherjee, Guanghui Wang. Stereo correspondence based on curveletdecomposition, support weights, and disparity calibration [J]. SPIC Journal ofOptical Engineering,2010,49(4):077003-1-8.
[50] Satyajit Anil Adhyapak, Nasser Kehtarnavaz, Mihai Nadin. Stereo matching viaselective multiple windows [J]. Journal of Electronic Imaging,2007,16(1).
[51] A Fusiello, V Roberto, E Trucco. Symmetric stereo with multiple windowing [J].Journal of Pattern Recognition and Artificial Intelligence,2000,14(8):1053–1066.
[52] J Jeon, C Kim, Y Sung Ho. Sharp and dense disparity maps using multiple windows,IEEE Pacific Rim Conference on Multimedia,2002[C]. pp.1057-1064.
[53] A Fusiello, Roberto V, Trucco E. Efficient Stereo with Multiple Windowing, IEEEComputer Society Conference on Computer Vision and Pattern Recognition,1997[C]. pp.858-863.
[54] D Scharsterein, R Szeliski. A taxonomy and evaluation of dense two-frame stereocorrespondence algorithms. Proceedings IEEE Workshop on Stereo andMulti-Baseline Vision, Dec.9-10,2001[C]. Kauai, HI.
[55] M J Hannah. Computer Matching of Areas in Stereo Images [D]. Stanford University,1974.
[56] P Anandan. A computational framework and an algorithm for the measurement ofvisual motion [J]. International Journal of Computer Vision,1989,2(3):283–310.
[57] L Matthies, R Szeliski, T Kanade. Kalman filter-based algorithms for estimatingdepth from image sequences [J]. International Journal of Computer Vision,1989,3:209-236.
[58] E P Simoncelli, E H Adelson, D J Heeger. Probability distributions of optical flow.In proceeding of Computer Vision and Patteren Recognition. June,1991[C]. pp.310-315. Hawaii.
[59] T Kanade. Development of a video-rate stereo machine, In Image UnderstandingWorkshop,1994[C]. pp.549-557. Monterey, CA. Morgan Kaufmann Publishers.
[60] Fu He, F Da. Stereo matching based on dissimilar intensity support, ChineseConference on Pattern Recognition, Oct.21-23,2010[C]. pp.1-5. Chongqing, China.
[61] Crouzil A, IRIT-UPS Toulouse France, Massip-Pailhes L, Castan S. A NewCorrelation Criterion Based on Gradient Fields Similarity, Proceedings of the13thInternational Conference on Pattern Recognition, Aug25-29,1996[C]. vol.1, pp.632-636. Vienna.
[62] Delaunoy Ama l, Prados Emmanuel, Gradient Flows for Optimizing TriangularMesh-based Surfaces: Applications to3D Reconstruction Problems Dealing withVisibility [J], International Journal of Computer Vision, November,2011,95(2):100-123.
[63] Leonardo De-Maeztu, Arantxa Villanueva, Rafael Cabeza, Stereo matching usinggradient similarity and locally adaptive support-weight [J], Pattern RecognitionLetters, October,2011,32(13):1643–1651.
[64] Simon Hermann, Tobi Vaudrey.The Gradient-A Powerful and Robust CostFunction for Stereo Matching,25th International Conference of Image and VisionComputing New Zealand, Nov.8-9,2010[C]. pp.1-8. Queenstown.
[65] D Scharstein. Matching images by comparing their gradient fields. Conference A:Computer Vision& Image Processing, Proceedings of the12th IAPRInternational Conference on Pattern Recognition, Oct.9-13,1994[C]. vol.1, pp.572–575. Jerusalem.
[66] R Zabih, J Woodfill. Non-parametric local transforms for computing visualcorrespondence, Third European Conference on Computer Vision Stockholm, May2–6,1994[C]. vol.II, pp.151-158. Sweden.
[67] Banks J，Bennamout M. Reliablity analysis of the rank transform for stereomatching [J]. IEEE Transactions on Systems, Man, and Cybernetics-Prat B：Cybernetics,2001,31(6).
[68] Banks J, Bennamoun M, Corke P. Non-parametric techniques for fast and robuststereo matching, In IEEE TENCON’97Speech and Image Technologies forComputing and Telecommunications Proceedings, Dec.4-4,1997[C]. pp.365-368.Brisbane, Australia.
[69]马璇,朱世强,赖小波. RANK变换在立体匹配中的应用研究[J],系统仿真学报,2007,19(9).
[70] R Bolles, H Baker, M Hannah. The JISCT stereo evaluation, Image UnderstandingWorkshop, pp.263-274,1993[C].
