摘要
随着科学技术的进步，数字图像序列已成为一种重要的信息载体并广泛应用于国防、工业生产、文化传媒、以及医疗诊断等领域。目标提取发现和提取目标在图像序列中的时间和空间分布，是图像序列分析中的核心科学问题之一，在科学研究和工程应用上都有着十分重要的意义。近年来，低帧频（low-frame-rate）图像序列（帧频≤5帧/秒）在移动成像、无线数据传输及存储容量受限等复杂场景中得到了广泛应用。在低帧频图像序列中，相邻帧时间间隔长、目标时空连贯性差、目标外观与尺度变化剧烈，给目标提取带来了新的挑战。
本文围绕低帧频图像序列，研究了目标提取的三个关键技术：目标检测、目标跟踪以及图像序列目标分割。目标检测是目标提取的起点和支撑；目标跟踪获得目标的时空连贯性；图像序列目标分割得到目标在序列中的时空区域。本文的主要研究成果如下：
一、提出了一种预筛选Hough森林目标检测算法。针对现有的Hough森林算法存在的随机抽取图像块样本利用效率低、无效样本干扰大的问题，提出了基于图像块表述质量的随机样本预筛选思路；利用量化灰度级空间相关直方图描述图像块样本的灰度与局部空间结构的统计特性，并结合二维熵建立图像块表述质量的度量，进而建立了预筛选Hough森林目标检测算法。多组数据集上的实验结果证明了所提出的预筛选机制能够降低随机森林的不确定性并提高Hough森林的目标检测性能。
二、在低帧频图像序列目标跟踪方面，针对目标大尺度变化的难题，提出了一种基于核的变尺度目标跟踪算法（SIKBOT）。首先提出了一种基于集合分析的目标相似性度量，进而结合均值漂移过程求解尺度维加权核密度函数的模值搜索问题。在目标跟踪迭代中，并行地使用尺度维与空间维上的两个均值漂移过程估计目标的尺度与位置，实现变尺度目标跟踪。与现有主流方法相比，该方法提高了对变尺度目标跟踪的能力。
三、在低帧频图像序列目标跟踪方面，针对目标形状不规则的难题，提出了一种基于目标形状的Epanechnikov核函数（形状核，shaped kernel）并用于本文提出的SIKBOT方法中，形成基于形状核的变尺度目标跟踪方法（SK+SIKBOT）。所提出的形状核可以避免背景噪声对目标建模的影响，而且严格满足均值漂移算法收敛的充分条件。多组目标跟踪实验证明SK+SIKBOT方法不但提高了目标跟踪的精度与鲁棒性，也提高了目标跟踪的效率。
四、针对低帧频图像序列目标分割中分割误差累积的问题，提出了一种时空Grab Cut算法。将Grab Cut图像分割算法的不完整标记与迭代式估计的思想推广到图像序列目标分割中，建立了目标/背景统计分布的帧间传递机制，有效克服了图像序列目标分割错误的累积问题。实验结果证明了所提出的时空Grab Cut算法的性能超过了目前主流的基于图割的图像序列目标分割算法。
With the development of scientific technology, image sequences become anextremely important data source and have been extensively applied in many fields, e.g.national defense, industry production, entertainment, media broadcast, and medicaldiagnosis, etc. Object extraction is to find and extract the space-time distribution objectsin image sequences. As a fundamental problem in image sequence analysis, objectextraction receives attentions from both scientists and engineers. Low-frame-rate imagesequences (frame rate≤5fps) have, for the past few years, been put into use in manycomplicated scenes, such as moving imaging platforms, wireless surveillance or limitedstorages. Low-frame-rate image sequences bring object extraction new challenges, e.g.,the long temporal interval between adjacent frames, weak spatiotemporal coherence,and intensive change in object appearance and scale.
In this thesis, we focus on three key techniques for object extraction inlow-frame-rate image sequence object detection, object tracking and image sequenceobject segmentation. Object detection is the start point for object extraction; objecttracking explores the spatiotemporal coherence of objects; object segmentation obtainsspatiotemporal regions occupied by the objects. The main contributions of thedissertation are summarized as follows:
1. To solve the problem of underutilization of the randomly extracted patches,we proposed a prescreening mechanism for the Hough forest based on therepresentation quality. The gray level spatial correlation histogram (GLSCH) wasintroduced and improved to characterize the randomly extracted patches. Then weemployed2D image entropy to measure the representation quality of the patches andconstructed the prescreening-based Hough forest. Extensive experiments on standarddatabase demonstrated the proposed pre-screening mechanism decreased the uncertaintyHough forest and improved the detection performance.
