智能系统学报Issue(4)：289-294,6.DOI:10.3969/j.issn.1673-4785.2011.04.002
综合颜色和小波纹理特征的免疫粒子滤波视觉跟踪
An immune particle filter video tracking method based on color and wavelet texture
摘要
Abstract
Focusing on the drawbacks of standard particle filters in visual tracking, a new immune particle filter was proposed based on color and wavelet texture feature to avoid " sample impoverishment" and incomplete feature. Since the wavelet feature was added, the poor performance of only one feature was improved when the environment changed. In addition, the proposed immune optimization algorithm improved the diversity of particles, especially the decline of the impact of sample impoverishment when occlusion occurs. At last, the effectiveness and robustness of the proposed method was verified by experiments. In contrast to the single feature tracking method, the proposed method can track targets with complex backgrounds more steadily.关键词
粒子滤波/免疫粒子滤波/小波纹理特征/相似性度量/视觉跟踪Key words
particle filter/ immune particle filter/ wavelet texture/ similarity measurement分类
信息技术与安全科学引用本文复制引用
韩华,丁永生,郝矿荣..综合颜色和小波纹理特征的免疫粒子滤波视觉跟踪[J].智能系统学报,2011,(4):289-294,6.基金项目
国家自然科学基金重点资助项目(61134009) （61134009）
国家自然科学基金资助项目(60975059,60775052) （60975059,60775052）
国家ITER计划国内配套研究资助项目(2010GB108004) （2010GB108004）
教育部高等学校博士学科点专项科研基金资助项目(20090075110002) （20090075110002）
上海市优秀学术带头人计划资助项目(11XD1400100) （11XD1400100）
上海市科学技术委员会重点基础研究资助项目(10JC1400200,09JC1400900) （10JC1400200,09JC1400900）
上海市科学技术委员会技术标准专项资助项目(10DZ0506500). （10DZ0506500）
