计算机科学 ›› 2018, Vol. 45 ›› Issue (7): 252-258.doi: 10.11896/j.issn.1002-137X.2018.07.044
吴玮,郑娟毅,杜乐
WU Wei, ZHENG Juan-yi ,DU Le
摘要: 传统的Camshift算法以颜色直方图为特征对目标进行跟踪,对刚性目标的跟踪具有较强的鲁棒性。当目标受到颜色相近的干扰物干扰或者部分遮挡时,其跟踪效果和准确度不太理想。为此,提出一种多特征融合的Camshift目标跟踪算法。首先,对目标的颜色特征、边缘特征和空间信息进行提取和处理,得到颜色空间直方图和空间边缘方向直方图;然后,分别在Camshift算法框架下得到目标匹配中心位置,采用每一帧图像的相似度向量得到权值系数,通过自适应加权融合的方法得到最优中心位置。实验结果表明,相较于传统的Camshift目标跟踪算法和改进的复杂特征融合的Meanshift算法,所提方法能够更有效地克服颜色干扰、目标重叠遮挡对跟踪效果的影响,避免了目标在跟踪过程中丢失的问题,突破了传统方法的局限性。
中图分类号:
[1]YILMAZ A,JAVED O,SHAH M.Object tracking:A survey[J].ACM Computing Surveys,2006,38(4):1-45. [2]杨杰,张翔.视频目标检测和跟踪及其应用[M].上海:上海交通大学出版社,2012. [3]BRADSKI G R.Computer vision face tracking for use in a perceptual user interface[C]∥IEEE Workshop on Applications of Computer Vision.1998:214-219. [4]SIRIKUNTAMAT N,SATOH S,CHALIDABHONGSE T H.Vehicle tracking in low hue contrast based on CAMShift and background subtraction[C]∥International Joint Conference on Computer Science and Software Engineering.IEEE,2015:58-62. [5]XIU C,BA F.Target tracking based on the improved Camshift method[C]∥Chinese Control and Decision Conference.IEEE,2016:3600-3604. [6]ZHU Y,XU B.A robust object tracking method combiningshape descriptor and adaptive background camshift[C]∥International Conference on Natural Computation.IEEE,2015:930-933. [7]HAN Z,ZHANG R,WEN L,et al.Moving Object TrackingMethod Based on Improved Camshift Algorithm[C]∥International Conference on Industrial Informatics-Computing Technology,Intelligent Technology,Industrial Information Integration.IEEE,2016:91-95. [8]YANG X,FEI S M,LI G,et al.Improved Mean shift tracking algorithm based on complicated feature fusion.Control and Decision,2014,29(7):1297-1300.(in Chinese) 杨欣,费树岷,李刚,等.基于复杂特征融合的改进Meanshift目标跟踪[J].控制与决策,2014,29(7):1297-1300. [9]LIU W J,ZHANG Y J.Edge-color-histogram and Kalman filter-based real-time object tracking .Journal of Tsinghua University Science and Technology,2008,48(7):1104-1107.(in Chinese) 刘惟锦,章毓晋.基于Kalman滤波和边缘直方图的实时目标跟踪[J].清华大学学报自然科学版,2008,48(7):1104-1107. [10]YIMIT A,HAGIHARA Y,MIYOSHI T,et al.2-D direction histogram based entropic thresholding[J].Neurocomputing,2013,120(10):287-297. [11]SHEN X M,ZHOU J X,XU T W.Minority Costume Image Retrieval by Fusion of Color Histogram and Edge Orientation Histogram∥International Conference on Computer and Information Science.IEEE,2016:1-7. [12]KATJA N,ESTHER K M.Object tracking with an adaptive color-based particle filter[C]∥Proceedings of the 24th DAGM Symposium on Pattern Recognition.2002:353-360. [13]NAVNEET D,BILL T.Histograms of oriented gradients for human detection[C]∥Computer Society Conference on Computer Vision & Pattern Recognition.2005:886-893. [14]HAN B,LUO Y.Accurate face detection by combining multiple classifiers using locally assembled histograms of orientedgra-dients[C]∥International Conference on Audio.2012:106-111. [15]ZHANG Y K,LI Y F,SUN J G.Image retrieval based on multi-feature fusion[J].Journal of Computer Applications,2015,35(2):495-498.(in Chinese) 张永库,李云峰,孙劲光.基于多特征融合的图像检索[J].计算机应用,2015,35(2):495-498. [16]QU J B.Convergence analysis of kernel based Camshift algorithm.Journal on Numerical Methods and Computer Applications,2011,32(2):148-158.(in Chinese) 曲巨宝.基于核函数的Camshift算法的收敛性分析[J].数值计算与计算机应用,2011,32(2):148-158. [17]XU W J,HOU Z Q,YU W S, et al. Fusing mulit feature object tracking algorithm based on background weighting .Journal of Air Fouce Engineering Universitu(Natrnal Science Edition),2015,16(3):71-76.(in Chinese) 许婉君,侯志强,余旺盛,等.背景加权的多特征融合目标跟踪算法[J].空军工程大学学报(自然科学版),2015,16(3):71-76. |