2. We developed a novel scale invariant kernel-based object trackingalgorithm (SIKBOT) for tracking fast scaling objects in low-frame-rate imagesequences. We first proposed a novel set analysis based object similarity measure andthen employed the mean shift procedure to estimate the object scale. During eachiteration in tracking, object scale and object position were simultaneously estimated bytwo mean shift procedures in parallel. Compared with state-of-the-art methods, theproposed SIKBOT method improved the performance for tracking fast scaling objects.
3. To accurately describe irregular-shaped objects during tracking, weproposed a new object-shape-based Epanechnikov kernel (shaped kernel, SK), whichwas then combined with the proposed SIKBOT algorithm to construct the shaped-kernelSIKBOT algorithm (SK+SIKBOT). The proposed shaped kernel can alleviate the influence of the background noise during object modeling. Moreover, the Epanechnikovprofile guarantees the strict convergence of the mean shift procedures. Extensiveexperiements demonstrated the proposed shaped kernel achieved improvements in bothaccuracy and efficiency.
4. To overcome the error accumulation problem in low-frame-rate imagesequence object segmentation, we proposed a novel spatiotemporal Grab Cut algorithm.The object/background distribution propagation mechanism was established by tracking.Then by introducing the concepts of incomplete labeling and iterative estimation of theGrab Cut, we effectively alleviated the problem of error accumulation. Experimentalresults demonstrated the proposed spatiotemporal outperformed the state-of-the-artGraph Cuts-based image sequence object segmentation algorithms.
引文
[1]王润生.图像理解[M].长沙:国防科技大学出版社,1995.
[2]孙即祥.现代模式识别[M].北京:高等教育出版社,2008.
[3]孙浩.运动成像平台近景视频运动目标检测技术研究[D].长沙:中国人民解放军国防科学技术大学,2011.
[4]王威.基于图像的视频事件分析方法[D].长沙:中国人民解放军国防科学与技术大学,2009.
[5] X. Song, M. Cheng, B. Wang et al. Computer-Aided Preoperative Planning forLiver Surgery Based on Ct Images[J]. Procedia Engineering,2011,24(0):133-137.
[6] G. Funka-Lea, Y. Boykov, C. Florin et al. Automatic Heart Isolation for CtCoronary Visualization Using Graph-Cuts[C]. IEEE,614-617.
[7] A. Yilmaz, O. Javed, and M. Shah. Object Tracking: A Survey[J]. ACM Comput.Surv,2006,38(4)45.
[8] X. Bai, J. Wang, D. Simons et al. Video Snapcut: Robust Video Object CutoutUsing Localized Classifiers[J]. Acm Transactions on Graphics,2009,28(3).
[9] C. Y. Chung, and H. H. Chen. Video Object Extraction Via Mrf-Based ContourTracking[J]. IEEE Transactions on Circuits and Systems for Video Technology,2010,20(1):149-155.
[10] M. P. Kumar, P. H. S. Torr, and A. Zisserman. Obj Cut[C]. IEEE Computer Soc,18-25.
[11] J. M. Morel, and G. S. Yu. Asift: A New Framework for Fully Affine InvariantImage Comparison[J]. SIAM Journal on Imaging Sciences,2009,2(2):438-469.
[12] K. Rematas, and B. Leibe. Efficient Object Detection and Segmentation with aCascaded Hough Forest Ism[C]. IEEE,
[13] J. Gall, A. Yao, N. Razavi et al. Hough Forests for Object Detection, Tracking,and Action Recognition[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,2011,33(11):2188-2202.
[14] S. Maji, and J. Malik. Object Detection Using a Max Margin HoughTransform[C]. IEEE,1038-1045.
[15] P. F. Felzenszwalb, R. B. Girshick, D. McAllester et al. Object Detection withDiscriminatively Trained Part-Based Models[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2010,32(9):1627-1645.
[16] O. Barinova, V. Lempitsky, and P. Kohli. On Detection of Multiple ObjectInstances Using Hough Transforms[C]. IEEE Computer Soc,2233-2240.
