红外多目标实时跟踪方法的研究
|
摘要
靶场是对新兴武器装备性能进行试验检验和技术指标测试、鉴定而专门建造的设施齐备的广大地域，通过测量目标弹体飞行弹道参数来揭示被测试武器装备内在性能和固有特性已成为靶场采用的最直接最有效的手段。红外多目标跟踪技术，就是根据红外图像不同帧之间的差异性，对各个目标进行数据分析，得到各个目标不同的观测集合或轨迹，并通过分析目标的位置、速度、加速度、灰度特性等信息，预测并准确捕获其在下一帧图像上的运动位置及运动状态，实现对目标弹体及武器的试验检验过程。
靶场测量中的多目标跟踪技术具有以下几个难题：首先，图像信噪比低、对比度差，目标尺寸小、形状和纹理等信息匮乏，图像背景复杂，目标与背景交织在一起，目标通常难以辨别，造成红外多目标跟踪十分困难。其次，目标交叉运动过程中的数据关联问题，目标的丢失与重现现象，也是多目标跟踪的难点之一。目前国内外的研究重点都集中在红外单目标的跟踪上，而红外多目标的跟踪领域却研究较少，因此，开展红外多目标实时跟踪方面的研究具有重要的学术意义和实用价值。
结合当前工程实际需求，本文主要针对靶场测量中数量在10个以上的多批次连发目标群的实时跟踪问题展开研究，开展的研究工作主要有：1、对基于数字图像处理技术的图像预处理与多目标实时跟踪算法进行了讨论研究，为实现整个系统功能提供理论基础；2、针对多批次连发目标数量较多、目标运动随机性较大、目标跟踪过程数据量大、跟踪过程必须实时等特点，提出了一种新的多站组合跟踪方法；3、靶场测量中的多连发目标跟踪过程为实时跟踪过程，因此需要对数据进行高速实时处理，本文对数字图像处理的硬件平台进行设计及研究。
本文首先对面阵红外探测器的成像特性进行分析，研究讨论了红外图像中各个结构层次的特性，针对这种特性的差异性，采用基于稀疏编码理论的图像预处理算法，建立各个结构层次的超完备字典，可以针对不同的结构层次实现相应的预处理算法，使图像处理结果更加稀疏、简洁。其次，分析了单目标与多目标跟踪的原理，提出了一种采用马尔科夫（Markov）跳变系统与核聚类相结合的红外多目标跟踪算法，通过构建Markov跳变非线性系统的多目标转移模型，自回归位移预测模型（ARMA）对多目标跟踪过程中数据关联分析，解决了多目标跟踪问题中的目标间交汇、遮挡等问题，利用核聚类算法对粒子进行优化采样来提高粒子的贡献率，解决粒子滤波过程中的退化问题。再次，针对多批次连发目标数量较多以及目标运动随机性较大等特点，提出了一种多站组合跟踪方案，通过分析多批次连发目标群的运动特性，定义了三种不同类目标群的概念，提出了基于多尺度多级FCM和智能群决策的目标群分类决策策略，将目标群智能分类为不同的类目标群，利用图像融合决策器对多台经纬仪进行决策控制，实现不同类目标群的实时跟踪。最后，根据图像融合决策器的功能及多目标跟踪的技术要求，设计FPGA+双多核DSP处理器架构的图像融合决策器平台，通过多核处理器的并行数据处理、任务分配与信息交互提升系统对红外图像数据的处理速度。最后将提出的预处理算法、多目标跟踪算法及目标群分类决策策略进行了硬件移植，并通过外场多项试验测试，验证了算法和硬件平台系统方案的正确性、实时性以及可靠性。
Range is a vast and well-equipped place that is specially built for testing anddetecting the performance of new weapons. What is more, range is also used to testand identify the technical indicators. In the range, we reveal the intrinsic andinherent performance characteristics of tested weaponry by measuring the flighttrajectory parameters of target missile, which is the most direct and effective means.Infrared multi-target tracking technology is that based on the difference betweendifferent frames of the infrared image, we analyze the data of each target to gaindifferent observation set or track, then through the analysis of the target position,velocity, acceleration, grayscale characteristics and other information, we accuratelypredict and capture the movement location and motion status of the next frame,finally we succeed in testing the experiments process of target projectile or weapon.However, multi-target tracking technology of measuring range meets the followingchallenges: on one hand, infrared multi-target tracking becomes very difficultbecause of low SNR, poor contrast ratio, small size of target, lack of informationabout shape and texture, complex image background and the mixture of objectivesand background that make identification very difficult. On the other hand, dataassociation problem coming from target consolidation and cross-motion process andthe phenomenon of target lose and reappearance make multi-target trackingtechnology get into trouble. Recently, scholars who come from home or abroad do not pay more attention to multi-target tracking research instead of being concentratedon single-target tracking.So infrared multi-target tracking study plays an importantrole in academic region and has practical value.
Combined with the actual needs of the current project, major research workincludes the following points:1, having a discussion on the image pre-processingand multi-target tracking algorithm based on digital image processing technology inorder to provide theoretical basis for whole system function;2, working out newtracking strategy using several Theodolites,for the large number of multiple andcontinuous targets, great randomness of characteristics of target motion;3, designingand studying the hardware platform of digital image processing so as to address themulti-target tracking problem in measurement of the range.
The paper analyzes the imaging characteristics of array infrared detector thenstudy and discuss the characteristic of each structure of infrared images. Accordingto the characteristic difference, we establish ultra-complete dictionary that help tocome up with corresponding pre-processing algorithm being aimed at different levelof structure to make the effect of image process more sparse and simple by adoptingimage preprocessing algorithm on the basis of sparse coding theory. At the sametime, the article proposes infrared multi-target tracking algorithm that is thecombination of MJNSs and kernel clustering. The algorithm can solve themulti-target tracking problems such as: target intersection, obstructions and so onthrough building the MJNSs model and ARMA model to analyze the related datagenerating from multi-target tracking process. In addition, by using kernel clusteringalgorithm, we optimize the sampling to improve the high contribution rate of theparticles. And we also propose a several Theodolites tracking program in accordancewith the great number of continuous shoot and high randomness of target motion. Byanalyzing the motion characteristics of the target group, we give different conceptionfor three various target groups and figure out classification and decision-makingstrategy of target group based on Intelligent Decision Control and multi-scale andmulti-level FCM. The strategy classifies automatically target group into different class target groups. We use Image Fusion Decider to control severaltheodolites,which is to track different target groups. We build Image Fusion Deciderplatform that consists of FPGA processor and dual multi-core DSP processorsaccording to the function of Image Fusion Decider and the requirement ofmulti-target tracking technology then raise the data processing speed of infraredimage with parallel data processing, task allocation and information exchange ofmulti-core processors.
Finally, we make hardware transplant based on the proposed pre-processingalgorithm, multi-target tracking algorithm and classification and decision-makingstrategies of target group. At the same time we verify the correctness and feasibilityof the algorithm and system solutions of hardware platform by testing a great deal ofactual range.
