Mobile robot human tracking using hierarchical features
JIA Song min, LU Ying bin, WANG Li jia, LI Xiu zhi, XU Tao
College of Electronic and Control Engineering, Beijing University of Technology, Beijing 100124, China；
Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing 100124,China；
Engineering Research Center of Digital Community, Ministry of Education, Beijing 100124, China
First, the outer orientation of pedestrian target was obtained through head-shoulder or color features. Then the improved SIFT features of the pixels were extracted to locate the target accurately in the coarse location area. The pedestrian scale change problem was solved based on SIFT features to figure out the scale variation of the pedestrian. Simultaneously, the updating mechanism of SIFT features template library was introduced into feature reservation priority level to solve the problem of temporary occlusion and deformation of the target. Aiming at the adaptive elliptic kernel function of the traditional Cam-Shift algorithm, the variablebandwidth and orientation ellipse kernel was combined with scale variations of the human and the Epanechnikov function, which reduced the interfere of background. Furthermore, the searching area for Harris operator was limited by the outler coarse location results, which improved the real-time performance of SIFT feature matching. The experimental results indicate that the proposed mobie robot human tracking algorithm can accomplish human tracking under conditions of target scale variations, temporary occlusion and deformation．
JIA Song min, LU Ying bin, WANG Li jia, LI Xiu zhi, XU Tao. Mobile robot human tracking using hierarchical features. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2016, 50(9): 1677-1683.
［1］ YOSHIMI T, NISHIYAMA M, SONOURA T, et al. Development of a person following robot with vision based target detection ［C］ ∥ Proceedings of the 2006 IROS International Conference on Intelligent Robots and Systems. Beijing: IROS, 2006: 5286-5291.
［2］ YUN W H, KIM D, LEE J. Person following with obstacle avoidance based on multilayered mean shift and force field method ［C］ ∥ Proceedings of the 2010 SMC International Conference on Systems, Man and Cybernetics. Turkey: SMC,2010: 3813-3816.
［3］ 曹腾,项志宇,刘济林.基于视差空间V截距的障碍物检测自主机器人［J］.浙江大学学报：工学版,2015,49(3)： 409-414．
CAO Teng, XIANG Zhiyu, LIU Jilin. Obstacle detection based on Vintercept in disparity space ［J］. Journal of Zhejiang University: Engineering Science, 2015, 49(3): 409414.
［4］ 邱雪娜.基于视觉的运动目标跟踪算法及其在移动机器人中的应用 ［D］. 上海：华东理工大学,2011．
QIU Xuena. Visiobased moving object tracking and its application to mobile robot ［D］ Shanghai: South China University of Technology, 2011．
［5］ 李杏华,刘硕. 基于计算机视觉的人体跟踪系统研究［D］.天津：天津大学,2008．
LI Xinghua, LIU Shuo. Research on computer vision based human tracking system ［D］. Tianjin: Tianjin University, 2008．
［6］ 王丽佳,贾松敏,王爽,等.采用改进的MeanShift算法的移动机器人行人跟踪［J］.光学紧密工程,2013,21(9)： 2364-2370．
WANG Lijia, JIA Songmin, WANG Shuang, et al. Person tracking of mobile robot using improved MeanShift ［J］. Optics and Precision Engineering, 2013, 21(9): 2364-2370．
［7］ 薛陈,朱明,陈爱华.鲁邦的基于改进MeanShift的目标跟踪［J］.光学精密工程,2010,18(1): 234-239．
XUE Chai,ZHU Ming,CHEN Aihua. Robust object tracking based on improved MeanShift algorithm ［J］. Optics and Precision Engineering, 2010, 18(1): 234-239.
［8］ LI J D. Research on camerabased human body tracking using improved camshift algorithm ［J］. International Journal on Smart Sensing and Intelligent Systems, 2015, 8(2): 1104-1122．
［9］ SUGA A, FUKUDA K, TAKIGUCHI T, et al. Object recognition and segmentation using SIFT and graph cuts ［C］ ∥ Proceedings of the 19th ICPR International Conference on Pattern Recognition. Tampa: ICPR, 2008:14．
［10］ LOWE D G. Distinctive image features from scaleinvariant keypoints ［J］. International Journal of Computer Vision, 2004, 60(2): 91-110．
［11］ 王丽佳,贾松敏,李秀智.多特征提取的双目机器人目标跟踪［J］.控制与决策,2013,28(10): 1568-1572．
WANG Lijia, JIA Songmin, LI Xiuzhi. Person following of binocular robot by extracting multiple features ［J］. Control and Decision, 2013, 28(10): 1568-1572.
［12］ NAUSHAD ALI M M, ABDULLAHALWADUD M, LEE S L. Moving object detection and tracking using particle filter ［J］. Applied Mechanics and Materials 2013, 321324(2): 1200-1204.
［13］ SONGMIN J, WANG S. Human tracking system based on adaptive multifeature meanshift for robot under the doublelayer locating mechanism ［J］. Advanced Robotics, 2014, 25(5): 564-577．
［14］ 蔺海峰, 马宇峰, 宋涛. 基于 SIFT 特征目标跟踪算法研究［J］. 自动化学报, 2010, 36(8): 1204-1208．
LIN Haifeng, MA Yufeng, Research on object tracking algorithm based on SIFT ［J］. Acta Automatica Sinica, 2010, 36(8): 1204-1208.
［15］ 徐进,沈敏一,杨力,等. 基于双目光束法平差的机器人定位与地形拼接［J］. 浙江大学学报：工学版, 2011, 45(7): 1141-1146.
XU Jin, SHEN Minyi, YANG Li, et al. Binocular bundle adjustment based localization and terrain stitching for robot ［J］. Journal of Zhejiang University: Engineering Science, 2011, 45(7): 1141-1146．
［16］ 董蓉, 李勃. 基于 SIFT 特征的目标多自由度 MeanShift 跟踪算法［J］.控制与决策, 2012,27(3)： 399-407．
DONG Rong, LI Bo. Multidegreeoffreedom MeanShift tracking algorithm based on SIFT feature ［J］. Control and Decision, 2012, 27(3): 399-407．
［17］ 王丽佳,贾松敏. 基于改进在线多示例学习算法的机器人目标跟踪［J］.自动化学报,2014,40(12): 2916-2925．
WANG Lijia, JIA Songmin. Person following for mobile robot using improved multiple instance learning ［J］. Acta Automatica Sinica, 2014, 40(12): 2916-2925．
