双边界限定下的运动目标跟踪方法
|
摘要
针对目标跟踪领域中复杂背景下的目标提取、目标与目标以及目标与背景之间的相互遮挡、阴影的处理、目标跟踪的实时性及鲁棒性等问题,提出一种双边界限定下的运动目标跟踪算法。利用金字塔光流法对目标关键点进行长时间无模型跟踪,并将其与检测到的前景关键点进行不重复地融合。根据融合得到的关键点,评估尺度因子及旋转因子,并利用全部前景关键点对目标中心点进行概率性表决。再利用双边界限定方法使边界上关键点对目标进行可靠描述,提升系统的鲁棒性。
Aiming at problems of target extraction in complex background,mutual occlusion between two targets,target and background,processing of shadow,real-time and robustness of target tracking,and so on,in target tracking field,a moving target tracking algorithm based on double boundary limitation is proposed. The pyramid optical flow method is used to track target key points for a long time without model,and it is fused with the detected foreground key points without repetition. The scale factor and twiddle factor are evaluated,according to the key points obtained by fusion,and the target center points are probabilistically voted by using all the foreground key points. The double boundary definition method is used to make the key points on the boundary reliably describe the target and improve the robustness of the system.
Aiming at problems of target extraction in complex background,mutual occlusion between two targets,target and background,processing of shadow,real-time and robustness of target tracking,and so on,in target tracking field,a moving target tracking algorithm based on double boundary limitation is proposed. The pyramid optical flow method is used to track target key points for a long time without model,and it is fused with the detected foreground key points without repetition. The scale factor and twiddle factor are evaluated,according to the key points obtained by fusion,and the target center points are probabilistically voted by using all the foreground key points. The double boundary definition method is used to make the key points on the boundary reliably describe the target and improve the robustness of the system.
引文
[1]焦再强.基于Zynq—7000的嵌入式数字图像处理系统设计与实现[D].太原:太原理工大学,2015.
[2]陈炳权,刘宏立,孟凡斌.数字图像处理技术的现状及其发展方向[J].吉首大学学报:自然科学版,2009,30(1):63-70.
[3] Scholz M,Konen W,Tombrock S,et al. Development of an endoscopic navigation system based on digital image processing[J].Computer Aided Surgery,2015,3(3):134-143.
[4]郭钊.嵌入式图像并行处理系统的研究与应用[D].兰州:兰州理工大学,2014.
[5]赵强.基于视觉信息的移动机器人目标识别算法研究[D].济南:山东大学,2012.
[6]刘洪伟.基于单目视觉的移动机器人目标识别与定位[D].济南:山东大学,2011.
[7] Ojha S,Sakhare S. Image processing techniques for object tracking in video surveillance:A survey[C]∥International Conference on Pervasive Computing,IEEE,2015:1-6.
[8] Adam A,Rivlin E,Shimshoni I. Robust fragments-based tracking using the integral histogram[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2006:798-805.
[9] Hua G,Wu Y. Measurement integration under inconsistency for robust tracking[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2006:650-657.
[10] Grabner M,Grabner H,Bischof H. Learning features for tracking[C]∥IEEE Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2007:1-8.
[11] Hare S. Efficient online structured output learning for keypointbased object tracking[C]∥IEEE Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2012:1894-1901.
[12] Park S Y,Choi S I,Moon J,et al. Real-time 3D registration of stereo-vision-based range images using GPU[C]∥Applications of Computer Vision Workshop,2009:1-6.
[13] Mo H,Information S O. Survey of clustering algorithms for big data[J]. Computer Science,2016,43(s1):380-383.
[14] Donoho D L,Elad M,Temlyakov V N. Stable recovery of sparse overcomplete representations in the presence of noise[J]. IEEE Transactions on Information Theory,2005,52(1):6-18.
[15] Friedman J H. Greedy function approximation:A gradient boosting machine[J]. Annals of Statistics,2001,29(5):1189-1232.
[16]周志华.机器学习[M].北京:清华大学出版社,2016.
[17]杨康,刘桂华,徐锋.基于BCH和HOG的Mean Shift跟踪算法[J].传感器与微系统,2018,37(7):138-141.
[18] Zhang K,Zhang L,Yang M H. Real-time compressive tracking[C]∥European Conference on Computer Vision,SpringerVerlag,2012:864-877.
[19] Kalal Z,Mikolajczyk K,Matas J. Tracking-learning-detection[J].IEEE Transon Pattern Anal Mach Intell,2012,34(7):1409-1422.
[2]陈炳权,刘宏立,孟凡斌.数字图像处理技术的现状及其发展方向[J].吉首大学学报:自然科学版,2009,30(1):63-70.
[3] Scholz M,Konen W,Tombrock S,et al. Development of an endoscopic navigation system based on digital image processing[J].Computer Aided Surgery,2015,3(3):134-143.
[4]郭钊.嵌入式图像并行处理系统的研究与应用[D].兰州:兰州理工大学,2014.
[5]赵强.基于视觉信息的移动机器人目标识别算法研究[D].济南:山东大学,2012.
[6]刘洪伟.基于单目视觉的移动机器人目标识别与定位[D].济南:山东大学,2011.
[7] Ojha S,Sakhare S. Image processing techniques for object tracking in video surveillance:A survey[C]∥International Conference on Pervasive Computing,IEEE,2015:1-6.
[8] Adam A,Rivlin E,Shimshoni I. Robust fragments-based tracking using the integral histogram[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2006:798-805.
[9] Hua G,Wu Y. Measurement integration under inconsistency for robust tracking[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2006:650-657.
[10] Grabner M,Grabner H,Bischof H. Learning features for tracking[C]∥IEEE Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2007:1-8.
[11] Hare S. Efficient online structured output learning for keypointbased object tracking[C]∥IEEE Conference on Computer Vision and Pattern Recognition,IEEE Computer Society,2012:1894-1901.
[12] Park S Y,Choi S I,Moon J,et al. Real-time 3D registration of stereo-vision-based range images using GPU[C]∥Applications of Computer Vision Workshop,2009:1-6.
[13] Mo H,Information S O. Survey of clustering algorithms for big data[J]. Computer Science,2016,43(s1):380-383.
[14] Donoho D L,Elad M,Temlyakov V N. Stable recovery of sparse overcomplete representations in the presence of noise[J]. IEEE Transactions on Information Theory,2005,52(1):6-18.
[15] Friedman J H. Greedy function approximation:A gradient boosting machine[J]. Annals of Statistics,2001,29(5):1189-1232.
[16]周志华.机器学习[M].北京:清华大学出版社,2016.
[17]杨康,刘桂华,徐锋.基于BCH和HOG的Mean Shift跟踪算法[J].传感器与微系统,2018,37(7):138-141.
[18] Zhang K,Zhang L,Yang M H. Real-time compressive tracking[C]∥European Conference on Computer Vision,SpringerVerlag,2012:864-877.
[19] Kalal Z,Mikolajczyk K,Matas J. Tracking-learning-detection[J].IEEE Transon Pattern Anal Mach Intell,2012,34(7):1409-1422.