靶场测量中的多目标跟踪技术具有以下几个难题：首先，图像信噪比低、对比度差，目标尺寸小、形状和纹理等信息匮乏，图像背景复杂，目标与背景交织在一起，目标通常难以辨别，造成红外多目标跟踪十分困难。其次，目标交叉运动过程中的数据关联问题，目标的丢失与重现现象，也是多目标跟踪的难点之一。目前国内外的研究重点都集中在红外单目标的跟踪上，而红外多目标的跟踪领域却研究较少，因此，开展红外多目标实时跟踪方面的研究具有重要的学术意义和实用价值。
结合当前工程实际需求，本文主要针对靶场测量中数量在10个以上的多批次连发目标群的实时跟踪问题展开研究，开展的研究工作主要有：1、对基于数字图像处理技术的图像预处理与多目标实时跟踪算法进行了讨论研究，为实现整个系统功能提供理论基础；2、针对多批次连发目标数量较多、目标运动随机性较大、目标跟踪过程数据量大、跟踪过程必须实时等特点，提出了一种新的多站组合跟踪方法；3、靶场测量中的多连发目标跟踪过程为实时跟踪过程，因此需要对数据进行高速实时处理，本文对数字图像处理的硬件平台进行设计及研究。
本文首先对面阵红外探测器的成像特性进行分析，研究讨论了红外图像中各个结构层次的特性，针对这种特性的差异性，采用基于稀疏编码理论的图像预处理算法，建立各个结构层次的超完备字典，可以针对不同的结构层次实现相应的预处理算法，使图像处理结果更加稀疏、简洁。其次，分析了单目标与多目标跟踪的原理，提出了一种采用马尔科夫（Markov）跳变系统与核聚类相结合的红外多目标跟踪算法，通过构建Markov跳变非线性系统的多目标转移模型，自回归位移预测模型（ARMA）对多目标跟踪过程中数据关联分析，解决了多目标跟踪问题中的目标间交汇、遮挡等问题，利用核聚类算法对粒子进行优化采样来提高粒子的贡献率，解决粒子滤波过程中的退化问题。再次，针对多批次连发目标数量较多以及目标运动随机性较大等特点，提出了一种多站组合跟踪方案，通过分析多批次连发目标群的运动特性，定义了三种不同类目标群的概念，提出了基于多尺度多级FCM和智能群决策的目标群分类决策策略，将目标群智能分类为不同的类目标群，利用图像融合决策器对多台经纬仪进行决策控制，实现不同类目标群的实时跟踪。最后，根据图像融合决策器的功能及多目标跟踪的技术要求，设计FPGA+双多核DSP处理器架构的图像融合决策器平台，通过多核处理器的并行数据处理、任务分配与信息交互提升系统对红外图像数据的处理速度。最后将提出的预处理算法、多目标跟踪算法及目标群分类决策策略进行了硬件移植，并通过外场多项试验测试，验证了算法和硬件平台系统方案的正确性、实时性以及可靠性。
Range is a vast and well-equipped place that is specially built for testing anddetecting the performance of new weapons. What is more, range is also used to testand identify the technical indicators. In the range, we reveal the intrinsic andinherent performance characteristics of tested weaponry by measuring the flighttrajectory parameters of target missile, which is the most direct and effective means.Infrared multi-target tracking technology is that based on the difference betweendifferent frames of the infrared image, we analyze the data of each target to gaindifferent observation set or track, then through the analysis of the target position,velocity, acceleration, grayscale characteristics and other information, we accuratelypredict and capture the movement location and motion status of the next frame,finally we succeed in testing the experiments process of target projectile or weapon.However, multi-target tracking technology of measuring range meets the followingchallenges: on one hand, infrared multi-target tracking becomes very difficultbecause of low SNR, poor contrast ratio, small size of target, lack of informationabout shape and texture, complex image background and the mixture of objectivesand background that make identification very difficult. On the other hand, dataassociation problem coming from target consolidation and cross-motion process andthe phenomenon of target lose and reappearance make multi-target trackingtechnology get into trouble. Recently, scholars who come from home or abroad do not pay more attention to multi-target tracking research instead of being concentratedon single-target tracking.So infrared multi-target tracking study plays an importantrole in academic region and has practical value.
Combined with the actual needs of the current project, major research workincludes the following points:1, having a discussion on the image pre-processingand multi-target tracking algorithm based on digital image processing technology inorder to provide theoretical basis for whole system function;2, working out newtracking strategy using several Theodolites,for the large number of multiple andcontinuous targets, great randomness of characteristics of target motion;3, designingand studying the hardware platform of digital image processing so as to address themulti-target tracking problem in measurement of the range.
The paper analyzes the imaging characteristics of array infrared detector thenstudy and discuss the characteristic of each structure of infrared images. Accordingto the characteristic difference, we establish ultra-complete dictionary that help tocome up with corresponding pre-processing algorithm being aimed at different levelof structure to make the effect of image process more sparse and simple by adoptingimage preprocessing algorithm on the basis of sparse coding theory. At the sametime, the article proposes infrared multi-target tracking algorithm that is thecombination of MJNSs and kernel clustering. The algorithm can solve themulti-target tracking problems such as: target intersection, obstructions and so onthrough building the MJNSs model and ARMA model to analyze the related datagenerating from multi-target tracking process. In addition, by using kernel clusteringalgorithm, we optimize the sampling to improve the high contribution rate of theparticles. And we also propose a several Theodolites tracking program in accordancewith the great number of continuous shoot and high randomness of target motion. Byanalyzing the motion characteristics of the target group, we give different conceptionfor three various target groups and figure out classification and decision-makingstrategy of target group based on Intelligent Decision Control and multi-scale andmulti-level FCM. The strategy classifies automatically target group into different class target groups. We use Image Fusion Decider to control severaltheodolites,which is to track different target groups. We build Image Fusion Deciderplatform that consists of FPGA processor and dual multi-core DSP processorsaccording to the function of Image Fusion Decider and the requirement ofmulti-target tracking technology then raise the data processing speed of infraredimage with parallel data processing, task allocation and information exchange ofmulti-core processors.
Finally, we make hardware transplant based on the proposed pre-processingalgorithm, multi-target tracking algorithm and classification and decision-makingstrategies of target group. At the same time we verify the correctness and feasibilityof the algorithm and system solutions of hardware platform by testing a great deal ofactual range.
引文
[1]洛刚,尚娜,胡晓枫等.国家靶场开展装备作战试验问题研究[J].装备学院学报,2013,01:114-11
[2]陈冬科,张迪.靶场试验关键过程控制[J].鱼雷技术,2013,01:54-57.
[3]周芳,周正虎.国外信息保障靶场建设[J].指挥信息系统与技术,2013,01:1-4+12
[4]石子坚.美国室内靶场现状研究[J].公安教育,2013,02:73-76
[5]崔侃,曹裕华.美军装备试验靶场建设发展及其启示[J].装备学院学报,2013,04:110-113.
[6]朱玮，赵立荣，柳玉晗等．光电经纬仪子母弹多目标提取方法[J]．光学精密工程，2008，16(11)：2140-2144
[7]杨露，金星，沈怀荣．利用光学成像进行子母弹多子弹抛射弹道测量[J]．弹箭与制导学报，2004，24(2)：58-61
[8]何小勇，吴钦章，高晓东．光电经纬仪子母弹多子弹轨迹交会法[J]．光电工程，2006，33(9)：15-18
[9]陈犇．靶场光测定位方法研究与应用[D]：［硕士学位论文］．长沙：国防科学技术大学，2011
[10]朱自谦，王淑兰．多目标跟踪(MTT)技术的发展[J]．航空电子技术，1995，3：48-51
[11]安毓英，曾晓东.光电探测原理[M].西安：西安电子科技大学出版社，2004
[12]王静雷.热红外图像异常目标检测算法研究[D].浙江大学,2013.
[13]卢刚.基于图像融合的红外弱小目标检测算法研究[D].西安电子科技大学,2013.
[14]王晓文,赵宗贵,庞秀梅等.基于频谱分析和图像融合的红外弱小目标增强[J].中国电子科学研究院学报,2013,03:258-265.
[15]王雨,蒋增波,李文刚.红外图像不均匀背景消除方法研究[A].陕西省光学学会、中国光学学会高速摄影与光子学专业委员会.2013年（第五届）西部光子学学术会议论文集[C].陕西省光学学会、中国光学学会高速摄影与光子学专业委员会:,2013:1.
[16]曹琦，王德江，张奇等.红外点目标检测中的能量积累[J].光学精密工程，2010，18(3):741-747.
[17]曾雅琼，陈钱．基于改进的双边滤波的单帧红外目标背景抑制[J]．红外技术，2011,33(9)：537-540
[18] Mohimani H.，Babaie-zadeh M.，Jutten C. A fast approach for overcompletesparse decomposition based on smoothed L0-norm[J].IEEE Trans. Signal. Processing，2009,057（1）:289-301
[19] WRIGHT J，YANG Y，GANESH A，et al Robust Face Recognition via SparseRepresentation[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2009,31(2):210-227.
[20] MAIRAL J，ELAD M，SAPIRO G. Sparse Representation for Color ImageRestoration[J].IEEE Trans on Image Processing，28,17（1）:53-69.
[21]赵高长,张磊,武风波.改进的中值滤波算法在图像去噪中的应用[J].应用光学,2011,04:678-682.
[22] Cao Y， Liu RM， Yang J. Infrared Small Targets Detection UsingPPCA[J].International Journal of Infrared and Millimeter Waves,2008,2（94）:385-395.
[23] DelMarco S，Agaian S. The design of wavelets for image enhancement andtarget detection[J]. Proc. SPIE，2009，7351.
[24]汪大宝，刘上干,寇小明．基于MRF的自适应正则化红外背景抑制算法[J]．红外与毫米波学报，2009，28(6)：440-444
[25]秦翰林，李佳，周慧鑫等.采用剪切波变换的红外弱小目标背景[J].红外与毫米波学报，2011.4.163-168.
[26]陈炳文，王文伟，秦前清.基于分数阶积分算子的红外弱小目标检测[J].控制与决策，2012,27（1）：147-151.
[27]张翔，张建奇，秦翰林等．刘菁基于对偶树复小波变换的红外目标背景抑制[J]．光子学报，2010,39(9)：1672-1677
[28]朱自谦，王淑兰．多目标跟踪(MTT)技术的发展[J]．航空电子技术，1995，3：48-51
[29] Mayor, Marco A.; Carroll, Robert L.A Multi-Target Track InitiationAlgorithm[J], American Control Conference,1987,1128-1130.
[30] Hongyan Zhu; Chongzhao Han; Yan Lin.Particle labeling PHD filter formulti-target track-valued estimates[J]. Information Fusion (FUSION),2011Proceedings of the14th International.2011,1-8.
[31] Han Guang-liang; Meng Bo. A new real time multi-target tracking algorithm invideo[J]. Computer Application and System Modeling (ICCASM),2010International,13:568-572
[32] Wei Yi; Morelande, M.R.; Lingjiang Kong. An Efficient Multi-FrameTrack-Before-Detect Algorithm for Multi-Target Tracking[J]. Selected Topics inSignal Processing,2013,7(3):421-434.
[33] Coraluppi, S.; Carthel, C. Optimality and ghosting phenomena in multi-targettracking[J]. Information Fusion (FUSION),2012,302-308.
[34] Weijie Kong; Wei Zheng. Fault diagnosis of train sensors based on evolutionarygenetic Particle Filter[J]. Intelligent Rail Transportation (ICIRT),2013IEEE,255-258.
[35] Jinlong Xian; Yulong Geng. Multi-user detection based on Gaussian particlefilter in impulsive noise[J]. Instrumentation and Measurement, Sensor Network andAutomation (IMSNA),2013,12(23-24):349-352.
[36]薛陈，朱明，陈爱华．一种鲁棒的基于改进Mean-shift的目标跟踪算法[J]．光学精密工程，2010,18(1)：234-239
[37]李荣华．基于AdaBoost与粒子滤波的人脸检测与跟踪[D]：［硕士学位论文］．南京：南京理工大学，控制理论与控制工程，2010
[38] Chang C and Ansara R. Kernel particle filter for visual tracking [J]. IEEE SignalProcessing Letters.2005,12(3):242-245
[39]蒋蔚．粒子滤波改进算法研究与应用[D]：[博士学位论文]．哈尔滨：哈尔滨工业大学，2010
[40]田隽．基于粒子滤波的视频目标跟踪算法的研究[D]：[博士学位论文].江苏：中国矿业大学，2010
[41] Ting Y, Ying W. Collaborative Tracking of Multiple Targets [C]. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition. Washington, DCIEEE Press,1995
[42]程水英，张剑云．裂变自举粒子滤波[J]：电子学报，2008,36(3)：500-504
[43] Gordon N, Salmond D, SMITH A. Novel approach to nonlinear/non-gaussianbayesian state estimation[J]. IEE Proceedings on Radar and Signal Processing,1993,140(2):107-113
[44] Shan C F, Tan T N, Wei Y CH. Real-time hand tracking using a mean-shiftembedded particle filter [J]. Pattern Recognition.2007,40:1958-1962
[45] Darroch J N, Lauritzen S L, Speed T P. Markov-field and log-linear models forcontingency tables [J]. Annals of Statistics,1980,8(3):522-539
[46] Edwards D. Introduction to Graphical Modelling [M]. Springer-Verlag,2ndedition，2000
[47] Gan ZH, JiangM. Articulated body tracking by immuneparticle filter [J]. LectureNotes in Computer Science,2006,4247:853.
[48] Clapp TC. Statisticalmethods for the communication data.[D]. Cambridne: University of Cambr idge,2000.
[49] Patrick P, Vermaak J. Data fusion for visual tracking with particle [J]. IEEE,2004,92(3):495-513
[50] Michaels Elad，Mariol A. T. Figueiredo，YIMa. On the role of sparse andredundant representations in image proeessing[J]. Proceedings of the IEEE，2010，98(6):972-982.
[51] A. Bruckstein，D.Donoho，and M.Elad. From sparse solutions of system ofequations to sparse modelling of signals and images[J].SIAM Rev.2009,5:34-81
[52] Otsu N. A threshold selection method from gray-level histogram [J]. IEEE Trans.on Systems, Man, and Cybernetics,1979,9(1):62-66
[53] Beghdadi A, Negrate AL, Lesegno P. V. Entropic thresholding using a blocksoure model [J]. Computer Models Image Process,1995,57:197-205
[54] Torr P.H.S, Zisserman A. Concerning Bayesian motion segmentation, modelaveraging, matching and the trifocal tensor [C]. Fifth European Conference onComputer Vision.Germany Freiburg: Springer Verlag,1998
[55] Changfa Xu, Minghong Pi, Chunxi Zhu, Guokuan Li. Interpolating-wavelet andits application [J]. In Proceedings of SPIE Vol,35-45
[56] L. H. Sun, Y. Z. Wang, Y. Q. He. Design and Implement of Infrared Small TargetReal-time Detection System Based on Pipeline Technology [C].27th InternationalCongress on High-Speed Photography and Photonics, Xi'an, China,2007. Proc. ofSPIE,6279-6283
[57]程建,周越,蔡念,杨杰.基于粒子滤波的红外目标跟踪[J].红外与毫米波学报,2006,02:113-117
[58]高建坡,王煜坚,杨浩,吴镇扬.以颜色和形状直方图为线索的粒子滤波人脸跟踪[J].中国图象图形学报,2007,03:466-473
[59]程水英,张剑云.粒子滤波评述[J].宇航学报,2008,04:1099-1111
[60]夏克寒,许化龙,张朴睿.粒子滤波的关键技术及应用[J].电光与控制,2005,06:1-4+19
[61]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策,2005,04:361-365+371
[62]李俊山,陈霞,李建华.不同地面背景下目标与背景红外辐射对比特性[J].红外与激光工程,2014,02:424-428
[63]陈晨.高超声速飞行器红外辐射特性分析[D].南京理工大学,2013
[64]李园园.目标红外辐射特性测试技术研究[D].南京理工大学,2013.