[1] | 张源, 康乐, 宫朝辉, 张志鸿. 基于Bi-LSTM的期货市场关联交易行为检测方法 Related Transaction Behavior Detection in Futures Market Based on Bi-LSTM 计算机科学, 2022, 49(7): 31-39. https://doi.org/10.11896/jsjkx.210400304 |
[2] | 程成, 降爱莲. 基于多路径特征提取的实时语义分割方法 Real-time Semantic Segmentation Method Based on Multi-path Feature Extraction 计算机科学, 2022, 49(7): 120-126. https://doi.org/10.11896/jsjkx.210500157 |
[3] | 郁舒昊, 周辉, 叶春杨, 王太正. SDFA:基于多特征融合的船舶轨迹聚类方法研究 SDFA:Study on Ship Trajectory Clustering Method Based on Multi-feature Fusion 计算机科学, 2022, 49(6A): 256-260. https://doi.org/10.11896/jsjkx.211100253 |
[4] | 李鹏祖, 李瑶, Ibegbu Nnamdi JULIAN, 孙超, 郭浩, 陈俊杰. 基于多特征融合的重叠组套索脑功能超网络构建及分类 Construction and Classification of Brain Function Hypernetwork Based on Overlapping Group Lasso with Multi-feature Fusion 计算机科学, 2022, 49(5): 206-211. https://doi.org/10.11896/jsjkx.210300049 |
[5] | 瞿中, 陈雯. 基于空洞卷积和多特征融合的混凝土路面裂缝检测 Concrete Pavement Crack Detection Based on Dilated Convolution and Multi-features Fusion 计算机科学, 2022, 49(3): 192-196. https://doi.org/10.11896/jsjkx.210100164 |
[6] | 牛富生, 郭延哺, 李维华, 刘文洋. 基于序列特征融合的蛋白质可溶性预测 Protein Solubility Prediction Based on Sequence Feature Fusion 计算机科学, 2022, 49(1): 285-291. https://doi.org/10.11896/jsjkx.201100117 |
[7] | 张曼, 李杰, 朱新忠, 沈霁, 成昊天. 基于改进DCGAN算法的遥感数据集增广方法 Augmentation Technology of Remote Sensing Dataset Based on Improved DCGAN Algorithm 计算机科学, 2021, 48(6A): 80-84. https://doi.org/10.11896/jsjkx.200700185 |
[8] | 吕金娜, 邢春玉, 李莉. 基于多特征融合的细粒度视频人物关系抽取 Video Character Relation Extraction Based on Multi-feature Fusion and Fine-granularity Analysis 计算机科学, 2021, 48(4): 117-122. https://doi.org/10.11896/jsjkx.200800160 |
[9] | 栾晓, 李晓双. 基于多特征融合的人脸活体检测算法 Face Anti-spoofing Algorithm Based on Multi-feature Fusion 计算机科学, 2021, 48(11A): 409-415. https://doi.org/10.11896/jsjkx.210100181 |
[10] | 原晓佩, 陈小锋, 廉明. 基于Haar-like和LBP的多特征融合目标检测算法 Improved Multi-feature Fusion Algorithm for Target Detection Based on Haar-like and LBP 计算机科学, 2021, 48(11): 219-225. https://doi.org/10.11896/jsjkx.201100174 |
[11] | 吴宏涛, 刘力源, 孟颖, 荣亚鹏, 李路凯. 动态多特征融合的道路遗洒物威胁度分析方法 Novel Threat Degree Analysis Method for Scattered ObJects in Road Traffic Based on Dynamic Multi-feature Fusion 计算机科学, 2020, 47(6A): 196-205. https://doi.org/10.11896/JsJkx.190900066 |
[12] | 胡宇佳, 甘伟, 朱敏. 基于多特征融合的增强子-启动子相互作用预测综述 Enhancer-Promoter Interaction Prediction Based on Multi-feature Fusion 计算机科学, 2020, 47(5): 64-71. https://doi.org/10.11896/jsjkx.191100027 |
[13] | 金堃, 陈少昌. 步态识别现状与发展 Status and Development of Gait Recognition 计算机科学, 2019, 46(6A): 30-34. |
[14] | 吴璠, 李寿山, 周国栋. 基于LSTM和多特征组合的电影评论专业程度分类 Movie Review Professionalism Classification Using LSTM and Features Fusion 计算机科学, 2019, 46(6A): 74-79. |
[15] | 王晓, 邹泽伟, 李勃勃, 王静. 基于多特征融合的彩色图像声呐目标检测 Target Detection in Colorful Imaging Sonar Based on Multi-feature Fusion 计算机科学, 2019, 46(6A): 177-181. |
|