HLBP与颜色特征自适应融合的粒子滤波目标跟踪改进算法.PDFVIP

  • 9
  • 0
  • 约2.77万字
  • 约 8页
  • 2018-10-15 发布于天津
  • 举报

HLBP与颜色特征自适应融合的粒子滤波目标跟踪改进算法.PDF

HLBP与颜色特征自适应融合的粒子滤波目标跟踪改进算法

第 18卷第 1期 南京师范大学学报(工程技术版) Vol18 No1 2018年3月 JOURNAL OF NANJING NORMAL UNIVERSITY(ENGINEERING ANDTECHNOLOGY EDITION) Marꎬ2018 - doi:10.3969/ j.issn.1672 1292.2018.01.008 HLBP与颜色特征自适应融合的 粒子滤波目标跟踪改进算法 卞  乐ꎬ李天峰ꎬ韦  怡ꎬ曾毓敏 (南京师范大学物理科学与技术学院ꎬ江苏 南京 210023) [摘要]  针对只采用颜色特征的经典粒子滤波目标跟踪算法无法适用相同颜色干扰情况的缺陷ꎬ提出一种结 合HLBP特征与颜色特征的自适应粒子滤波跟踪算法. 该算法采用Haar型局部二值模式算子(Haar local binary patternꎬHLBP)提取的HLBP纹理特征与颜色特征结合ꎬ通过自适应权值动态调整颜色特征和纹理特征在追踪过 程中的比重ꎬ实现颜色纹理特征的自适应融合. 实验表明ꎬ该算法改进了相同颜色干扰情况下的追踪效果ꎬ并在 目标被遮挡的情况下仍能持续稳定地追踪ꎬ提高了追踪的准确度和适用性. [关键词]  粒子滤波ꎬHLBP纹理特征ꎬ颜色特征ꎬ自适应权值 - - - [中图分类号]TP391  [文献标志码]A  [文章编号]1672 1292(2018)01 0056 08 Improved Particle Filtering Target Tracking Algorithm for HLBP and Color Feature Adaptive Fusion Bian LeꎬLiTianfengꎬWeiYiꎬZeng Yumin (School of Physics Science and TechnologyꎬNanjing Normal UniversityꎬNanjing210023ꎬChina) Abstract:In this paperꎬan adaptive particle filter tracking algorithm combining HLBP feature and color feature is proposed.This algorithm isbased on imperfection of the classical particle filter target tracking algorithm:it only uses the color featureto track anddoesnotperformeffectivelyinthesamecolorinterferencecondition.OuralgorithmusestheHaar local binary model operator toextract the HLBP texturefeatureandcombineitwiththecolorfeature.Throughtheopposed modelꎬwe dynamicallyadjustadaptiveweightsofthecolorcharacteristicsandtexturefeaturesinthetrackingprocessꎬwhich achievethe adaptive fusion between the color feature and texture feature. Experiments show that the tracking result is greatly strengt

文档评论(0)

1亿VIP精品文档

相关文档