[65]李建勋,童中翔,王超哲,柴栋,张志波.飞机红外辐射特性分析与易损性评估[J].激光与红外,2013,02:180-185.
[66]杨词银,张建萍,曹立华.地基空间目标红外辐射特性测量技术[J].仪器仪表学报,2013,02:304-310.
[67]张辉，赵保军，唐林波等．变维卡尔曼滤波实现运动目标的跟踪[J]．光学学报，2010，30(5)：1291-1296
[68]严柏军，郑链，王克勇．基于不变矩特征匹配的快速目标检测算法[J]．红外技术，2001，23(6)：8-11
[69]梁敏，刘贵喜．基于自适应混合滤波的多目标跟踪算法[J]．光学学报，2010，30(9)：2554-2561
[70]邹江威，陈曾平．应用形态学与图像流法的空间小目标提取方法[J]．光电工程，2005，32(4)：13-15
[71]侯洁,辛云宏.基于高通滤波和图像增强的红外小目标检测方法[J].红外技术,2013,05:279-284.
[72]刘渭清.一种数字全通滤波器的最小二乘设计法[J].西南师范大学学报(自然科学版),2013,12:90-94
[73]蒋永馨，金俣欣，王孝通等．基于马尔可夫随机场的运动目标检测[J]．计算机工程与应用，2010，46(15)：182-184
[74]孔军，汤心溢，蒋敏等．采用尺度空间特征点匹配的红外目标跟踪[J]．红外与激光工程，2011,40(11)：2104-2109
[75]陶珂.小波去噪质量评价方法研究[D].中南大学,2012
[76] Ran He; Wei-Shi Zheng; Bao-Gang Hu; Xiang-Wei Kong.Two-StageNonnegative Sparse Representation for Large-Scale Face Recognition. NeuralNetworks and Learning Systems, IEEE Transactions,2013,24(1):35-46.
[77] Julazadeh, A.; Marsousi, M.; Alirezaie, J. Classification based on sparserepresentation and Euclidian distance[J]. Visual Communications and ImageProcessing (VCIP),2012IEEE,1-6.
[78]路锦正.基于稀疏表示的图像超分辨率重构技术研究[D].电子科技大学,2013.
[79]宋相法.基于稀疏表示和集成学习的若干分类问题研究[D].西安电子科技大学,2013.
[80] Yuanyuan Zuo; Bo Zhang. General image classification based on sparserepresentation[J]. Cognitive Informatics (ICCI),20109th IEEE InternationalConference on.2010,223-229.
[81]钟九生.基于稀疏表示的光学遥感影像超分辨率重建算法研究[D].南京师范大学,2013.
[82] Haoliang Yuan; Yang Lu; Lina Yang; Huiwu Luo; Yuan Yan Tang. Sparserepresentation using contextual information for hyperspectral image classification[J].Cybernetics (CYBCONF),2013IEEE International Conference on,2013,138-143.
[83]匡金骏.基于稀疏表示的图像分类与目标跟踪研究[D].重庆大学,2013.
[84] Samadi, S.; Cetin, M.; Masnadi-Shirazi, M.A.Sparse Signal Representation forComplex-Valued Imaging[J]. Digital Signal Processing Workshop and5th IEEESignal Processing Education,2009,4-7:365-370.
[85]汪桃林.基于稀疏表示的短语音说话人识别[D].电子科技大学,2013.
[86]刘钢.基于压缩感知和稀疏表示理论的图像去噪研究[D].电子科技大学,2013.
[87] Hongkai Xiong; Zhiming Pan; Xinwei Ye; Chang Wen Chen.SparseSpatio-Temporal Representation With Adaptive Regularized Dictionary Learning forLow Bit-Rate Video Coding[J]. Circuits and Systems for Video Technology, IEEETransactions on,2013.23(4):710-728.
[88]闫永帅.基于稀疏表示的高光谱遥感图像分类[D].电子科技大学,2013.
[89]王小娜.基于字典学习的图像稀疏表示及其在压缩感知中的应用[D].燕山大学,2013.
[90]兰诚栋,陈亮,卢涛.利用位置权重稀疏表示的人脸超分辨率算法[J].武汉大学学报(信息科学版),2013,01:27-30.
[91]孔繁锵,井庆丰,计振兴.图像压缩感知的自适应方向提升稀疏表示及重构算法[J].宇航学报,2013,01:121-127.
[92] Bar-Shalom Y, Fortmann T. Tracking and data association [M]．Acad. Press.New York.1988
[93] D. L. Alspach, H. W. Sorenson. Nonlinear Bayesian Estimation using GaussianSum Approximation [J]．IEEE Trans. on AC,1972,17(4):439-448
[94] Julier S, Uhlmann J. A new extension of the Kalman filter to nonlinear systems
[C]．Proc. SPIE,1997，3068:182-193
[95] S. Zhou, R. Chellappa, B. Moghaddam. Visual Tracking and Recognition usingAppearance-adaptive Models in Particle Filters [J]. IEEE Trans. Image Proc.2004,13(11):1491-1505
[96]胡诗国,方洋旺,蔡文新等.连续Markov跳变系统最优控制[J].控制与决策,2013,03:396-401.
[97]孟波,高存臣,考永贵等.一类时滞不确定随机Markovian跳变系统的滑模控制[J].中国海洋大学学报(自然科学版),2013,03:104-112.
[98]陈佳,方洋旺,楼顺天.随机跳变摄动系统的鲁棒H_∞控制[J].控制与决策,2013,02:279-284.
[99]叶丹,范泉涌.转移概率部分已知的Markov跳变系统鲁棒H_2控制[J].东北大学学报(自然科学版),2013,02:153-156.
[100]岳生伟,焦贤发.具有Markov跳的时滞中立系统的均方指数稳定[J].合肥工业大学学报(自然科学版),2013,04:477-482.
[101]刘晓华,吕娜. Markov跳变系统的输出反馈鲁棒预测控制[J].控制理论与应用,2013,11:.
[102]李科，徐克虎.特征融合和聚类核函数平滑采样优化的粒子滤波目标跟踪方法[J].计算机科学，2012.39（4）210-215.
[103] R. Edmondson, M. Rodgers, M. Banish. Ranganath.Single-Frame ImageProcessing Techniques for Low-SNR Infrared Imagery [J]. Infrared Technology andApplications. Proc. of SPIE.2008,6940:6943
[104] SHAN C F, TAN T N, WEI Y CH. Real-time hand tracking using amean-shift embedded particle filter [J]. Pattern Recognition.2007,40:1958-1962
[105] Yaghoobi M，Daudet L，Davies M. Parametric dictionary design for sparsecoding[J].IEEE Transactions on Signal Processing，2009，57（12）：4800-4810.
[106]刘伟.基于人工神经网络的遥感影像模式分类研究[D].沈阳工业大学,2005
[107]单建丰,郁道银,陈晓冬等.基于模糊聚类和C-V模型的水平集方法图像分割[J].中国介入影像与治疗学,2013,08:498-502.
[108]段军,位保振.改进的模糊C_均值聚类医学图像分割算法[J].微型机与应用,2013,16:36-37+41.
[109]曲福恒.一类模糊聚类算法研究及其应用[D]:[博士学位论文].吉林：吉林大学，2009.
[110]陈厦,方方,胡战利.模糊聚类算法综述[J].生命科学仪器,2013,06:33-37.
[111]倪迎花.图像处理中的模糊算法及实现分析[J].信息与电脑(理论版),2013,08:53-54.
