计算机工程

• 图形图像处理 • 上一篇    下一篇

基于HOG特征优化的夜间行人快速识别方案

汤琳 1,2,李敏 2,柳波 2   

  1. (1.中国科学院成都计算机应用研究所,成都 610041; 2.绵阳师范学院数学与计算机科学学院,四川 绵阳 621000)
  • 收稿日期:2014-11-03 出版日期:2015-07-15 发布日期:2015-07-15
  • 作者简介:汤琳(1982-),女,讲师、博士研究生,主研方向:图像处理,模式识别;李敏,副教授、博士;柳波,讲师、硕士。
  • 基金项目:

    四川省教育厅基金资助项目(14ZA0257);四川省教育厅青年基金资助项目(12ZB261)。

Fast Identification Scheme for Nighttime Pedestrian Based on HOG Feature Optimization

TANG Lin  1,2,LI Min  2,LIU Bo  2   

  1. (1.Chengdu Institute of Computer Application,Chinese Academy of Sciences,Chengdu 610041,China; 2.School of Math and Computer Science,Mianyang Normal University,Mianyang 621000,China)
  • Received:2014-11-03 Online:2015-07-15 Published:2015-07-15

摘要:

针对夜间行人检测成像尺度不一所引起的类内方差较大、实时性不足等问题,在统计学习的应用原理下,设计基于熵加权和快速分类支持向量机(FCSVM)优化的头部校验夜间行人快速识别方案。应用熵加权原理改进梯度直方图特征,引入三分支结构的支持向量机识别目标,同时利用FCSVM降低运算开销,确保实时性,通过头部校验方法分析评估误检目标,提高图像匹配的准确度。实验结果表明,该方案在夜间环境下能有效区分远红外行人目标,在充分确保行人检测实时性的基础上,在市区、郊区等不同应用环境中均具有较好的识别效果。

关键词: 夜间行人检测, 统计学习, 熵加权, 快速分类支持向量机, 头部校验

Abstract:

For the larger intra-class variance and inadequate real-time problems caused by factors of imaging scales difference in nighttime pedestrian detection,this paper designs a rapid dentify program for nighttime pedestrians based on entropy weight and header calibration of Fast Classification Support Vector Machine(FCSVM) optimization under the application of statistical learning principles.The program utilizes entropy weight to improve the feature of gradient histogram,introduces three branch structure SVM to identify the target further,and uses rapid classification FCSVM to reduce the overhead required computation and to ensure real-time.Through the header calibration method to analyze and assess error detection goals,it further improves the accuracy of image matching.Experimental results show that the scheme can distinguish far infrared pedestrian goals effectively at night environment,and have good recognition effect in urban,suburban and other different application environments on the basis of ensuring pedestrian real-time fully.

Key words: nighttime pedestrian detection, statistical learning, entropy weight, Fast Classification Support Vector Machine(FCSVM), header calibration

中图分类号: