融合颜色和梯度直方图的免疫粒子滤波算法
|
摘要
针对标准粒子滤波在目标与背景颜色相似和存在遮挡时易出现跟踪丢失的问题,提出一种融合颜色和梯度直方图信息的免疫粒子滤波算法.颜色直方图能对目标进行全局描述,梯度方向直方图含有相应的结构信息,通过融合多特征信息,使得单纯依靠颜色特征不能很好适应环境变化的情况得到了有效改善.同时通过加入免疫粒子算法,提高了粒子的多样性,尤其是在目标被遮挡时,减少了"样贫"的影响.实验结果显示,采用综合直方图特征的改进粒子滤波算法在背景颜色干扰及目标被遮挡时仍能稳定地跟踪目标,具有较强的鲁棒性.
Focusing on fails to track with standard particle filters when object and background have similar color and object is occluded,a new algorithm of immune particle filter object tracking based on histograms of color and oriented gradient is proposed.Color histogram is the global description of targets in color image,histogram of oriented gradient contains some construction information.The gradient feature is added,the poor performance of only one feature is improved when the environment changes.In addition,the proposed immune optimization algorithm improved the diversity of particles,especially the decline of the impact of sample impoverishment when occlusion occurs.Experimental results show that the proposed method is able to track the object stably when background color interference and object is occluded,which has a strong robustness.
Focusing on fails to track with standard particle filters when object and background have similar color and object is occluded,a new algorithm of immune particle filter object tracking based on histograms of color and oriented gradient is proposed.Color histogram is the global description of targets in color image,histogram of oriented gradient contains some construction information.The gradient feature is added,the poor performance of only one feature is improved when the environment changes.In addition,the proposed immune optimization algorithm improved the diversity of particles,especially the decline of the impact of sample impoverishment when occlusion occurs.Experimental results show that the proposed method is able to track the object stably when background color interference and object is occluded,which has a strong robustness.
引文
[1]Cao Yi-qin,Xiao Jin-sheng,Huang Xiao-sheng.Mean-shift tracking algorithm based on boundary detection and sandbags kernel func tion[J].Application Research of Computers,2015,32(11):216-223.
[2]Tang Ji-yong,Zhong Yuan-chang,et al.Mean shift object tracking algorithm of adaptive threshold kirsch-lbp texture features[J].Computer Science,2015,42(8):234-239.
[3]Gu Xin-fang,Mao Yao-bin,Li Qiu-jie.Visual tracking algorithms based on Mean shift[J].Computer Science,2013,39(12):16-24.
[4]Xu Chao,Gao Min,Yang Yao.Self-tuning hierarchical Kalmanparticle filter for efficient target tracking[J].Infrared and Laser Engineering,2015,44(6):312-320.
[5]Maroulas V,Stinis P.Improved particle filters for multi-target tracking[J].Journal of Computational Physics,2012,231(2):602-611.
[6]Nummiaro K,Koller-Meier E,Van Gool L.An adaptive colorbased particle filter[J].Image and Vision Computing,2003,21(1):99-110.
[7]Isard M,Blake A.Condensation-conditional density propagation for visual tracking[J].International Journal of Computer Vision,1998,29(1):5-28.
[8]Zeng Wei,Zhu Gui-bin,Chen Jie,et al.Multiple feature fusion tracking with particle filter algorithm[J].Computer Application,2010,30(3):644-645.
[9]Tian Hui,Shen Ting-zhi,Zhao San-yuan,et al.Comprehensive color and texture feature of particle filter for human face tracking algorithm[J].Journal of Beijing Institute of Technology,2010,30(4):469-473.
[10]Mao Xiao-bo,Zhu Dong-wei,Chen Tie-jun.Particle filter object tracking based on histograms of oriented grads and colors[J].Journal of Zhengzhou University,2012,33(4):265-270.
[11]Weng S K,Kuo C M,Tu S K.Video object tracking using adaptive Kalman filter[J].Journal of Visual Communication and Image Representation,2006,17(6):1190-1208.
[12]Dong En-zeng,Su Li-ya,Fu Yan-hong,et al.Fusion of color texture feature adaptive particle filter tracking algorithm[J].Computer Measurement&Control,2014,22(4):125-129.
[1]曹义亲,肖金胜,黄晓生.基于边界沙包核函数的Mean-Shift跟踪算法[J].计算机应用研究,2015,32(11):216-223.
[2]唐继勇,仲元昌,等.基于自适应阈值Kiesch-LBP纹理特征的均值漂移目标跟踪算法[J].计算机科学,2015,42(8):234-239.
[3]顾幸方,茅耀斌,李秋洁.基于Mean Shift的视觉目标跟踪算法[J].计算机科学,2013,39(12):16-24.
[8]曾伟,朱桂斌,陈杰,等.多特征融合的鲁棒粒子滤波跟踪算法[J].计算机应用,2010,30(3):644-645.
[9]田卉,沈庭芝,赵三元,等.综合颜色和纹理特征的粒子滤波人脸跟踪算法[J].北京理工大学学报,2010,30(4):469-473.
[10]毛晓波,朱东伟,陈铁军.融合颜色和梯度方向直方图的粒子滤波跟踪算法[J].郑州大学学报,2012,33(4):265-270.
[12]董恩增,苏丽娅,付艳红,等.融合颜色纹理特征的自适应粒子滤波跟踪算法[J].计算机测量与控制,2014,22(4):125-129.
[2]Tang Ji-yong,Zhong Yuan-chang,et al.Mean shift object tracking algorithm of adaptive threshold kirsch-lbp texture features[J].Computer Science,2015,42(8):234-239.
[3]Gu Xin-fang,Mao Yao-bin,Li Qiu-jie.Visual tracking algorithms based on Mean shift[J].Computer Science,2013,39(12):16-24.
[4]Xu Chao,Gao Min,Yang Yao.Self-tuning hierarchical Kalmanparticle filter for efficient target tracking[J].Infrared and Laser Engineering,2015,44(6):312-320.
[5]Maroulas V,Stinis P.Improved particle filters for multi-target tracking[J].Journal of Computational Physics,2012,231(2):602-611.
[6]Nummiaro K,Koller-Meier E,Van Gool L.An adaptive colorbased particle filter[J].Image and Vision Computing,2003,21(1):99-110.
[7]Isard M,Blake A.Condensation-conditional density propagation for visual tracking[J].International Journal of Computer Vision,1998,29(1):5-28.
[8]Zeng Wei,Zhu Gui-bin,Chen Jie,et al.Multiple feature fusion tracking with particle filter algorithm[J].Computer Application,2010,30(3):644-645.
[9]Tian Hui,Shen Ting-zhi,Zhao San-yuan,et al.Comprehensive color and texture feature of particle filter for human face tracking algorithm[J].Journal of Beijing Institute of Technology,2010,30(4):469-473.
[10]Mao Xiao-bo,Zhu Dong-wei,Chen Tie-jun.Particle filter object tracking based on histograms of oriented grads and colors[J].Journal of Zhengzhou University,2012,33(4):265-270.
[11]Weng S K,Kuo C M,Tu S K.Video object tracking using adaptive Kalman filter[J].Journal of Visual Communication and Image Representation,2006,17(6):1190-1208.
[12]Dong En-zeng,Su Li-ya,Fu Yan-hong,et al.Fusion of color texture feature adaptive particle filter tracking algorithm[J].Computer Measurement&Control,2014,22(4):125-129.
[1]曹义亲,肖金胜,黄晓生.基于边界沙包核函数的Mean-Shift跟踪算法[J].计算机应用研究,2015,32(11):216-223.
[2]唐继勇,仲元昌,等.基于自适应阈值Kiesch-LBP纹理特征的均值漂移目标跟踪算法[J].计算机科学,2015,42(8):234-239.
[3]顾幸方,茅耀斌,李秋洁.基于Mean Shift的视觉目标跟踪算法[J].计算机科学,2013,39(12):16-24.
[8]曾伟,朱桂斌,陈杰,等.多特征融合的鲁棒粒子滤波跟踪算法[J].计算机应用,2010,30(3):644-645.
[9]田卉,沈庭芝,赵三元,等.综合颜色和纹理特征的粒子滤波人脸跟踪算法[J].北京理工大学学报,2010,30(4):469-473.
[10]毛晓波,朱东伟,陈铁军.融合颜色和梯度方向直方图的粒子滤波跟踪算法[J].郑州大学学报,2012,33(4):265-270.
[12]董恩增,苏丽娅,付艳红,等.融合颜色纹理特征的自适应粒子滤波跟踪算法[J].计算机测量与控制,2014,22(4):125-129.