[112] Esmin, A.A.A.; Coelho, R.A. Consensus Clustering Based on Particle SwarmOptimization Algorithm[J]. Systems, Man, and Cybernetics (SMC),2013(10):2280-2285.
[113]朱林,雷景生,毕忠勤等.一种基于数据流的软子空间聚类算法[J].软件学报,2013,11:2610-2627.
[114]刘建生,乔尚平,匡奕群.基于差分粒子群和模糊聚类的彩色图像分割算法[J].江西理工大学学报,2013,05:66-71.
[115]燕蜻.混合型多属性群决策方法研究[D].山西大学,2012
[116]陈晓红,杨立.基于多属性群决策方法的障碍诊断模型[J].科研管理,2013,03:115-121
[117]谭旭,毛太田,张少丁等.基于粒计算的多属性群决策分析[J].四川大学学报(工程科学版),2013,04:140-148
[118]刘翔.突发危机事件应急群决策支持系统的设计与实现[D].复旦大学,2012.
[119]刘兮.基于多粒度二元语义信息的几类信息集结算子及其在多属性群决策中的应用[D].安徽大学,2012
[120]姚成玉,陈东宁,冯中魁等.基于贝叶斯网络和理想解动态群决策的故障诊断方法[J].中国机械工程,2013,16:2235-2241
[121]石福丽,许永平,杨峰.考虑专家偏好关联的群决策方法及其应用[J].控制与决策,2013,03:391-395+401
[122]蔡付龄,廖貅武,杨娜.基于案例信息的多准则群决策分类方法[J].管理科学学报,2013,02:22-32
[123]王文全,黄胜,王超等.舰船通道布置方案的直觉模糊多属性群决策方法[J].上海交通大学学报,2013,06:894-899
[124]戚筱雯,梁昌勇,黄永青等.基于混合型评价矩阵的多属性群决策方法[J].系统工程理论与实践,2013,02:473-481
[125]燕蜻,梁吉业.混合多属性群决策中的群体一致性分析方法[J].中国管理科学,2011,06:133-140.
[126] Xilinx inc, Virtex-5FPGA Integrated Endpoint Block for PCI ExpressDesigns[EB/OL]. UG197, July22,2009. http://www.xilinx.com.
[127] Xilinx inc, Virtex-5FPGA Integrated Endpoint Block for PCI ExpressDesigns: DDR2SDRAM DMA Initiator Demonstration Platform[EB/OL]. XAPP859,July31,2008. http://www.xilinx.com.
[128]孟思远,刘建,奚杰等.一种多核DSP的动态存储器分配设计方法[J].微电子学与计算机,2008,11:21-24.
[129]李鑫,姜明.多核DSP高速实时信号处理系统设计[J].光学技术,2012,01:116-120.
[130]徐力,王沁,史少波.一种基于异构多核DSP的IEEE802.11a接收端基带处理的研究和实现[J].计算机应用研究,2012,01:241-245.
[131]徐力,史少波,王沁.面向SDR应用的多核DSP低功耗设计[J].电子科技大学学报,2012,01:136-141.
[132]孙科林,周维超,吴钦章等.光纤实时传输的多核DSP图像处理系统(英文)[J].光电工程,2012,04:136-144.
[133]汪东,陈书明.一种面向异构多核DSP的数据流传输控制引擎设计与评测[J].国防科技大学学报,2009,05:19-23
[134]敖天勇,陈杰,刘建,许汉荆,奚杰,张伟风.基于CostarⅡ的异构多核DSP设计与实现[J].微电子学与计算机,2010,03:59-62.
[135]刘建,陈杰,敖天勇,许汉荆.片上异构多核DSP同步与通信的实现[J].电子科技大学学报,2010,04:528-531+536.
[136]方兴,陈书明.支持细粒度并行性开发的多核DSP快速核间通信机制[J].计算机工程与科学,2009,04:130-133.
[137]辛熠,杨瑞民.多核DSP编程技术研究[J].电子测试,2011,09:81-85.
[138]吴家铸,田希,刘继福等.计算高度密集型应用在异构多核DSP上的运行方法研究[J].微电子学与计算机,2011,05:45-47+52.
[139]陈书明,汪东,陈小文等.一种面向多核DSP的小容量紧耦合快速共享数据池[J].计算机学报,2008,10:1737-1744.
[140]汪东,陈书明. DSCF:一种面向共享存储多核DSP的数据流分簇前向技术[J].计算机研究与发展,2008,08:1446-1453.
[141]单祥茹. TI多核DSP再出新品高性能计算时代即将到来[J].中国电子商情(基础电子),2012,Z1:29-30.
[142]赵海霖，韦文，汤俊．多DSP平台下性能预测算法[J]：信息与电子工程，2011，9(4)，472-477
[143]尹九阳，方滨兴，张宏莉．基于单程序多数据流并行软件的性能预测法[J]：清华大学学报(自然科学版)，2005，45(4)，509-512
[144]汪向阳,张云峰.基于双DSP的图像处理系统硬件设计[J].仪器仪表学报,2005,S1:646-647+656.
[145]谢明,吉书鹏,段哲民.一种基于高速双DSP的柔性机载实时图像跟踪系统研究[J].红外技术,2005,02:99-104.
[146]田小永,古志民.基于TMS320C55x DSP的代码优化[J].计算机应用,2003,01:56-59.
[147]邱春武.基于并行处理单元的代码优化方法研究[D]：[硕士学位论文].北京：北京邮电大学,2008.
[148]施虹宇. TMS320DM642上的代码优化研究[D]：[硕士学位论文].北京：北京邮电大学,2013.
[149]源江科技.VC编程技巧[M].上海科学普及出版社.2002.
[150]胡海生，李升亮.Visual C++6.0编程学习捷径[M].清华大学出版社.2003.
[151]王诚，薛小刚，钟信潮.FPGA/CPLD设计工具Xilinx ISE使用详解[M].人民邮电出版社.2004:95-159.
[152] Paragios N, Rsmesh V. A MRF-based approach for real-time subwaymonitoring [C]. Processing of Computer Vision and Pattern Recognition,2001:295-313
[153]郭伟．复杂背景下红外目标检测与跟踪[D]：[博士学位论文]．西安：西安电子科技大学，2008
[154] Alexis P.Tzannes, Dana H.Brooks. Point Target Detection in IR ImageSequences. A Hypothesis-testing Approach Based on Target and Clutter TemporalProfile Modeling [J]. Opt. Eng,2000,39(8).2270-2278
[155] Christopher WK. Infrared target detection: signal and noise sensitivityanalysis thesis [C].SPIE,1989.AD207-215
[156] Nishiguchi K, Kobayashi M, Ichikawa A.Small Target Detection from ImageSequences Using Recursive Max Filter[C].SPIE.1995.153-165
[2]陈冬科,张迪.靶场试验关键过程控制[J].鱼雷技术,2013,01:54-57.
[3]周芳,周正虎.国外信息保障靶场建设[J].指挥信息系统与技术,2013,01:1-4+12
[4]石子坚.美国室内靶场现状研究[J].公安教育,2013,02:73-76
[5]崔侃,曹裕华.美军装备试验靶场建设发展及其启示[J].装备学院学报,2013,04:110-113.
[6]朱玮，赵立荣，柳玉晗等．光电经纬仪子母弹多目标提取方法[J]．光学精密工程，2008，16(11)：2140-2144
[7]杨露，金星，沈怀荣．利用光学成像进行子母弹多子弹抛射弹道测量[J]．弹箭与制导学报，2004，24(2)：58-61
[8]何小勇，吴钦章，高晓东．光电经纬仪子母弹多子弹轨迹交会法[J]．光电工程，2006，33(9)：15-18
[9]陈犇．靶场光测定位方法研究与应用[D]：［硕士学位论文］．长沙：国防科学技术大学，2011
[10]朱自谦，王淑兰．多目标跟踪(MTT)技术的发展[J]．航空电子技术，1995，3：48-51
[11]安毓英，曾晓东.光电探测原理[M].西安：西安电子科技大学出版社，2004
[12]王静雷.热红外图像异常目标检测算法研究[D].浙江大学,2013.
