摘要
视频运动目标识别、跟踪是计算机视觉研究领域的重要课题之一,在视频监控、人机交互、机器人视觉导航以及智能交通控制中具有广泛的应用前景。而实现海上运动目标的自动检测、识别和跟踪对于统计海面上来往船只信息、防止船只碰撞以及海上安防都具有重要的意义。
本文主要研究海面运动船只的识别与跟踪技术。首先概述了海上运动目标检测和跟踪的研究现状;对目前主要的显著区域提取、运动目标识别和跟踪方法进行了简要概述;提出了基于视觉注意和HOG特征相融合的海上船只目标检测方法;利用多特征融合的粒子滤波算法对运动目标进行了跟踪。本文的主要研究工作和创新点包括以下几点:
1、对视觉注意力模型进行了深入研究,提出了基于改进的视觉注意力模型的船只显著区域检测方法。在获得的显著图的基础上,通过阈值分割和区域提取快速获得船只显著区域,为进一步的船只目标识别打下了良好的基础。
2、在视觉注意力模型和HOG特征的基础上,提出了视觉注意和HOG特征双向融合的船只目标检测机制。模拟人类视觉中有意识“主动寻找”与无意识“被动受吸引”相交互的视觉过程。通过视觉注意快速获得显著性船只区域,减小目标搜索区域;在显著性船只区域内,通过HOG特征和学习机制实现船只目标的精确识别。同时本文还提出在显著性船只区域内通过边缘检测和区域跟踪相结合来快速获得船只候选区域的方法。实验表明,这种双向融合机制对海上船只目标检测具有准确性高、处理速度快的特点,基本能够满足复杂场景下实时处理的要求。
3、对粒子滤波方法进行了深入研究并应用于海面船只跟踪系统中,通过HSV颜色直方图和形状特征相融合的方法,提高了跟踪系统的准确性和鲁棒性。在对多目标、目标交叉遮挡等情况下的跟踪也表现出了良好的性能。
Video moving targets detection、recognition and tracking are one of the most important issues in the field of computer vision. It has wide application prospects in many areas, such as video surveillance、human-computer interaction、robot visual navigation and intelligent traffic control. Implement of maritime objects detection、recognition and tracking is significant for statistics of vessel information、avoiding sea collision between vessels and coastal defense.
This paper mainly researches maritime ship identification and tracking technology. First, it describes current development status of maritime moving target detection and tracking research, and then outlines some main methods of salient region extraction, moving target identification and tracking, and we present a method of ship detection based on the visual attention and HOG feature. Finally we use the particle filter of multi-feature combining method to track the moving targets. The main research innovations and contributions are summarized as follows:
1. We take a deep research into the model of visual attention, and propose a method of salient ship region detection based on improved visual attention. Threshold division and region extraction are used to get salient ship region quickly based on saliency map to lay the foundation for ship identification.
2. Based on the attention model and HOG feature, we present a ship target detection mechanism of visual attention and HOG feature two-way integration, and simulate the process of interactive visual about the human visual awareness in the "active search" and unconscious "passive attracted by". We quickly get salient ship region through the visual attention, reduce the search region, and then achieve the precise recognition in the saliency ship region through the HOG feature and Learning mechanism. Simultaneously, we propose a method to get candidate ship location quickly by edge detection and region extraction. Experiment results show that this two-way integration mechanism of ship target detection has feature of high accuracy, fast processing speed, and it basically meet the requirements of real-time processing for complex scene.
3. We explore the particle filtering method and put it into the ship tracking system. We improve the accuracy and robustness of the tracking system through the combination of HSV color histogram with shape features. The system also shows a good performance in many situations such as multi-target tracking, target under cross-blocked, and so on.
引文
[1].WU Qi-ying;LI Cui-hua Novel method for moving maritime objects detection.2007,43(14):213-216.
[2].A.W.Smith,M.K.Teal,P.Voles.The statistical characterization of the sea for the segmentation of maritime images[C].Conference on Video/Image Processing and Multimedia Communications,2003,489--494.
