摘要
针对图像跟踪领域中因遮挡产生的漂移问题,提出一种基于直方图比的背景加权的Mean Shift算法和Kalman预测滤波器融合的方法。本文方法通过改进目标模型来优化Bhattacharyya系数值,增大目标正常跟踪状态下和遮挡状态下Bhattacharyya系数的差值,提高遮挡判定的有效性,进而提高遮挡时的跟踪性能。通过实验证明,基于直方图比的背景加权的Mean Shift算法和Kalman预测滤波器融合的方法可有效解决遮挡跟踪问题。
Aiming at the drift caused by occlusion in the field of image tracking,a fusion method based on Mean Shift and histogram ratio background weighted algorithm and Kalman predictive filter was proposed.The Bhattacharyya coefficients were optimized by improving the target model.The difference between the Bhattacharyya coefficients in the normal tracking state and in the occlusion state was increased.The effectiveness of occlusion judgement was improved so the performance of tracking in occlusion was improved.The experiments show that this method can effectively solve the occlusion tracking problem.
作者
吴水琴
毛耀
刘琼
李志俊
WU Shui-qin;MAO Yao;LIU Qiong;LI Zhi-jun(Key Laboratory of Optical Engineering,Chinese Academy of Science,Chengdu 610209,China;Institute of Optics and Electronics,Chinese Academy of Science,Chengdu 610209,China;Chinese Academy of Science,Beijing 101400,China)
出处
《液晶与显示》
CAS
CSCD
北大核心
2019年第2期188-193,共6页
Chinese Journal of Liquid Crystals and Displays