[71]卢思军,唐振民.彩色图像非参数变换立体匹配研究[J],计算机工程与应用,2009,45(2).
[72] Z Gu, X Su, Y Liu, Q Zhang. Local stereo matching with adaptive support-weight,rank transform, and disparity calibration [J]. Pattern Recognition Letter,2008,29:1230-1235.
[73]周龙,徐贵力,李开宇,王彪,田裕鹏,陈欣.基于Census变换和改进自适应窗口的立体匹配算法[J].航空学报,2012,第5期.
[74] S Birchfield, C Tomasi. Depth discontinuities by pixel-to-pixel stereo [J].International Journal of Computer Vision,1998,35(3):1073-1080.
[75] R D Arnold. Automated stereo perception. Technical Report AIM-351, ArtificialIntelligence Laboratory, Stanford University,1983.
[76] A F Bobick, S S Intille. Large occlusion stereo [J]. International Journal ofComputer Vision,1999,33(3):181-200.
[77] W E L Grimson. Computational experiments with a feature based stereo algorithm[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1985,7(1):17–34.
[78] S B Pollard, J E W Mayhew, J P Frisby. PMF:A stereo correspondence algorithmusing a disparity gradient limit [J]. Perception,1985,14(4):449-470.
[79] K Prazdny. Detection of binocular disparities [J]. Biological Cybernetics,1985,52(2):93-99.
[80] Sch fer R. Review and future directions for3D-video, Proceedings of the PictureCoding,2006[C]. pp.1-11. Beijing, China.
[81] Kang S B. A survey of image-based rendering techniques [J]. VideoMetrics,1999:2-16.
[82] Heung Yeung Shum, Sing Bing Kang. A review of image-based renderingtechniques, Proceedings of IEEE/SPIE Visual Communications and ImageProcessing, May30,2000[C]. pp.2-13.
[83] Luat Do, Svitlana Zinger, de With P H N. Objective quality analysis forfree-viewpoint DIBR, IEEE International Conference on Image Processing,September26-29,2010[C]. pp.2629-2632. HongKong.
[84] Daniel Berj ó n, Alexander Hornung, Francisco Mor á n, Aljoscha Smolic.Evaluation of backward mapping DIBR for FVV applications[C]. IEEEInternational Conference on Multimedia and Expo, Madrid, July11-15,2011. IEEE,2011:1-6.
[85] T W Ryan, R T Gray, B R Hunt. Prediction of correlation errors in stereo-pairimages [J]. Optical Engineering,1980,19(3):312–322.
[86] B D Lucas, T Kanade. An iterative image registration technique with an applicationin stereo vision. In Seventh International Joint Conference on Artificial Intelligence(IJCAI-81), pp.674–679,1981[C]. Vancouver.
[87] Q Tian, M N Huhns. Algorithms for subpixel registration [J]. Computer Vision,Graphics, and Image Processing,1986,35(2):220-233.
[88] L Matthies, R Szeliski, T Kanade. Kalman filter-based algorithms for estimatingdepth from image sequences [J]. International Journal of Computer Vision,1989,3(3):209-236.
[89] S B Kang, R Szeliski, J Chai. Handling occlusions in dense multi-view stereo.Proceedings of the2001IEEE Computer Society Conference on Computer Visionand Pattern Recognition,2001[C].
[90] S Birchfield, C Tomasi. A pixel dissimilarity measure that is insensitive to imagesampling [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(4):401–406.
[91] D Scharstein. View Synthesis Using Stereo Vision, Lecture Notes in ComputerScience.1999, vol.1583. Springer-Verlag.
[92] Xiaofeng Ding, Lizhong Xu, Huibin Wang, Xin Wang, Guofang Lv. Stereo depthestimation under different camera calibration and alignment errors [J]. AppliedOptics,2011,50(10):1289-1301.
[93]方恒,吴怀宇.基于MRF和颜色空间的立体图像匹配算法[J].计算机技术与发展,2008,18(12):28-30.
[94]何华君,卢朝阳,焦卫东,郭大波.基于梯度和MRF模型的视差估计算法[J].西安电子科技大学学报,2007,34(3):373-376.
[95]郑肇葆,黄桂兰.航空影像纹理分类的最小二乘法和问题的分析[J].测绘学报，1996,25(2).
[96]张蓓蕾，基于马尔可夫场理论的单目图像深度估计研究[D].东华大学,2010.
[97]郑肇葆，马尔柯夫场在影像纹理模拟中的应用[J].测绘学报,1993,22(4).
[98] S Geman, D Geman. Stochastic Relaxation, Gibbs Distributions,and the BayesianRestoration of Images [J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1984,6(6):721-741.
[99]谢琅,杨艳,曹阳,王立恒.一种改进的三步搜索块运动估计算法[J].武汉大学学报.2005,51(5):625-628.
[100]刘跃军,苏静.一种改进三步搜索算法的设计与实现[J].2008,29(4):62-64.
[101]林舒静.视频压缩中运动估计算法的研究[D].西南大学,2009.
[102] C Connolly, T Fleiss. A study of efficiency and accuracy in the transformation fromRGB to CIELAB color space. IEEE International Conference on Image Processing,1997[C], pp.1046-1048.
[103] Klaus A, Sormann M, Karner K. Segment-Based Stereo Matching Using BeliefPropagation and a Self-Adapting Dissimilarity Measure. In Proceedings of18thInternational Conference on Pattern Recognition, August20-24,2006[C]. vol.3, pp.15-18. Hong Kong, China.
[104] El Mahassani, E D. New Robust Matching Cost Functions for Stereo Vision. InProceedings of9th Biennial Conference of the Australian Pattern RecognitionSociety on Digital Image Computing Techniques and Applications,2007[C]. pp.144-150.
[105] S E Palmer. Vision Science: Photons to Phenomenology [M]. Bradford Books, MITPress,1999.
[106] D Marr. Vision. WH Freeman and Company [M].1982.
[107] S W Zucker, R A Hummel. Toward a Low-Level Description of Dot Clusters:Labeling Edge, Interior and Noise Points [J]. Computer Graphics and ImageProcessing,1979,9:213-233.
[108] N Ahuja, M Tuceryan. Extraction of Early Perceptual Structure in Dot Patterns:Integrating Region, Boundary, and Component Gestalt [J]. Computer Vision,Graphics, and Image Processing,1989,48:304-356.
[109] J Shi, J Malik. Normalized Cuts and Image Segmentation [J]. IEEE TransactionPattern Analysis and Machine Intelligence,2000,22(8):888-905.
[110] G Yingnan, Z Yan, C Hexin. Improved belief propagation based on RGB vectormeasure for stereo matching, in Proceedings of2011International Conference onWireless Communications and Signal Processing,2011[C]. pp.1–5. NanJing.
[111] F Tombari, S Mattoccia, L Di Stefano. Segmentation-based adaptive support foraccurate stereo correspondence, IEEE Pacific-Rim Symposium on Image and VideoTechnology,2007[C]:427-438.
[112] Q Yang, L Wang, N Ahuja. A constant-space belief propagation algorithm for stereomatching, IEEE Conference Computer Vision and Pattern Recognition,2010[C]:1458-1465.
[113] N Papadakis, V Caselles. Multi-label depth estimation for graph cuts stereoproblems [J]. Journal of Mathematical Imaging and Vision,2010,38(1):70-82.
[114] D Miyazaki, Y Matsushita, K Ikeuchi. Interactive shadow removal from a singleimage using hierarchical graph cut [J]. Transactions on Computer Vision andApplications,2010,2:235-252.
[115] W Hu, K Zhang, L Sun, J Li, Y Li, S Yang. Virtual support window foradaptive-weight stereo matching, IEEE Visual Communications and ImageProcessing, Nov.6-9,2011[C]. pp.1-4. Tainan.
[116] L Nalpantidis, A Gasteratos. Biologically and psychophysically inspired adaptivesupport weights algorithm for stereo correspondence [J]. Robotics and AutonomousSystems,2010,58(5):457-464.
[117] Ilkwon Park, Hyeran Byun. Depth map refinement using multiple patch-baseddepth image completion via local stereo warping [J]. Journal of OpticalEngineering,2010,49(7).
[118] R Gupta, S Y Cho. A correlation-based approach for real-time stereo matching.International Symposium on Visual Computing. Advances in Visual Computing-6thInternational Symposium, November29-December1,2010[C]. Las Vegas, NV,USA.
[119] C Rhemann, A Hosni, M Bleyer, C Rother, and M Gelautz. Fast cost-volumefiltering for visual correspondence and beyond,2011IEEE Conference on ComputerVision and Pattern Recognition,2011[C]. pp.3017-3024.
[120] Q Yang, C Engels, A Akbarzadeh. Near real-time stereo for weakly-textured scenes.British Machine Vision Conference,2008[C]:80-87.
[121] Z Wang, Z Zheng. A region based stereo matching algorithm using cooperativeoptimization, IEEE Conference on Computer Vision and Pattern Recognition,2008[C]:1-8.