[13]卢刚.基于图像融合的红外弱小目标检测算法研究[D].西安电子科技大学,2013.
[14]王晓文,赵宗贵,庞秀梅等.基于频谱分析和图像融合的红外弱小目标增强[J].中国电子科学研究院学报,2013,03:258-265.
[15]王雨,蒋增波,李文刚.红外图像不均匀背景消除方法研究[A].陕西省光学学会、中国光学学会高速摄影与光子学专业委员会.2013年（第五届）西部光子学学术会议论文集[C].陕西省光学学会、中国光学学会高速摄影与光子学专业委员会:,2013:1.
[16]曹琦，王德江，张奇等.红外点目标检测中的能量积累[J].光学精密工程，2010，18(3):741-747.
[17]曾雅琼，陈钱．基于改进的双边滤波的单帧红外目标背景抑制[J]．红外技术，2011,33(9)：537-540
[18] Mohimani H.，Babaie-zadeh M.，Jutten C. A fast approach for overcompletesparse decomposition based on smoothed L0-norm[J].IEEE Trans. Signal. Processing，2009,057（1）:289-301
[19] WRIGHT J，YANG Y，GANESH A，et al Robust Face Recognition via SparseRepresentation[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2009,31(2):210-227.
[20] MAIRAL J，ELAD M，SAPIRO G. Sparse Representation for Color ImageRestoration[J].IEEE Trans on Image Processing，28,17（1）:53-69.
[21]赵高长,张磊,武风波.改进的中值滤波算法在图像去噪中的应用[J].应用光学,2011,04:678-682.
[22] Cao Y， Liu RM， Yang J. Infrared Small Targets Detection UsingPPCA[J].International Journal of Infrared and Millimeter Waves,2008,2（94）:385-395.
[23] DelMarco S，Agaian S. The design of wavelets for image enhancement andtarget detection[J]. Proc. SPIE，2009，7351.
[24]汪大宝，刘上干,寇小明．基于MRF的自适应正则化红外背景抑制算法[J]．红外与毫米波学报，2009，28(6)：440-444
[25]秦翰林，李佳，周慧鑫等.采用剪切波变换的红外弱小目标背景[J].红外与毫米波学报，2011.4.163-168.
[26]陈炳文，王文伟，秦前清.基于分数阶积分算子的红外弱小目标检测[J].控制与决策，2012,27（1）：147-151.
[27]张翔，张建奇，秦翰林等．刘菁基于对偶树复小波变换的红外目标背景抑制[J]．光子学报，2010,39(9)：1672-1677
[28]朱自谦，王淑兰．多目标跟踪(MTT)技术的发展[J]．航空电子技术，1995，3：48-51
[29] Mayor, Marco A.; Carroll, Robert L.A Multi-Target Track InitiationAlgorithm[J], American Control Conference,1987,1128-1130.
[30] Hongyan Zhu; Chongzhao Han; Yan Lin.Particle labeling PHD filter formulti-target track-valued estimates[J]. Information Fusion (FUSION),2011Proceedings of the14th International.2011,1-8.
[31] Han Guang-liang; Meng Bo. A new real time multi-target tracking algorithm invideo[J]. Computer Application and System Modeling (ICCASM),2010International,13:568-572
[32] Wei Yi; Morelande, M.R.; Lingjiang Kong. An Efficient Multi-FrameTrack-Before-Detect Algorithm for Multi-Target Tracking[J]. Selected Topics inSignal Processing,2013,7(3):421-434.
[33] Coraluppi, S.; Carthel, C. Optimality and ghosting phenomena in multi-targettracking[J]. Information Fusion (FUSION),2012,302-308.
[34] Weijie Kong; Wei Zheng. Fault diagnosis of train sensors based on evolutionarygenetic Particle Filter[J]. Intelligent Rail Transportation (ICIRT),2013IEEE,255-258.
[35] Jinlong Xian; Yulong Geng. Multi-user detection based on Gaussian particlefilter in impulsive noise[J]. Instrumentation and Measurement, Sensor Network andAutomation (IMSNA),2013,12(23-24):349-352.
[36]薛陈，朱明，陈爱华．一种鲁棒的基于改进Mean-shift的目标跟踪算法[J]．光学精密工程，2010,18(1)：234-239
[37]李荣华．基于AdaBoost与粒子滤波的人脸检测与跟踪[D]：［硕士学位论文］．南京：南京理工大学，控制理论与控制工程，2010
[38] Chang C and Ansara R. Kernel particle filter for visual tracking [J]. IEEE SignalProcessing Letters.2005,12(3):242-245
[39]蒋蔚．粒子滤波改进算法研究与应用[D]：[博士学位论文]．哈尔滨：哈尔滨工业大学，2010
[40]田隽．基于粒子滤波的视频目标跟踪算法的研究[D]：[博士学位论文].江苏：中国矿业大学，2010
[41] Ting Y, Ying W. Collaborative Tracking of Multiple Targets [C]. Proceedings ofIEEE Conference on Computer Vision and Pattern Recognition. Washington, DCIEEE Press,1995
[42]程水英，张剑云．裂变自举粒子滤波[J]：电子学报，2008,36(3)：500-504
[43] Gordon N, Salmond D, SMITH A. Novel approach to nonlinear/non-gaussianbayesian state estimation[J]. IEE Proceedings on Radar and Signal Processing,1993,140(2):107-113
[44] Shan C F, Tan T N, Wei Y CH. Real-time hand tracking using a mean-shiftembedded particle filter [J]. Pattern Recognition.2007,40:1958-1962
[45] Darroch J N, Lauritzen S L, Speed T P. Markov-field and log-linear models forcontingency tables [J]. Annals of Statistics,1980,8(3):522-539
[46] Edwards D. Introduction to Graphical Modelling [M]. Springer-Verlag,2ndedition，2000
[47] Gan ZH, JiangM. Articulated body tracking by immuneparticle filter [J]. LectureNotes in Computer Science,2006,4247:853.
[48] Clapp TC. Statisticalmethods for the communication data.[D]. Cambridne: University of Cambr idge,2000.
[49] Patrick P, Vermaak J. Data fusion for visual tracking with particle [J]. IEEE,2004,92(3):495-513
[50] Michaels Elad，Mariol A. T. Figueiredo，YIMa. On the role of sparse andredundant representations in image proeessing[J]. Proceedings of the IEEE，2010，98(6):972-982.