[17] S. Schulter, C. Leistner, P. M. Roth et al. On-Line Hough Forests[C]. BMVAPress,128.1-128.11.
[18] X. Hou, and L. Zhang. Saliency Detection: A Spectral Residual Approach[C].New York: IEEE,2007:2280-2287.
[19] Z. Zivkovic. Improved Adaptive Gaussian Mixture Model for BackgroundSubtraction[C]. New York: IEEE,2008:28-31.
[20] P. Viola, and M. J. Jones. Robust Real-Time Face Detection[J]. InternationalJournal of Computer Vision,2004,57(2):137-154.
[21] P. Viola, and M. Jones. Rapid Object Detection Using a Boosted Cascade ofSimple Features[C]. New York: IEEE,2001:511-518.
[22] B. Babenko, M. H. Yang, and S. Belongie. Robust Object Tracking with OnlineMultiple Instance Learning[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2011,33(8):1619-1632.
[23] D. Comaniciu, V. Ramesh, and P. Meer. Kernel-Based Object Tracking[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2003,25(5):564-577.
[24] J. S. Hu, C. W. Juan, and J. J. Wang. A Spatial-Color Mean-Shift ObjectTracking Algorithm with Scale and Orientation Estimation[J]. PatternRecognition Letters,2008,29(16):2165-2173.
[25] Z. Kalal, J. Matas, and K. Mikolajczyk. P-N Learning: Bootstrapping BinaryClassifiers by Structural Constraints[C]. New York: IEEE,2010:49-56.
[26] J. Kwon, and K. M. Lee. Tracking by Sampling Trackers[C]. NEW YORK:IEEE,2011:1195-1202.
[27] J. Ning, L. Zhang, D. Zhang et al. Scale and Orientation Adaptive Mean ShiftTracking[J]. IET Computer Vision,2012,6(1):52-61.
[28] S. Stalder, H. Grabner, and L. Van Gool. Cascaded Confidence Filtering forImproved Tracking-by-Detection[C]. BERLIN: SPRINGER-VERLAG,2010:369-382.
[29] N. Vaswani, Y. Rathi, A. Yezzi et al. Deform Pf-Mt: Particle Filter with ModeTracker for Tracking Nonaffine Contour Deformations[J]. IEEE Transactions onImage Processing,2010,19(4):841-857.
[30] S. Wang, H. Lu, F. Yang et al. Superpixel Tracking[C]. New York: IEEE,2001:1323-1330.
[31] A. Yilmaz. Object Tracking by Asymmetric Kernel Mean Shift with AutomaticScale and Orientation Selection[C].140-145.
[32] A. Yilmaz. Kernel-Based Object Tracking Using Asymmetric Kernels withAdaptive Scale and Orientation Selection[J]. Machine Vision and Applications,2011,22(2):255-268.
[33] S. Beucher. The Watershed Transformation Applied to Image Segmentation[J].Scanning Microscopy International,1991,6299--314.
[34] L. Vincent, and P. Soille. Watersheds in Digital Spaces: An Efficient AlgorithmBased on Immersion Simulations[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,1991,13(6):583-598.
[35] J. B. Shi, and J. Malik. Normalized Cuts and Image Segmentation[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2000,22(8):888-905.
[36] Y. Y. Boykov, and M. P. Jolly. Interactive Graph Cuts for Optimal Boundary&Region Segmentation of Objects in N-D Images[C].105-112.
[37] C. Rother, V. Kolmogorov, and A. Blake."Grabcut"-Interactive ForegroundExtraction Using Iterated Graph Cuts[J]. Acm Transactions on Graphics,2004,23(3):309-314.
[38] P. F. Felzenszwalb, and D. P. Huttenlocher. Efficient Graph-Based ImageSegmentation[J]. International Journal of Computer Vision,2004,59(2):167-181.
[39] C. M. Li, C. Y. Xu, C. Gui et al. Level Set Evolution without Re-Initialization:A New Variational Formulation[C]. IEEE,430-436.
[40] T. F. Chan, and L. A. Vese. A Level Set Algorithm for Minimizing theMumford-Shah Functional in Image Processing[C]. LOS ALAMITOS: IEEECOMPUTER SOC,2001:161-168.