[3].P.Voles,M.Teal,J.Sanderson.Target identification in a complex maritime scene[C].IEEE Colloquium on Motion Analysis and Tracking,London,UK:1999,15/1-15/4.
[4].J.G.Sanderson,M.K.Teal,T.J.Ellis.Characterisation of a Complex Maritime Scene using Fourier Space Analysis to Identify Small Craft[J].Image Processing And Its Applications,1999,250 - 254.
[5].任明武,曹雨龙,杨静宇等.复杂条件下的船舶目标检测的图象预处理[J].计算机工程,2000,10:68-70.
[6].M.W.Ren,Y.L.Cao,Z.M.Tang,etc.One effective method for ship recognition in ship locks[C].SPIE Conf.Signal Processing,Sensor Fusion and Target Recognition.vol.3720,1999,467-472.
[7].Yulong Cao,Jingyu Yang,Mingwu Ren.Online detecting ships in lock using optical flow method[C].Proc.SPIE Int.Soc.Opt.Eng.1999:759.
[8].周颢.复杂背景下的运动目标分割和识别的关键技术研究[D].上海:上海交通大学硕士学位论文,2004.
[9].王劲松.序列图像运动检测研究及不变矩方法在船只识别中的应用[D].合肥:中国科技大学硕士学位论文,2002.
[10].D.Noton and L.W.Stark,"Scanpaths in Eye Movements During Pattern Perception,"Science,vol.17,no.1,1971,P308-311.
[11].Claudio M.Privitera,Lawrence W.Stark,Algorithms for Defining Visual Regions-of-Interest:Comparison with Eye Fixations.IEEE Transactions On Patern Analysis and Machine Intelligence,2000.22(9):P970-982.
[12].李弼成,彭天强等.智能图像处理技术.北京:电子工业出版社,2004.07
[13].Ivar.Balslev,K.P.Dcring,R.D.Eriksen.Weighted central moments in Pattern Reognition[J].Pattenr Recongition Letters,2000,21:381-384.
[14].Reddy B.S,Chatterji B.N.An fft-based technique for translation,rotation and scale-invariant image registration[J].IEEE Transactionson Image Processing,1996,5(8):1266-1271.
[15].JiZhou,Jioaying shi.A robust algorithm for feature Point matching[J].Computers&GraPhics.2002,26:429-436.
[16].XioalongDai, SimakaKhorram.A feature-Based Image Registartion Algorithm Using Improved Chain-Code Rersentation Combined with Invariant Moments [C].IEEE Trnas.on Geoscience and Remote Sensing.1999,37(5):235-238.
[17].Mallat S.A .Theory for Multiersolution Signal Deeomposition:The Wavelet Representation [J].IEEE Trnas.PaRem Anal Mach Intell.1989,11(7):674-693.
[18].N.Matthews,P.An,D.Chamley,and C.Harris.Vehicle Detection and Recognition in Greyscale Imagery [J].Control Eng.Practice,vol.4,pp.473-479,1996.
[19].H.Schneiderman.A Statistical Approach to 3D Object Detection Applied to Faces and Cars [J].2000.CMU-RI-TR-00-06.
[20].Jaesik Choi.Realtime On-Road Vehicle Detection with Optical Flows and Haar-like feature detector[R].A final reprt of a course CS543 (Computer Vision,prof.Li Fei-Fei).2006.
[21].Z.Sun,G Bebis and R.Miller.On-road vehicle detection using Gabor filters and support vector machines [J].IEEE 14th Int.Conf.Digital Signal Processing,Greece,Jul.2002,pp.1019-1022.
[22].Z.Sun,G Bebis and R.Miller.Improving the performance of on-road vehicle detection by combining Gabor and wavelet features[C].IEEE 5th Int.Conf.Intelligent Transportation Systems,Singapore,Sep.2002,pp.130-135.
[23].T.Kato,Y.Ninomiya and I.Masaki.Preceding vehicle recognition based on learning from sample images [J].IEEE Trans.Intell.Transp.Syst,vol.3,pp.252-260,Dec.2002.
[24].MeyerF,Bouthemy P.Region-basedtraekingu;ingaffinemotionmodelsinlong imagesequenees.CVGIP:Imageunderstandir, 1994,60(2):119-140.
[25].BaseleB,DerieheR.Region tracking through image sequences.In:Proceedings Of IEEE International Conference of Computer Vision,1995,302-307.
[26].Aggarwal J K,Cai Q,Human motion analysis:A review.Computer Vision and Image Understanding,1999,73 (3):428-440.
[27].Paragios N,Deriche R.Geodesic active contours and level sets for the detection and tracking of moving object [J].IEEE Trans Pattern Analysis and Machine Intelligence.2000,22(3):266-280.
[28].Peterfreund N.Robust tracking of position and velocity with Kalman snakes [J].IEEE Trans Pattern Analysis and Machine Intelligence.2000,22(6):564-569.
[29].Isard M,Blake A.Contour tracking by stochastic propagation of conditional density[C].Proc European Conference on Computer Vision,Cambridge,1996,343-356.
[30].Vieren C,Cabestaing F,Postaire J.Catching Moving Object with Snake for Motion Tracking [J].Pattern Recognition Letters.1995, 16:679-985.
[31].W.J.Rucklidge.Efficiently Locating Objects Using Hausdorff Distance [J].Internation J of Computer Vision.1997,24(3):251 -270.
[32].J.Shi,C.Tomasi.Good Feature to Track[C].IEEE Conference on Computer Vision and Pattern Recognition.1994,593-600.
[33].L.Itti,C.Koch,E.Niebur.A model of saliency-based visual attention for rapid scene analysis [J].IEEE Transaction on Pattern Analysis and Machine Intelligence,1998,20(11):1254-1259.
[34].Lindeberg T.Detecting salient blob-like image structures and their scales with a scale-space primal sketch:a method for focus-of-attention[J].International Journal of Computer Vision,1993,11(3):283-318.
[35].YF Ma,HJ Zhang.Contrast-based image attention analysis by using fuzzy growing[C].Proc.of ACM Multimedia 2003,Berkeley,2003,374-381.
[36].龙甫荟,郑南宁.一种引入注意机制的视觉计算模型[J].中国图象图形学报,1998,3(7):592-595.
[37].张鹏,王润生.基于视点转移和视区追踪的图像显著区域检测[J].软件学报,2004,15(6):891-898.
[38].N.Ouerhani,H.Hugli.Computing visual attention from scene depth[C].Proc.ICPR 2000,IEEE Computer Society Press,2000,375-378.
[39].Y Sun,R.Fisher.Object-based visual attention for computer vision[J].Artificial Intelligence,2003,146(1):77-123.
[40].YF Ma,HJ Zhang.A model of motion attention for video skimming[C].Proc.of ICIP,2002,1:129-132.
[41].L.Itti,C.Gold,C.Koch.Visual attention and target detection in cluttered natural scenes[J].Optical Engineering,2001,40(9):1784-1793.
[42].H Zabrodsky,S Peleg.Attentive transmission[J].Journal of Visual Communication and Image Representation,1990,1(2):189-198
[43].叶聪颖.建筑目标识别中检测与分割技术的研究.厦门大学计算机系硕士毕业论文:55-56.2005.
[44].Ridler TW,Calvard S.Picture thresholding using an iterative selection method.IEEE-SMC,1978,9:630-632.
[45].Dalal N,Triggs B.Histograms of oriented gradients for human detection[A].In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C],San Diego,CA,USA,2005,1:886-893.
[46].Q.Zhu,S.Avidan,M- C.Yeh,and K-T.Cheng,"Fast Human Detection Using a Cascade of Histograms of Oriented Gradients," Proc.IEEE Intemational Conference.On Computer Vision and Pattem Recognition,2006
[47].Fporikli," Intesral histogram:A fast way to extract histograms in Cartesian spaces,"CVPR,2005.
[48].VIOLA P,JONES M,"Rapid Objeet Detection Using a Boosted Cascade of Simple Features",Proc.of International Conference on ComPuter Vision and Pattern Recognition,2001,1:511-518.
[49].Vapnik V.The Nature of Statistical Learning Theory[M].New York:Springer-Verlag.1995.
[50].Cortes C,Vapnik V.Support Vector Networks[J].Machine Learning.1995.
[51].张学工.关于统计学习理论与支持向量机[J].自动化学报,第26卷第1期.2000.
[52].Burges C J C.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery,2,121-167.1998.
[53].T.-F.Wu,C.-J.Lin,and R.C.Weng.Probability Estimates for Multi-class Classification by Pairwise Coupling[J].Journal of Machine Learning Research,2004.
[54].王惠锋,孙正兴.基于内容的图像检索中的语义处理方法[J].中国图像图形学报,6(10):945-952.2001
[55].Libsvm.http://www.csie.ntu.edu.tw/~cjlin/libsvm/
[56].Canny John.A computational approach to edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1986,8(6):679-698
[57].Milan Sonka,Vaclav Hlavac,Roger Boyle.Image Processing,Analysis,and Machine Vision [M].2nd ed.Beijing:People's Posts & Telecommunications Publishing House,2003(in Chinese)
[58].M Sanjeev ArulamPalam,Simon Maskell,Neil Gordon.A tutorial on Particle filters for Online nonlinear/non-Gaussian Bayesian tracking[J].IEEE Transactions on Signal Processing,2002,50(2):174-188.
[59].Xingrong Li and Jue Yang.Sequential Monte Carlo Methods for Collaborative Multi-Sensor Tracking.IEEE I-4244-1513-06/072007 1-6.
[60].Oliver Frank,Juan Nieto,Jose Guivant.Multiple Targets Tracking Using Sequential Monte Carlo Methods and Statistical Data Association.Proceedings of the 2003 IEEE/RSJ Inti.Conference on Intelligent Robots and Systems Las Vegas,Nevada.Oetober2003.2718-2723.
[61].Hue C,Le Cadre,J-P Perez P.Sequential Monte Carlo Methods for Multiple Target Tracking and Data Fusion.Signal Processing,IEEE Transactions on Volume 50,Issue 2,Feb,2002Page(s):309-325.
[62].D Crisan and A Doueet.A Survey of convergence results on Particle filtering methods.For Practitioners[J].IEEE Trans.Speech and Audio Proc.,2002,10(3):173-185.
[63].P.Brasnett,L.Mihaylova,D.Bull,et al.Sequential Monte Carlo tracking by fusing multiple cues in video sequences.Image and Vision Computing,2007,25:1217-1227
[64].P.Brasnett,L.Mihaylova,N.Canagarajah,et al.Particle Filtering with Multiple Cues for Object Tracking in Video Sequences.Proc.of SPIE's Annual Symp.EI ST,2005,5685430-441:
[65].D.Serby,E.K.Meier,L.V.Gool.Probabilistic Object Tracking Using Multiple Features.Proceedings of the 17th International Conference on Pattern Recognition,2004,2:184-187
[66].C.H.Shen,A.Hengel.Probabilistic Multiple Cue Integration for Particle Filter Based Tracking.Proc.7th Digital Image Computing:Techniques and Applications,2003,1:399-408
[67].P.Li,F.Chaumette.Image Cues Fusion for Object Tracking Based on Particle Filter.AMDO,2004,2004,3174:99-107
[68].周宏仁,敬忠良,王培德.机动目标跟踪.北京:国防工业出版社,1991,263-321
[69].付梦印,邓志红,张继伟.Kalman滤波理论及其在导航系统中的应用.北京:科学出版社,2003,79-90
[70].陆恺,田蔚风.最优估计理论及其在导航系统中的应用.上海:上海交通大学出版社,1990,135-147.122
[71].Huttenlocker D.P.Klanderman G.A.and Rucklidge W.J.Comparing Images Using the Hausdorff Distance[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1993,15(9):850-863