[51] A. Bruckstein，D.Donoho，and M.Elad. From sparse solutions of system ofequations to sparse modelling of signals and images[J].SIAM Rev.2009,5:34-81
[52] Otsu N. A threshold selection method from gray-level histogram [J]. IEEE Trans.on Systems, Man, and Cybernetics,1979,9(1):62-66
[53] Beghdadi A, Negrate AL, Lesegno P. V. Entropic thresholding using a blocksoure model [J]. Computer Models Image Process,1995,57:197-205
[54] Torr P.H.S, Zisserman A. Concerning Bayesian motion segmentation, modelaveraging, matching and the trifocal tensor [C]. Fifth European Conference onComputer Vision.Germany Freiburg: Springer Verlag,1998
[55] Changfa Xu, Minghong Pi, Chunxi Zhu, Guokuan Li. Interpolating-wavelet andits application [J]. In Proceedings of SPIE Vol,35-45
[56] L. H. Sun, Y. Z. Wang, Y. Q. He. Design and Implement of Infrared Small TargetReal-time Detection System Based on Pipeline Technology [C].27th InternationalCongress on High-Speed Photography and Photonics, Xi'an, China,2007. Proc. ofSPIE,6279-6283
[57]程建,周越,蔡念,杨杰.基于粒子滤波的红外目标跟踪[J].红外与毫米波学报,2006,02:113-117
[58]高建坡,王煜坚,杨浩,吴镇扬.以颜色和形状直方图为线索的粒子滤波人脸跟踪[J].中国图象图形学报,2007,03:466-473
[59]程水英,张剑云.粒子滤波评述[J].宇航学报,2008,04:1099-1111
[60]夏克寒,许化龙,张朴睿.粒子滤波的关键技术及应用[J].电光与控制,2005,06:1-4+19
[61]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策,2005,04:361-365+371
[62]李俊山,陈霞,李建华.不同地面背景下目标与背景红外辐射对比特性[J].红外与激光工程,2014,02:424-428
[63]陈晨.高超声速飞行器红外辐射特性分析[D].南京理工大学,2013
[64]李园园.目标红外辐射特性测试技术研究[D].南京理工大学,2013.
[65]李建勋,童中翔,王超哲,柴栋,张志波.飞机红外辐射特性分析与易损性评估[J].激光与红外,2013,02:180-185.
[66]杨词银,张建萍,曹立华.地基空间目标红外辐射特性测量技术[J].仪器仪表学报,2013,02:304-310.
[67]张辉，赵保军，唐林波等．变维卡尔曼滤波实现运动目标的跟踪[J]．光学学报，2010，30(5)：1291-1296
[68]严柏军，郑链，王克勇．基于不变矩特征匹配的快速目标检测算法[J]．红外技术，2001，23(6)：8-11
[69]梁敏，刘贵喜．基于自适应混合滤波的多目标跟踪算法[J]．光学学报，2010，30(9)：2554-2561
[70]邹江威，陈曾平．应用形态学与图像流法的空间小目标提取方法[J]．光电工程，2005，32(4)：13-15
[71]侯洁,辛云宏.基于高通滤波和图像增强的红外小目标检测方法[J].红外技术,2013,05:279-284.
[72]刘渭清.一种数字全通滤波器的最小二乘设计法[J].西南师范大学学报(自然科学版),2013,12:90-94
[73]蒋永馨，金俣欣，王孝通等．基于马尔可夫随机场的运动目标检测[J]．计算机工程与应用，2010，46(15)：182-184
[74]孔军，汤心溢，蒋敏等．采用尺度空间特征点匹配的红外目标跟踪[J]．红外与激光工程，2011,40(11)：2104-2109
[75]陶珂.小波去噪质量评价方法研究[D].中南大学,2012
[76] Ran He; Wei-Shi Zheng; Bao-Gang Hu; Xiang-Wei Kong.Two-StageNonnegative Sparse Representation for Large-Scale Face Recognition. NeuralNetworks and Learning Systems, IEEE Transactions,2013,24(1):35-46.
[77] Julazadeh, A.; Marsousi, M.; Alirezaie, J. Classification based on sparserepresentation and Euclidian distance[J]. Visual Communications and ImageProcessing (VCIP),2012IEEE,1-6.
[78]路锦正.基于稀疏表示的图像超分辨率重构技术研究[D].电子科技大学,2013.
[79]宋相法.基于稀疏表示和集成学习的若干分类问题研究[D].西安电子科技大学,2013.
[80] Yuanyuan Zuo; Bo Zhang. General image classification based on sparserepresentation[J]. Cognitive Informatics (ICCI),20109th IEEE InternationalConference on.2010,223-229.
[81]钟九生.基于稀疏表示的光学遥感影像超分辨率重建算法研究[D].南京师范大学,2013.
[82] Haoliang Yuan; Yang Lu; Lina Yang; Huiwu Luo; Yuan Yan Tang. Sparserepresentation using contextual information for hyperspectral image classification[J].Cybernetics (CYBCONF),2013IEEE International Conference on,2013,138-143.
[83]匡金骏.基于稀疏表示的图像分类与目标跟踪研究[D].重庆大学,2013.
[84] Samadi, S.; Cetin, M.; Masnadi-Shirazi, M.A.Sparse Signal Representation forComplex-Valued Imaging[J]. Digital Signal Processing Workshop and5th IEEESignal Processing Education,2009,4-7:365-370.
[85]汪桃林.基于稀疏表示的短语音说话人识别[D].电子科技大学,2013.
[86]刘钢.基于压缩感知和稀疏表示理论的图像去噪研究[D].电子科技大学,2013.
[87] Hongkai Xiong; Zhiming Pan; Xinwei Ye; Chang Wen Chen.SparseSpatio-Temporal Representation With Adaptive Regularized Dictionary Learning forLow Bit-Rate Video Coding[J]. Circuits and Systems for Video Technology, IEEETransactions on,2013.23(4):710-728.
[88]闫永帅.基于稀疏表示的高光谱遥感图像分类[D].电子科技大学,2013.
[89]王小娜.基于字典学习的图像稀疏表示及其在压缩感知中的应用[D].燕山大学,2013.
[90]兰诚栋,陈亮,卢涛.利用位置权重稀疏表示的人脸超分辨率算法[J].武汉大学学报(信息科学版),2013,01:27-30.
[91]孔繁锵,井庆丰,计振兴.图像压缩感知的自适应方向提升稀疏表示及重构算法[J].宇航学报,2013,01:121-127.
[92] Bar-Shalom Y, Fortmann T. Tracking and data association [M]．Acad. Press.New York.1988
[93] D. L. Alspach, H. W. Sorenson. Nonlinear Bayesian Estimation using GaussianSum Approximation [J]．IEEE Trans. on AC,1972,17(4):439-448
[94] Julier S, Uhlmann J. A new extension of the Kalman filter to nonlinear systems
[C]．Proc. SPIE,1997，3068:182-193
[95] S. Zhou, R. Chellappa, B. Moghaddam. Visual Tracking and Recognition usingAppearance-adaptive Models in Particle Filters [J]. IEEE Trans. Image Proc.2004,13(11):1491-1505
[96]胡诗国,方洋旺,蔡文新等.连续Markov跳变系统最优控制[J].控制与决策,2013,03:396-401.
[97]孟波,高存臣,考永贵等.一类时滞不确定随机Markovian跳变系统的滑模控制[J].中国海洋大学学报(自然科学版),2013,03:104-112.
[98]陈佳,方洋旺,楼顺天.随机跳变摄动系统的鲁棒H_∞控制[J].控制与决策,2013,02:279-284.
[99]叶丹,范泉涌.转移概率部分已知的Markov跳变系统鲁棒H_2控制[J].东北大学学报(自然科学版),2013,02:153-156.
[100]岳生伟,焦贤发.具有Markov跳的时滞中立系统的均方指数稳定[J].合肥工业大学学报(自然科学版),2013,04:477-482.
[101]刘晓华,吕娜. Markov跳变系统的输出反馈鲁棒预测控制[J].控制理论与应用,2013,11:.
[102]李科，徐克虎.特征融合和聚类核函数平滑采样优化的粒子滤波目标跟踪方法[J].计算机科学，2012.39（4）210-215.
[103] R. Edmondson, M. Rodgers, M. Banish. Ranganath.Single-Frame ImageProcessing Techniques for Low-SNR Infrared Imagery [J]. Infrared Technology andApplications. Proc. of SPIE.2008,6940:6943
[104] SHAN C F, TAN T N, WEI Y CH. Real-time hand tracking using amean-shift embedded particle filter [J]. Pattern Recognition.2007,40:1958-1962
[105] Yaghoobi M，Daudet L，Davies M. Parametric dictionary design for sparsecoding[J].IEEE Transactions on Signal Processing，2009，57（12）：4800-4810.