[41] L. A. Vese, and T. F. Chan. A Multiphase Level Set Framework for ImageSegmentation Using the Mumford and Shah Model[J]. International Journal ofComputer Vision,2002,50(3):271-293.
[42] V. Caselles, F. Catte, T. Coil et al. A Geometric Model for Active Contours inImage Processing[J]. Numedsche Mathematik,199333.
[43] M. Kass, A. Witkin, and D. Terzopoulos. Snake：Active Contour Models[J].International Journal of Computer Vision,1988,1, No.211.
[44] A. Wedel, T. Schoenemann, T. Brox et al. Warpcut-Fast ObstacleSegmentation in Monocular Video[J]. Pattern Recognition, Proceedings,2007,4713264-273.
[45] S. Kwak, W. Nam, B. Han et al. Learning Occlusion with Likelihoods for VisualTracking[C]. New York: IEEE,2011:1551-1558.
[46] A. Humayun, O. Mac Aodha, and G. J. Brostow. Learning to Find OcclusionRegions[C]. New York: IEEE,2011:
[47] N. Apostoloff, and A. Fitzgibbon. Learning Spatiotemporal T-Junctions forOcclusion Detection[C]. New York: IEEE,2005:553-559.
[48] V. P. Namboodiri, A. Ghorawat, and S. Chaudhuri. Improved Kernel-BasedObject Tracking under Occluded Scenarios[C]. BERLIN: SPRINGER-VERLAG,2006:504-515.
[49]吴磊.视觉语言分析:从底层视觉特征表达到语义距离学习[D].合肥:中国科学技术大学,2010.
[50] H. Bay, A. Ess, T. Tuytelaars et al. Speeded-up Robust Features (Surf)[J].Computer Vision and Image Understanding,2008,110(3):346-359.
[51]李安平.复杂环境下的视频目标跟踪算法研究[D].上海:上海交通大学,2006.
[52] G. Bradski. Computer Vision Face Tracking for Use in a Perceptual UserInterface[J]. Intel Technology Journal,1998,2(Q2):1-15.
[53] G. R. Bradski. Real Time Face and Object Tracking as a Component of aPerceptual User Interface[C].214-219.
[54] Y. Ke, R. Sukthankar, and M. Hebert. Event Detection in Crowded Videos[C].New York: IEEE,2007:8.
[55] F. Porikli, and O. Tuzel. Object Tracking in Low-Frame-Rate Video[C].Bellingham: Spie-Int Soc Optical Engineering,2005:72-79.
[56] P. Agouris, A. Stefanidis, S. Gyftakis et al. Differential Object ExtractionMethods for Automated Gis Updates[J]. Int. Archives of Photogrammetry&Remote Sensing,2002,34(4):781-785.
[57] V. Bucha, and S. Ablameyko. Interactive Objects Extraction from RemoteSensing Images[J]. NATO Science for Peace and Security Series C:Environmental Security,2007225-238.
[58] S. Y. Chen, and J. W. Hsieh. Boosted Road Sign Detection and Recognition[C].NEW YORK: IEEE,2008:3823-3826.
[59] S. Hinz. Automatic Object Extraction for Change Detection and Gis Update[J].The International Archives of the Photogrammetry, Remote Sensing and SpatialInformation Sciences,2008, XXXVII(B4):277-284.
[60] S. Hinz, and A. Baumgartner. Automatic Extraction of Urban Road Networksfrom Multi-View Aerial Imagery[J]. ISPRS Journal of Photogrammetry andRemote Sensing,2003,58(1-2):83-98.
[61] S. H. Hsu, and C. L. Huang. Road Sign Detection and Recognition UsingMatching Pursuit Method[J]. Image and Vision Computing,2001,19(3):119-129.
[62]宋晓.肝脏ct图像自动分割与虚拟手术切除技术研究[D].厦门:厦门大学,2012.
[63]庄玲.基于三维图像序列的左心室分割与运动联合估计[D].杭州:浙江大学,2008.
[64] N. Sebe, A. W. M. Smenlders, and M. S. Leww. Video Retrieval andSummarization[J]. Computer Vision and Image Understanding,2003,92(2-3):141-146.
[65] J. Philbin, O. Chum, M. Isard et al. Object Retrieval with Large Vocabulariesand Fast Spatial Matching[C].1545-1552.
[66] J. Sivic, and A. Zisserman. Efficient Visual Search of Videos Cast as TextRetrieval[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(4):591-606.
[67] Y. Li, J. Sun, and H. Y. Shum. Video Object Cut and Paste[J]. AcmTransactions on Graphics,2005,24(3):595-600.
[68] J. Wang, P. Bhat, R. A. Colburn et al. Interactive Video Cutout[J]. AcmTransactions on Graphics,2005,24(3):585-594.
[69] B. L. Price, B. S. Morse, and S. Cohen. Livecut: Learning-Based InteractiveVideo Segmentation by Evaluation of Multiple Propagated Cues[C]. New York:IEEE,2009:779-786.
[70] M. Stamm, and K. J. R. Liu. Live Video Object Tracking and SegmentationUsing Graph Cuts[C]. New York: IEEE,2008:1576-1579.
[71] H. L. Li, K. N. Ngan, and Q. Liu. Faceseg: Automatic Face Segmentation forReal-Time Video[J]. IEEE Transactions on Multimedia,2009,11(1):77-88.
[72] F. Porikli, F. Bashir, and S. Huifang. Compressed Domain Video ObjectSegmentation[J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2010,20(1):2-14.
[73] N. P. Cuntoor, A. Basharat, A. G. A. Perera et al. Track Initialization in LowFrame Rate and Low Resolution Videos[C]. Piscataway, NJ Institute ofElectrical and Electronics Engineers Inc.,2010:3640-3644.
[74] Y. Li, H. Ai, T. Yamashita et al. Tracking in Low Frame Rate Video: A CascadeParticle Filter with Discriminative Observers of Different Lifespans[C]. NewYork: IEEE,2007:1752-1759.
[75] K. Palaniappan, F. Bunyak, P. Kumar et al. Efficient Feature Extraction andLikelihood Fusion for Vehicle Tracking in Low Frame Rate Airborne Video[C].Piscataway, NJ IEEE Computer Society,1-8.
[76] Y. Li, H. Ai, T. Yamashita et al. Tracking in Low Frame Rate Video: A CascadeParticle Filter with Discriminative Observers of Different Life Spans[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2008,30(10):1728-1740.
[77] A. Mori, Y. Makihara, and Y. Yagi. Gait Recognition Using Period-Based PhaseSynchronization for Low Frame-Rate Videos[C]. Piscataway, NJ Institute ofElectrical and Electronics Engineers Inc.,2010:2194-2197.
[78] X. Jiangjian, C. Hui, H. Sawhney et al. Vehicle Detection and Tracking in WideField-of-View Aerial Video[C]. New York: IEEE,2010:679-684.
[79] P. Li, and C. Wang. Object of Interest Extraction in Low-Frame-Rate ImageSequences and Application to Mobile Mapping Systems[J]. Optical Engineering,2012,51(6):06720101-06720112.
[80] N. Dalal, and B. Triggs. Histograms of Oriented Gradients for HumanDetection[C]. IEEE Computer Soc,886-893.
[81] B. Alexe, T. Deselaers, and V. Ferrari. What Is an Object?[C]. IEEE,73-80.
[82] C. X. Dai, Y. F. Zheng, and X. Li. Pedestrian Detection and Tracking in InfraredImagery Using Shape and Appearance[J]. Computer Vision and ImageUnderstanding,2007,106(2-3):288-299.
[83] H. Sun, C. Wang, B. L. Wang et al. Pyramid Binary Pattern Features forReal-Time Pedestrian Detection from Infrared Videos[J]. Neurocomputing,2011,74(5):797-804.
[84] J. Rabin, J. Delon, and Y. Gousseau. A Statistical Approach to the Matching ofLocal Features[J]. SIAM Journal on Imaging Sciences,2009,2(3):931-958.
[85] J. C. Niebles, B. Y. Han, F. F. Li et al. Efficient Extraction of Human MotionVolumes by Tracking[C].655-662.
[86] J. Gall, and V. Lempitsky. Class-Specific Hough Forests for ObjectDetection[C].1022-1029.
[87] J. Shotton, A. Blake, and R. Cipolla. Multiscale Categorical Object RecognitionUsing Contour Fragments[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2008,30(7):1270-1281.
[88] B. Leibe, A. Leonardis, and B. Schiele. Robust Object Detection withInterleaved Categorization and Segmentation[J]. International Journal ofComputer Vision,2008,77(1-3):259-289.
[89] M. J. Swain, and D. H. Ballard. Color Indexing[J]. International Journal ofComputer Vision,1991,7(1):11-32.
[90] D. G. Lowe. Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[91] S. J. Roberts, D. Husmeier, I. Rezek et al. Bayesian Approaches to GaussianMixture Modeling[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1998,20(11):1133-1142.
[92] J. Matas, O. Chum, M. Urban et al. Robust Wide-Baseline Stereo fromMaximally Stable Extremal Regions[J]. Image and Vision Computing,2004,22(10):761-767.
[93] C. J. C. Burges. A Tutorial on Support Vector Machines for PatternRecognition[J]. Data Mining and Knowledge Discovery,1998,2(2):121-167.
[94] J. Sivic, F. Schaffalitzky, and A. Zisserman. Object Level Grouping for VideoShots[J]. International Journal of Computer Vision,2006,67(2):189-210.
[95] C. M. Bishop. Pattern Recognition and Machine Learning (Information Scienceand Statistics)[M]: Springer-Verlag New York, Inc.,2006.
[96] L. Breiman. Random Forests[J]. Machine Learning,2001,45(1):5-32.
[97] V. Kumar, and I. Patras. A Discriminative Voting Scheme for Object DetectionUsing Hough Forests[C]. BMVA Press,3.1-3.10.
[98] C. H. Lampert, M. B. Blaschko, and T. Hofmann. Efficient Subwindow Search:A Branch and Bound Framework for Object Localization[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2009,31(12):2129-2142.
[99] C. H. Lampert, M. B. Blaschko, and T. Hofmann. Beyond Sliding Windows:Object Localization by Efficient Subwindow Search[C]. New York: IEEE,2008:1897-1904.
[100] D. H. Ballard. Generalizing the Hough Transform to Detect Arbitrary Shapes[J].Pattern Recognition,1981,13(2):111-122.
[101] M. Andriluka, S. Roth, and B. Schiele. People-Tracking-by-Detection andPeople-Detection-by-Tracking[C]. New York: IEEE,2008:1873-1880.
[102] M. Isard, and A. Blake. Condensation-Conditional Density Propagation forVisual Tracking[J]. International Journal of Computer Vision,1998,29(1):5-28.
[103] C. Bibby, and I. Reid. Robust Real-Time Visual Tracking Using Pixel-WisePosteriors[C]. BERLIN: SPRINGER-VERLAG BERLIN,831-844.
[104] P. M. Djuric, J. H. Kotecha, J. Q. Zhang et al. Particle Filtering[J]. IEEE SignalProcessing Magazine,2003,20(5):19-38.
[105] F. Gustafsson, F. Gunnarsson, N. Bergman et al. Particle Filters for Positioning,Navigation, and Tracking[J]. IEEE Transactions on Signal Processing,2002,50(2):425-437.
[106] J. L. Barron, D. J. Fleet, and S. S. Beauchemin. Performance of Optical-FlowTechniques[J]. International Journal of Computer Vision,1994,12(1):43-77.
[107] B. K. P. Horn, and B. G. Schunck. Determining Optical-Flow[J]. ArtificialIntelligence,1981,17(1-3):185-203.
[108] A. Yilmaz, X. Li, and M. Shah. Contour-Based Object Tracking with OcclusionHandling in Video Acquired Using Mobile Cameras[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,2004,261531-1536.
[109] R. T. Collins. Mean-Shift Blob-Tracking through Scale Space[C].234-240.
[110] J. G. Allen, R. Y. D. Xu, and J. S. Jin. Object Tracking Using CamshiftAlgorithm and Multiple Quantized Feature Spaces[C].3-7.
[111] N. S. Peng, J. Yang, and Z. Liu. Mean Shift Blob Tracking with KernelHistogram Filtering and Hypothesis Testing[J]. Pattern Recognition Letters,2005,26(5):605-614.
[112] K. M. Yi, H. S. Ahn, and J. Y. Choi. Orientation and Scale Invariant Mean ShiftUsing Object Mask-Based Kernel[C]. New York: IEEE,3121-3124.
[113] B. Babenko, M. H. Yang, and S. Belongie. Visual Tracking with OnlineMultiple Instance Learning[C].983-990.
[114] Z. Kalal, K. Mikolajczyk, and J. Matas. Tracking-Learning-Detection[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2012,34(7):1409-1422.
[115] X. Ren, and J. Malik. Tacking as Repeated Figure/Ground Segmentation[C].New York:2007:1582-1589.
[116] S. J. Sun, D. R. Haynor, and Y. Kim. Semiautomatic Video Object SegmentationUsing Vsnakes[J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2003,13(1):75-82.
[117] V. Kolmogorov, A. Criminisi, A. Blake et al. Bi-Layer Segmentation ofBinocular Stereo Video[C]. LOS ALAMITOS, CA: IEEE COMPUTER SOC,2005:407-414.
[118] A. Criminisi, G. Cross, A. Blake et al. Bilayer Segmentation of Live Video[C].53-60.
[119] P. Yin, A. Criminisi, J. Winn et al. Tree-Based Classifiers for Bilayer VideoSegmentation[C]. New York: IEEE,2007:295-302.
[120] H. Yuchi, L. Qingshan, and D. Metaxas. Video Object Segmentation byHypergraph Cut[C]. IEEE,1738-1745.
[121] J. Y. A. Wang, and E. H. Adelson. Spatio-Temporal Segmentation of VideoData[C]. M.I.T. Media Laboratory Vision and Modeling Group,12.
[122] F. Zhong, X. Y. Qin, and Q. S. Peng. Transductive Segmentation of Live Videowith Non-Stationary Background[C]. New York: IEEE Computer Soc,2189-2196.
[123] N. Razavi, J. Gall, and L. Van Gool. Backprojection Revisited: ScalableMulti-View Object Detection and Similarity Metrics for Detections[C].Springer-Verlag Berlin,620-633.
[124] J. Lezama, K. Alahari, J. Sivic et al. Track to the Future: Spatio-Temporal VideoSegmentation with Long-Range Motion Cues[C].
[125] N. Akae, A. Mansur, Y. Makihara et al. Video from Nearly Still: An Applicationto Low Frame-Rate Gait Recognition[C]. New York: IEEE,2012:1537-1543.
[126] E. Shechtman, Y. Caspi, and M. Irani. Space-Time Super-Resolution[J]. IEEETrans. Pattern Anal. Mach. Intell.,2005,27(4):531-545.
[127] A. K. Jain, R. P. W. Duin, and J. C. Mao. Statistical Pattern Recognition: AReview[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(1):4-37.
[128] K. Mikolajczyk, T. Tuytelaars, C. Schmid et al. A Comparison of Affine RegionDetectors[J]. International Journal of Computer Vision,2005,65(1-2):43-72.
[129] T. Tuytelaars, and K. Mikolajczyk. Local Invariant Feature Detectors: ASurvey[J]. Found. Trends. Comput. Graph. Vis.,2008,3(3):177-280.
[130] C. S. Won, D. K. Park, and S. J. Park. Efficient Use of Mpeg-7Edge HistogramDescriptor[J]. Etri Journal,2002,24(1):23-30.
[131]冈萨雷斯.数字图像处理[M].北京:电子工业出版社,2006.
[132] A. Rosenfeld, and E. Johnston. Angle Detection on Digital Curves[J]. IEEETrans. Comput.,1973,22(9):875-878.
[133]徐玲.基于图像轮廓的角点检测方法研究[D].重庆:重庆大学,2009.
[134] T. Lindeberg. Feature Detection with Automatic Scale Selection[J]. InternationalJournal of Computer Vision,1998,30(2):79-116.
[135] M. S. Kumari, and B. H. Shekar. Color-Sift Model: A Robust and an AccurateShot Boundary Detection Algorithm[C]. SPIE,
[136] Y. Ke, and R. Sukthankar. Pca-Sift: A More Distinctive Representation forLocal Image Descriptors[C]. New York:2004:506-513.
[137] H. Ming-Kuei. Visual Pattern Recognition by Moment Invariants[J]. InformationTheory, IRE Transactions on,1962,8(2):179-187.
[138] T. Ojala, M. Pietikainen, and T. Maenpaa. Multiresolution Gray-Scale andRotation Invariant Texture Classification with Local Binary Patterns[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(7):971-987.
[139] R. O. Duda, P. E. Hart, and D. G. Stork. Pattern Classification (2nd Edition)[M]:Wiley-Interscience,2000.
[140] Z. Lei, T. Fang, H. Huo et al. Rotation-Invariant Object Detection of RemotelySensed Images Based on Texton Forest and Hough Voting[J]. IEEETransactions on Geoscience and Remote Sensing,2012,50(4):1206-1217.
[141] V. Ferrari, L. Fevrier, F. Jurie et al. Groups of Adjacent Contour Segments forObject Detection[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,2008,30(1):36-51.
[142] P. F. Felzenszwalb, and D. P. Huttenlocher. Pictorial Structures for ObjectRecognition[J]. International Journal of Computer Vision,2005,61(1):55-79.
[143] D. Hoiem, C. Rother, and J. Winn.3d Layoutcrf for Multi-View Object ClassRecognition and Segmentation[C]. New York: IEEE,2007:580-587.
[144]陈进.高光谱图像分类方法研究[D].长沙:中国人民解放军国防科学技术大学,2010.
[145]雷震.随机森林及其在遥感影像处理中应用研究[D].上海:上海交通大学,2012.
[146] Y. Freund, and R. E. Schapire. A Decision-Theoretic Generalization of on-LineLearning and an Application to Boosting[J]. Journal of Computer and SystemSciences,1997,55(1):119-139.
[147]罗大鹏.基于在线学习理论的目标检测技术[D].武汉:华中科技大学,2010.
[148] D. Comaniciu, and P. Meer. Mean Shift: A Robust Approach toward FeatureSpace Analysis[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,2002,24(5):603-619.
[149] B. Leibe, and B. Schiele. Scale-Invariant Object Categorization Using aScale-Adaptive Mean-Shift Search[J]. Lecture Notes in Computer Science,2004,3175145-153.
[150] M. Godec, P. M. Roth, and H. Bischof. Hough-Based Tracking of Non-RigidObjects[C]. New York: IEEE,2011:81-88.
[151] Y. Xiao, Z. Cao, and T. Zhang. Entropic Thresholding Based on Gray-LevelSpatial Correlation Histogram[C]. IEEE,1742-1745.
[152] M. Hossny, S. Nahavandi, D. Creighton et al. Image Fusion Performance MetricBased on Mutual Information and Entropy Driven Quadtree Decomposition[J].Electronics Letters,2010,46(18):1266-U45.
[153] Z. Zivkovic, and B. Krose. An Em-Like Algorithm for Color-Histogram-BasedObject Tracking[C]. New York: IEEE,2004:798-803.
[154] Y. Z. Cheng. Mean Shift, Mode Seeking, and Clustering[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,1995,17(8):790-799.
[155] A. Vedaldi, and S. Soatto. Quick Shift and Kernel Methods for ModeSeeking[C]. BERLIN, GERMANY SPRINGER-VERLAG BERLIN,2008:705-718.
[156] Y. Boykov, and G. Funka-Lea. Graph Cuts and Efficient N-D ImageSegmentation[J]. International Journal of Computer Vision,2006,70(2):109-131.
[157]郁生阳.基于能量最小化图割的图像与视频目标精确分割研究[D].上海交通大学,2009.
[158] J. F. Ning, L. Zhang, D. Zhang et al. Interactive Image Segmentation byMaximal Similarity Based Region Merging[J]. Pattern Recognition,2010,43(2):445-456.
[159]钟平.面向图像标记的随机场模型研究[D].长沙:中国人民解放军国防科学技术大学,2008.
[160] D. M. Greig, B. T. Porteous, and A. H. Seheult. Exact Maximum a PosterioriEstimation for Binary Images[J]. Journal ofthe Royal Statistical Society. SeriesB (Methodological),1989,51(2):271-279.
[161] L. R. Ford, and D. R. Fulkerson. Maximal Flow through a Network[J]. CanadianJournal of Mathematics,1956,8399-404.
[162] A. V. Goldberg, and R. E. Tarjan. A New Approach to the Maximum FlowProblem[C]. NY, USA: ACM,1986:136-146.
[163]刘献如.视频图像序列目标跟踪算法及其应用研究[D].长沙:中南大学,2011.
[164]张鹏林.复杂场景视频序列图像运动物体提取方法研究[D].武汉:武汉大学,2005.
[165] G. Bradski, and A. Kaehler. Learning Opencv: Computer Vision with theOpencv Library[M]: O'Reilly,2008.