[2]池凌鸿.立体匹配算法的研究与应用[D].中国科技大学，2011.
[3] L G，Roberts. Machine perception of three-dimensional solids, In Optical andElectroptical Information Processing,1965[C]. MIT Press, Cambridge.
[4] D.Marr. Vision, W.H.Freeman and Company [M]. San Francisco,1982.
[5] Xing Yinglin, Fang Lin. A Stereo Matching Method Based on Chain Code Vector [J].Computer Science and Engineering,2009,(1):372-375.
[6]伊世明.立体匹配中若干问题的研究[D].国防科学技术大学,2006.
[7] Z Zhang, R Deriche, O D Faugeras, Q.T. Luong. A robust technique for matchingtwo uncalibrated images through the recovery of the unknown epipolar geometry [J].Artificial Intelligence,1994,78(1):87–119.
[8] J H McIntosh, K M Mutch, Matching straight lines Computer Vision [J]. Graphicsand Image Processing,1998,43(3).
[9]王继阳，文贡坚，李德仁.直线特征立体匹配中的不确定性问题[J].信号处理，2010,26(5).
[10]高峰,文贡坚.利用仿射几何的仿射不变特征提取方法[J].中国图象图形学报,2011,16(3):389~397.
[11]施陈博.快速图像配准和高精度立体匹配算法研究[D].清华大学博士论文,2011.
[12] Maxime Lhuillier, Long Quan. Match Propagation for Image-Based Modeling andRendering [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(8):1140-1146.
[13] Bin Fan, Chunlei Huo, Chunhong Pan, Qingqun Kong. Registration of Optical andSAR Satellite Images by Exploring the Spatial Relationship of the Improved SIFT[J]. IEEE Geoscience and Remote Sensing Letters,2013,10(4):657-661.
[14] Bin Fan, Fuchao Wu, Zhanyi Hu. Robust Line Matching through Line-pointInvariants [J]. Pattern Recognition,2012,45(2):794-805.
[15] Bin Fan, Fuchao Wu, Zhanyi Hu. Towards Reliable Matching of ImagesContaining Repetitive Patterns [J]. Pattern Recognition Letters,2011,32(14):1851-1859.
[16] Zhenhua Wang, Bin Fan, Fuchao Wu. Local Intensity Order Pattern for FeatureDescription, In Proceedings of International Conference on Computer Vision, Nov.6-13,2011[C]. Barcelona, Spain.
[17] Bin Fan, Fuchao Wu, Zhanyi Hu, Aggregating Gradient Distributions into IntensityOrders: A Novel Local Image Descriptor, In Proceedings of IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, June20-25,2011[C]. Colorado Springs, USA.
[18] A. Klaus, M. Sormann, K. Karner. Segment-based stereo matching using beliefpropagation and a self-adapting dissimilarity measure, In18th InternationalConference Pattern Recognition, August20-24,2006[C]. pp.15-18, Hong Kong,China.
[19] Quanquan Gu, Jie Zhou. Belief propagation on Riemannian manifold for stereomatching [J]. Image Processing,2008:1788-1791.
[20] J Sun, N N Zheng, H Y Shum. Stereo matching using belief propagation [J].Transactions on Pattern Analysis and Machine Intelligence,2003,25(7):787-800.
[21] F Z Wang, D G Huang, G Sen. Belief propagation stereo matching based ondifferential geometry constraint of disparity, in Proceedings of the InternationalConference on Digital Manufacturing and Automation, December18-20,2010[C].vol.1, pp.324-327. Chendu, China.
[22] Yedidia, J S, Freeman. Understanding Belief Propagation and Its Generalizations[J]. Exploring Artificial Intelligence in the New Millennium,2003:239-236.
[23] L Hong, G Chen. Segment-based stereo matching using graph cuts, in Proceedingsof IEEE Conference on Computer Vision and Pattern Recognition, June2-27,2004[C], vol.1, pp.74-81.
[24] V Kolmogorov, R Zabih. Multi-camera scene reconstruction via graph cuts,European Conference on Computer Vision, May28-31,2002[C], vol.3, pp.82-96.Berlin: Springer.
[25] V Kolmogorov, R Zabih. What Energy Funetions can be minimized via Graph Cuts?[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2004,26(2):147-159.
[26] Y Boykov, O Veksler, R.Zabih. Fast Approximate Energy Minimization via GraphCuts [J]. Pattern Analysis and Machine Intelligence,2001,23(11):1222-1239.
[27] Cheng Lei, Jason Selzer, Yee-Hong Yang. Region-Tree Based Stereo Using DynamicProgramming Optimization, IEEE Conference on Computer Vision and PatternRecognition, June17-22,2006[C], vol.2, pp.2378-2385.
[28] Won Pyo Dong, Yun Seok Lee, Chang Sung Jeong. A Stereo Matching UsingVariable Windows and Dynamic Programming,18th Australian Joint Conference onArtificial Intelligence, December5-9,2005[C]. Sydney, Australia.
[29] Yuichi Ohta, Takehiko Yamamoto, Katsuo Ikeda. Collinear Trinocular Stereo UsingTwo Level Dynamic Programming,9th International Conference on PatternRecognition, Nov14-17,1988[C], vol.2, pp.658-662.
[30] Yi Deng, Xueyin Lin. A Fast Line Segment Based Dense Stereo Algorithm UsingTree Dynamic Programming,9th European Conference on Computer Vision, May7-13,2006[C]. Graz, Austria.
[31] A Criminisi, A Blake, C Rother, J Shotton. Effcient Dense Stereo with Occlusionsfor New View-Synthesis by Four-State Dynamic Programming [J]. InternationalJournal of Computer Vision,2007,71(1):89-110.
[32] Minglun Gong, Yee Hong Yang, Real-Time Stereo Matching Using OrthogonalReliability-Based Dynamic Programming [J]. IEEE Transactions on ImageProcessing,2007,16(3):879-884.
[33] Lei Yang, Chao Lu, Jian Sun, Approximate dynamic programming for continuousstate and control problems [J]. Control and Automation,2009:1415-1420.
[34] T Kanada, M Okutomi. A stereo matching algorithm with an adaptive window:theory and experiment [J]. IEEE Transaction Pattern Analysis and MachineIntelligence,1994,16(9):920-932.
[35] K Zhang, J Lu, G Lafruit. Scalable stereo matching with locally adaptive polygonapproximation, in Proceedings of the15th IEEE International Conference ImageProcessing, Oct.12-15,2008[C], pp.313-316. San Diego, CA.
[36] G. Luo, X Yang, Q Xu. Fast stereo matching algorithm using adaptive window [J].International Symposiums Information Processing,2008:25-30.
[37] X Su, T M Khoshgoftaar. Arbitrarily-shaped window based stereo matching usingthe go-light optimization algorithm, in IEEE International Conference ImageProcessing, Sept.16-Oct.19,2007[C]. pp. VI-556-559. San Antonio, TX.
[38] M Okutomi, T Kanade. A locally adaptive window for signal matching, ThirdInternational Conference on Computer Vision, Dec.4-7,1990[C]. pp.190-199.Osaka.
[39] Y Boykov, O Veksler, R Zabih. A variable window approach to early vision [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(12):1283–1294.
[40] O Veksler. Stereo matching by compact windows via minimum ratio cycle. EighthIEEE International Conference on Computer Vision, Jul.7-14,2001[C]. vol.1, pp.540-547.
[41] L De Maeztu, A Villanueva, R Cabgeza. Stereo matching using gradient similarityand locally adaptive supportweight [J]. Pattern Recognition Letter,2011,32:1643-1651.
[42] J Lu, K Zhang, G Lafruit, F Catthoor. Real-time stereo matching: a cross-based localapproach, in IEEE International Conference on Acoustics, Speech and SignalProcessing, April19-24,2009[C]. pp.733-736. Taipei.
[43] A Hosni, M Bleyer, M Gelautz. Local stereo matching using geodesic supportweights,16th IEEE International Conference Image Processing, Nov.7-10,2009[C].pp.2093-2096. Cairo.
[44] Z Gu, X Su, Y Liu, Q Zhang. Local stereo matching with adaptive support-weight,rank transform and disparity calibration [J]. Pattern Recognition Letter,2008,29:1230-1235.
[45] Zhengang Zhai, Yao Lu, Hong Zhao. Stereo Matching with Adaptive ArbitrarySupport-Pixel Set and Adaptive Support-Weight，International Conference onComputational Intelligence and Security, Dec.11-14,2009[C]. vol.1, pp.598-602.Beijing.
[46] Y Xu, D Wang, T Feng, H Y Shum. Stereo Computation using Radial AdaptiveWindows,16th International Conference on Pattern Recognition,2002[C]. vol.3, pp.595-598.
[47] K J Yoon, I S Kweon. Adaptive support-weight approach for correspondence search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2006,28:650-656.
[48] F Tombari, S Mattoccia, L Di Stefano. Segmentation based adaptive support foraccurate stereo correspondence, in Proceedings of the Pacific-Rim Symposium onImage and Video Technology, Decemeber17-19,2007[C]. pp.427-428. Santiago.
[49] Dibyendu Mukherjee, Guanghui Wang. Stereo correspondence based on curveletdecomposition, support weights, and disparity calibration [J]. SPIC Journal ofOptical Engineering,2010,49(4):077003-1-8.
[50] Satyajit Anil Adhyapak, Nasser Kehtarnavaz, Mihai Nadin. Stereo matching viaselective multiple windows [J]. Journal of Electronic Imaging,2007,16(1).
[51] A Fusiello, V Roberto, E Trucco. Symmetric stereo with multiple windowing [J].Journal of Pattern Recognition and Artificial Intelligence,2000,14(8):1053–1066.
[52] J Jeon, C Kim, Y Sung Ho. Sharp and dense disparity maps using multiple windows,IEEE Pacific Rim Conference on Multimedia,2002[C]. pp.1057-1064.
[53] A Fusiello, Roberto V, Trucco E. Efficient Stereo with Multiple Windowing, IEEEComputer Society Conference on Computer Vision and Pattern Recognition,1997[C]. pp.858-863.
[54] D Scharsterein, R Szeliski. A taxonomy and evaluation of dense two-frame stereocorrespondence algorithms. Proceedings IEEE Workshop on Stereo andMulti-Baseline Vision, Dec.9-10,2001[C]. Kauai, HI.
[55] M J Hannah. Computer Matching of Areas in Stereo Images [D]. Stanford University,1974.
[56] P Anandan. A computational framework and an algorithm for the measurement ofvisual motion [J]. International Journal of Computer Vision,1989,2(3):283–310.
[57] L Matthies, R Szeliski, T Kanade. Kalman filter-based algorithms for estimatingdepth from image sequences [J]. International Journal of Computer Vision,1989,3:209-236.
[58] E P Simoncelli, E H Adelson, D J Heeger. Probability distributions of optical flow.In proceeding of Computer Vision and Patteren Recognition. June,1991[C]. pp.310-315. Hawaii.
[59] T Kanade. Development of a video-rate stereo machine, In Image UnderstandingWorkshop,1994[C]. pp.549-557. Monterey, CA. Morgan Kaufmann Publishers.
[60] Fu He, F Da. Stereo matching based on dissimilar intensity support, ChineseConference on Pattern Recognition, Oct.21-23,2010[C]. pp.1-5. Chongqing, China.
[61] Crouzil A, IRIT-UPS Toulouse France, Massip-Pailhes L, Castan S. A NewCorrelation Criterion Based on Gradient Fields Similarity, Proceedings of the13thInternational Conference on Pattern Recognition, Aug25-29,1996[C]. vol.1, pp.632-636. Vienna.
[62] Delaunoy Ama l, Prados Emmanuel, Gradient Flows for Optimizing TriangularMesh-based Surfaces: Applications to3D Reconstruction Problems Dealing withVisibility [J], International Journal of Computer Vision, November,2011,95(2):100-123.
[63] Leonardo De-Maeztu, Arantxa Villanueva, Rafael Cabeza, Stereo matching usinggradient similarity and locally adaptive support-weight [J], Pattern RecognitionLetters, October,2011,32(13):1643–1651.
[64] Simon Hermann, Tobi Vaudrey.The Gradient-A Powerful and Robust CostFunction for Stereo Matching,25th International Conference of Image and VisionComputing New Zealand, Nov.8-9,2010[C]. pp.1-8. Queenstown.
[65] D Scharstein. Matching images by comparing their gradient fields. Conference A:Computer Vision& Image Processing, Proceedings of the12th IAPRInternational Conference on Pattern Recognition, Oct.9-13,1994[C]. vol.1, pp.572–575. Jerusalem.
[66] R Zabih, J Woodfill. Non-parametric local transforms for computing visualcorrespondence, Third European Conference on Computer Vision Stockholm, May2–6,1994[C]. vol.II, pp.151-158. Sweden.
[67] Banks J，Bennamout M. Reliablity analysis of the rank transform for stereomatching [J]. IEEE Transactions on Systems, Man, and Cybernetics-Prat B：Cybernetics,2001,31(6).
[68] Banks J, Bennamoun M, Corke P. Non-parametric techniques for fast and robuststereo matching, In IEEE TENCON’97Speech and Image Technologies forComputing and Telecommunications Proceedings, Dec.4-4,1997[C]. pp.365-368.Brisbane, Australia.
[69]马璇,朱世强,赖小波. RANK变换在立体匹配中的应用研究[J],系统仿真学报,2007,19(9).
[70] R Bolles, H Baker, M Hannah. The JISCT stereo evaluation, Image UnderstandingWorkshop, pp.263-274,1993[C].
[71]卢思军,唐振民.彩色图像非参数变换立体匹配研究[J],计算机工程与应用,2009,45(2).
[72] Z Gu, X Su, Y Liu, Q Zhang. Local stereo matching with adaptive support-weight,rank transform, and disparity calibration [J]. Pattern Recognition Letter,2008,29:1230-1235.
[73]周龙,徐贵力,李开宇,王彪,田裕鹏,陈欣.基于Census变换和改进自适应窗口的立体匹配算法[J].航空学报,2012,第5期.
[74] S Birchfield, C Tomasi. Depth discontinuities by pixel-to-pixel stereo [J].International Journal of Computer Vision,1998,35(3):1073-1080.
[75] R D Arnold. Automated stereo perception. Technical Report AIM-351, ArtificialIntelligence Laboratory, Stanford University,1983.
[76] A F Bobick, S S Intille. Large occlusion stereo [J]. International Journal ofComputer Vision,1999,33(3):181-200.
[77] W E L Grimson. Computational experiments with a feature based stereo algorithm[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1985,7(1):17–34.
[78] S B Pollard, J E W Mayhew, J P Frisby. PMF:A stereo correspondence algorithmusing a disparity gradient limit [J]. Perception,1985,14(4):449-470.
[79] K Prazdny. Detection of binocular disparities [J]. Biological Cybernetics,1985,52(2):93-99.
[80] Sch fer R. Review and future directions for3D-video, Proceedings of the PictureCoding,2006[C]. pp.1-11. Beijing, China.
[81] Kang S B. A survey of image-based rendering techniques [J]. VideoMetrics,1999:2-16.
[82] Heung Yeung Shum, Sing Bing Kang. A review of image-based renderingtechniques, Proceedings of IEEE/SPIE Visual Communications and ImageProcessing, May30,2000[C]. pp.2-13.
[83] Luat Do, Svitlana Zinger, de With P H N. Objective quality analysis forfree-viewpoint DIBR, IEEE International Conference on Image Processing,September26-29,2010[C]. pp.2629-2632. HongKong.
[84] Daniel Berj ó n, Alexander Hornung, Francisco Mor á n, Aljoscha Smolic.Evaluation of backward mapping DIBR for FVV applications[C]. IEEEInternational Conference on Multimedia and Expo, Madrid, July11-15,2011. IEEE,2011:1-6.
[85] T W Ryan, R T Gray, B R Hunt. Prediction of correlation errors in stereo-pairimages [J]. Optical Engineering,1980,19(3):312–322.
[86] B D Lucas, T Kanade. An iterative image registration technique with an applicationin stereo vision. In Seventh International Joint Conference on Artificial Intelligence(IJCAI-81), pp.674–679,1981[C]. Vancouver.
[87] Q Tian, M N Huhns. Algorithms for subpixel registration [J]. Computer Vision,Graphics, and Image Processing,1986,35(2):220-233.
[88] L Matthies, R Szeliski, T Kanade. Kalman filter-based algorithms for estimatingdepth from image sequences [J]. International Journal of Computer Vision,1989,3(3):209-236.
[89] S B Kang, R Szeliski, J Chai. Handling occlusions in dense multi-view stereo.Proceedings of the2001IEEE Computer Society Conference on Computer Visionand Pattern Recognition,2001[C].
[90] S Birchfield, C Tomasi. A pixel dissimilarity measure that is insensitive to imagesampling [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(4):401–406.
[91] D Scharstein. View Synthesis Using Stereo Vision, Lecture Notes in ComputerScience.1999, vol.1583. Springer-Verlag.
[92] Xiaofeng Ding, Lizhong Xu, Huibin Wang, Xin Wang, Guofang Lv. Stereo depthestimation under different camera calibration and alignment errors [J]. AppliedOptics,2011,50(10):1289-1301.
[93]方恒,吴怀宇.基于MRF和颜色空间的立体图像匹配算法[J].计算机技术与发展,2008,18(12):28-30.
[94]何华君,卢朝阳,焦卫东,郭大波.基于梯度和MRF模型的视差估计算法[J].西安电子科技大学学报,2007,34(3):373-376.
[95]郑肇葆,黄桂兰.航空影像纹理分类的最小二乘法和问题的分析[J].测绘学报，1996,25(2).
[96]张蓓蕾，基于马尔可夫场理论的单目图像深度估计研究[D].东华大学,2010.
[97]郑肇葆，马尔柯夫场在影像纹理模拟中的应用[J].测绘学报,1993,22(4).
[98] S Geman, D Geman. Stochastic Relaxation, Gibbs Distributions,and the BayesianRestoration of Images [J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1984,6(6):721-741.
[99]谢琅,杨艳,曹阳,王立恒.一种改进的三步搜索块运动估计算法[J].武汉大学学报.2005,51(5):625-628.
[100]刘跃军,苏静.一种改进三步搜索算法的设计与实现[J].2008,29(4):62-64.
[101]林舒静.视频压缩中运动估计算法的研究[D].西南大学,2009.
[102] C Connolly, T Fleiss. A study of efficiency and accuracy in the transformation fromRGB to CIELAB color space. IEEE International Conference on Image Processing,1997[C], pp.1046-1048.
[103] Klaus A, Sormann M, Karner K. Segment-Based Stereo Matching Using BeliefPropagation and a Self-Adapting Dissimilarity Measure. In Proceedings of18thInternational Conference on Pattern Recognition, August20-24,2006[C]. vol.3, pp.15-18. Hong Kong, China.
[104] El Mahassani, E D. New Robust Matching Cost Functions for Stereo Vision. InProceedings of9th Biennial Conference of the Australian Pattern RecognitionSociety on Digital Image Computing Techniques and Applications,2007[C]. pp.144-150.
[105] S E Palmer. Vision Science: Photons to Phenomenology [M]. Bradford Books, MITPress,1999.
[106] D Marr. Vision. WH Freeman and Company [M].1982.
[107] S W Zucker, R A Hummel. Toward a Low-Level Description of Dot Clusters:Labeling Edge, Interior and Noise Points [J]. Computer Graphics and ImageProcessing,1979,9:213-233.
[108] N Ahuja, M Tuceryan. Extraction of Early Perceptual Structure in Dot Patterns:Integrating Region, Boundary, and Component Gestalt [J]. Computer Vision,Graphics, and Image Processing,1989,48:304-356.
[109] J Shi, J Malik. Normalized Cuts and Image Segmentation [J]. IEEE TransactionPattern Analysis and Machine Intelligence,2000,22(8):888-905.
[110] G Yingnan, Z Yan, C Hexin. Improved belief propagation based on RGB vectormeasure for stereo matching, in Proceedings of2011International Conference onWireless Communications and Signal Processing,2011[C]. pp.1–5. NanJing.
[111] F Tombari, S Mattoccia, L Di Stefano. Segmentation-based adaptive support foraccurate stereo correspondence, IEEE Pacific-Rim Symposium on Image and VideoTechnology,2007[C]:427-438.
[112] Q Yang, L Wang, N Ahuja. A constant-space belief propagation algorithm for stereomatching, IEEE Conference Computer Vision and Pattern Recognition,2010[C]:1458-1465.
[113] N Papadakis, V Caselles. Multi-label depth estimation for graph cuts stereoproblems [J]. Journal of Mathematical Imaging and Vision,2010,38(1):70-82.
[114] D Miyazaki, Y Matsushita, K Ikeuchi. Interactive shadow removal from a singleimage using hierarchical graph cut [J]. Transactions on Computer Vision andApplications,2010,2:235-252.
[115] W Hu, K Zhang, L Sun, J Li, Y Li, S Yang. Virtual support window foradaptive-weight stereo matching, IEEE Visual Communications and ImageProcessing, Nov.6-9,2011[C]. pp.1-4. Tainan.
[116] L Nalpantidis, A Gasteratos. Biologically and psychophysically inspired adaptivesupport weights algorithm for stereo correspondence [J]. Robotics and AutonomousSystems,2010,58(5):457-464.
[117] Ilkwon Park, Hyeran Byun. Depth map refinement using multiple patch-baseddepth image completion via local stereo warping [J]. Journal of OpticalEngineering,2010,49(7).
[118] R Gupta, S Y Cho. A correlation-based approach for real-time stereo matching.International Symposium on Visual Computing. Advances in Visual Computing-6thInternational Symposium, November29-December1,2010[C]. Las Vegas, NV,USA.
[119] C Rhemann, A Hosni, M Bleyer, C Rother, and M Gelautz. Fast cost-volumefiltering for visual correspondence and beyond,2011IEEE Conference on ComputerVision and Pattern Recognition,2011[C]. pp.3017-3024.
[120] Q Yang, C Engels, A Akbarzadeh. Near real-time stereo for weakly-textured scenes.British Machine Vision Conference,2008[C]:80-87.
[121] Z Wang, Z Zheng. A region based stereo matching algorithm using cooperativeoptimization, IEEE Conference on Computer Vision and Pattern Recognition,2008[C]:1-8.