[106]刘伟.基于人工神经网络的遥感影像模式分类研究[D].沈阳工业大学,2005
[107]单建丰,郁道银,陈晓冬等.基于模糊聚类和C-V模型的水平集方法图像分割[J].中国介入影像与治疗学,2013,08:498-502.
[108]段军,位保振.改进的模糊C_均值聚类医学图像分割算法[J].微型机与应用,2013,16:36-37+41.
[109]曲福恒.一类模糊聚类算法研究及其应用[D]:[博士学位论文].吉林：吉林大学，2009.
[110]陈厦,方方,胡战利.模糊聚类算法综述[J].生命科学仪器,2013,06:33-37.
[111]倪迎花.图像处理中的模糊算法及实现分析[J].信息与电脑(理论版),2013,08:53-54.
[112] Esmin, A.A.A.; Coelho, R.A. Consensus Clustering Based on Particle SwarmOptimization Algorithm[J]. Systems, Man, and Cybernetics (SMC),2013(10):2280-2285.
[113]朱林,雷景生,毕忠勤等.一种基于数据流的软子空间聚类算法[J].软件学报,2013,11:2610-2627.
[114]刘建生,乔尚平,匡奕群.基于差分粒子群和模糊聚类的彩色图像分割算法[J].江西理工大学学报,2013,05:66-71.
[115]燕蜻.混合型多属性群决策方法研究[D].山西大学,2012
[116]陈晓红,杨立.基于多属性群决策方法的障碍诊断模型[J].科研管理,2013,03:115-121
[117]谭旭,毛太田,张少丁等.基于粒计算的多属性群决策分析[J].四川大学学报(工程科学版),2013,04:140-148
[118]刘翔.突发危机事件应急群决策支持系统的设计与实现[D].复旦大学,2012.
[119]刘兮.基于多粒度二元语义信息的几类信息集结算子及其在多属性群决策中的应用[D].安徽大学,2012
[120]姚成玉,陈东宁,冯中魁等.基于贝叶斯网络和理想解动态群决策的故障诊断方法[J].中国机械工程,2013,16:2235-2241
[121]石福丽,许永平,杨峰.考虑专家偏好关联的群决策方法及其应用[J].控制与决策,2013,03:391-395+401
[122]蔡付龄,廖貅武,杨娜.基于案例信息的多准则群决策分类方法[J].管理科学学报,2013,02:22-32
[123]王文全,黄胜,王超等.舰船通道布置方案的直觉模糊多属性群决策方法[J].上海交通大学学报,2013,06:894-899
[124]戚筱雯,梁昌勇,黄永青等.基于混合型评价矩阵的多属性群决策方法[J].系统工程理论与实践,2013,02:473-481
[125]燕蜻,梁吉业.混合多属性群决策中的群体一致性分析方法[J].中国管理科学,2011,06:133-140.
[126] Xilinx inc, Virtex-5FPGA Integrated Endpoint Block for PCI ExpressDesigns[EB/OL]. UG197, July22,2009. http://www.xilinx.com.
[127] Xilinx inc, Virtex-5FPGA Integrated Endpoint Block for PCI ExpressDesigns: DDR2SDRAM DMA Initiator Demonstration Platform[EB/OL]. XAPP859,July31,2008. http://www.xilinx.com.
[128]孟思远,刘建,奚杰等.一种多核DSP的动态存储器分配设计方法[J].微电子学与计算机,2008,11:21-24.
[129]李鑫,姜明.多核DSP高速实时信号处理系统设计[J].光学技术,2012,01:116-120.
[130]徐力,王沁,史少波.一种基于异构多核DSP的IEEE802.11a接收端基带处理的研究和实现[J].计算机应用研究,2012,01:241-245.
[131]徐力,史少波,王沁.面向SDR应用的多核DSP低功耗设计[J].电子科技大学学报,2012,01:136-141.
[132]孙科林,周维超,吴钦章等.光纤实时传输的多核DSP图像处理系统(英文)[J].光电工程,2012,04:136-144.
[133]汪东,陈书明.一种面向异构多核DSP的数据流传输控制引擎设计与评测[J].国防科技大学学报,2009,05:19-23
[134]敖天勇,陈杰,刘建,许汉荆,奚杰,张伟风.基于CostarⅡ的异构多核DSP设计与实现[J].微电子学与计算机,2010,03:59-62.
[135]刘建,陈杰,敖天勇,许汉荆.片上异构多核DSP同步与通信的实现[J].电子科技大学学报,2010,04:528-531+536.
[136]方兴,陈书明.支持细粒度并行性开发的多核DSP快速核间通信机制[J].计算机工程与科学,2009,04:130-133.
[137]辛熠,杨瑞民.多核DSP编程技术研究[J].电子测试,2011,09:81-85.
[138]吴家铸,田希,刘继福等.计算高度密集型应用在异构多核DSP上的运行方法研究[J].微电子学与计算机,2011,05:45-47+52.
[139]陈书明,汪东,陈小文等.一种面向多核DSP的小容量紧耦合快速共享数据池[J].计算机学报,2008,10:1737-1744.
[140]汪东,陈书明. DSCF:一种面向共享存储多核DSP的数据流分簇前向技术[J].计算机研究与发展,2008,08:1446-1453.
[141]单祥茹. TI多核DSP再出新品高性能计算时代即将到来[J].中国电子商情(基础电子),2012,Z1:29-30.
[142]赵海霖，韦文，汤俊．多DSP平台下性能预测算法[J]：信息与电子工程，2011，9(4)，472-477
[143]尹九阳，方滨兴，张宏莉．基于单程序多数据流并行软件的性能预测法[J]：清华大学学报(自然科学版)，2005，45(4)，509-512
[144]汪向阳,张云峰.基于双DSP的图像处理系统硬件设计[J].仪器仪表学报,2005,S1:646-647+656.
[145]谢明,吉书鹏,段哲民.一种基于高速双DSP的柔性机载实时图像跟踪系统研究[J].红外技术,2005,02:99-104.
[146]田小永,古志民.基于TMS320C55x DSP的代码优化[J].计算机应用,2003,01:56-59.
[147]邱春武.基于并行处理单元的代码优化方法研究[D]：[硕士学位论文].北京：北京邮电大学,2008.
[148]施虹宇. TMS320DM642上的代码优化研究[D]：[硕士学位论文].北京：北京邮电大学,2013.
[149]源江科技.VC编程技巧[M].上海科学普及出版社.2002.
[150]胡海生，李升亮.Visual C++6.0编程学习捷径[M].清华大学出版社.2003.
[151]王诚，薛小刚，钟信潮.FPGA/CPLD设计工具Xilinx ISE使用详解[M].人民邮电出版社.2004:95-159.
[152] Paragios N, Rsmesh V. A MRF-based approach for real-time subwaymonitoring [C]. Processing of Computer Vision and Pattern Recognition,2001:295-313
[153]郭伟．复杂背景下红外目标检测与跟踪[D]：[博士学位论文]．西安：西安电子科技大学，2008
[154] Alexis P.Tzannes, Dana H.Brooks. Point Target Detection in IR ImageSequences. A Hypothesis-testing Approach Based on Target and Clutter TemporalProfile Modeling [J]. Opt. Eng,2000,39(8).2270-2278
[155] Christopher WK. Infrared target detection: signal and noise sensitivityanalysis thesis [C].SPIE,1989.AD207-215
[156] Nishiguchi K, Kobayashi M, Ichikawa A.Small Target Detection from ImageSequences Using Recursive Max Filter[C].SPIE.1995.153-165
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |