Adaptive mixture Gaussian model based target detection using
omnidirectional vision
LIU Shi-rong 1, WANG Kai1, QIU Xue-na2
1. Institute of Automation, Hangzhou Dianzi University, Hangzhou 310018, China; 2. Institute of
Automation, East China University of Science and Technology, Shanghai 200237, China
Considering the advantage of omnidirectional vision sensor with large fieldofview, an omnidirectional vision target detection system was presented based on the adaptive mixture Gaussian model. A Hough transform algorithm was adopted to detect the center of the scene,and the omnidirectional image was transformed around the center. An adaptively updated background was modeled by the mixture Gaussian model for the transformed images. The moving object can be effectively segmented by the foreground segmentation method. The realtime performance of target detection can be improved by adjusting the sample frequency during image transforming and mixture Gaussian modeling. Experimental results demonstrate that the system can effectively detect the moving object under complex environments, and has more accuracy and robustness.
［1］ SHEIKN Y, SHAH M. Baysian modeling of dynamic scenes for object detection ［J］. IEEE Transportation on Pattern Analysis and Machine Intelligence, 2005, 27(11): 17781792.
［2］ BLACK J, ELLIS T. Multi camera image tracking ［J］. Image and Vision Computing, 2006, 24(11): 12561267.
［3］ LIU Hong, PI Wenkai. ZHA Hongbin. Motion detection for multiple moving targets by using an omnidirectional camera ［C］∥ Proceedings of the 2003 IEEE International Conference on Robotics, Intelligent Systems and Signal Processing. Changsha, China: IEEE, 2003: 422426.
［4］ ZHAN Mingzhu, CAO Huanrong. A new method of circle’s center and radius detection in image processing ［C］∥ Proceedings of the IEEE International on Automation and Logistics. Qingdao, China: IEEE, 2008: 22392242.
［5］ 原新,王亮,朱齐丹.基于全向视觉传感器的图像解算方法研究［J］.哈尔滨工业大学学报,2006,38(12)：21582161.
YUAN Xin,WANG Liang,ZHU Qidan. Image calculation research based on omnidirectional camera ［J］. Journal of Harbin Institute of Techeology, 2006,38(12)：21582161.
［6］ 席志红,吴自新,张曙.基于全视觉图像中心的二次逆向快速解算算法［J］.哈尔滨工程大学学报,2007,28(4)： 465468.
XI Zhihong,WU Zixin,ZHANG Shu. Quick double reverse algorithm for omnidirectional images based on the center of imaging ［J］. Journal of Harbin Engineering University, 2007,28(4): 465468.
［7］ 朱云芳,王贻术,杜歆.静态环境中基于光流的障碍物检测［J］.浙江大学学报:工学版,2008,42（6）： 923926.
ZHU Yunfang,WANG Yishu,DU Xin. Optical flow based obstacle detection in static environment ［J］.Journal of Zhejiang University: Engineering Science, 2008, 42（6）: 923926.
［8］ ANDERSON C H, BURT P J, VANDERWALL G S. Change detection and tracking using pyramid transform techniques ［C］∥ Proceedings of SPIEIntelligent Robots and Computer Vision. Boston: SPIE, 1985: 7278.
［9］ MITTAL A, PARAGIOS N. Motionbased background subtraction using adaptive kernel density estimation ［C］∥ Proceeding of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC, USA: IEEE, 2004: 302309.
［10］ DARSHYANG L. Effective Gaussian mixture learning for video background subtraction ［J］. IEEE Transactions on Pattern and Machine Intelligence, 2005, 27(5): 827832.
［11］ 王长军,朱善安.基于统计模型和活动轮廓的运动目标检测与跟踪［J］.浙江大学学报:工学版,2006,40（2）：249253.
WANG Changjun,ZHU Shanan. Motion detection and object tracking based on statistical model and active contour ［J］. Journal of Zhejiang University: Engineering Science, 2006, 40(2): 249253.
