期刊文献+

基于人体部位行人检测算法 被引量:2

Research on component-based pedestrian detection
下载PDF
导出
摘要 首先运用haar-like特征集,使用Adaboost算法训练全身和头肩、躯干和腿等部位检测器。全身检测器产生行人候选区域,再划分头肩、躯干和腿的待检测区域,然后将检测区域的检测结果组合分析,最后确定是否为行人。移动摄像机拍摄的视频图像结果表明,该方法在复杂街道上具有良好的检测效果。 With haar-like dataset,the Adaboost algorithm is used to train full-body detector and detectors of part of body such as head-shoulder,torso and leg,and then the pedestrian candidate areas were generated by full-body detector and the pedestrian candidate areas were divided into three parts of head-shoulder,torso and leg.The three body part detectors with the head,torso and leg were detected the corresponding parts of the area and analyze the combination of detection results,to further determine the pedestrian candidate whether people.Experimental results show the method has high performance in detecting pedestrians in cluttered background.
出处 《长春工业大学学报》 CAS 2017年第5期433-437,共5页 Journal of Changchun University of Technology
基金 吉林省科技厅基金资助项目(20160312002ZG)
关键词 ADABOOST算法 部位检测器 行人检测 Adaboost algorithm part detector pedestrian detection
  • 相关文献

参考文献1

二级参考文献4

  • 1Li B,Liu Y H.When eigenfaces are combined with wavelets[J].Knowledge-based Systems,2002,15(6):343-347.
  • 2Lai J H,Yuen P C.Face recognition using holistic Fourier invariant features[J].Pattern Recognition,2001,34(1):95-109.
  • 3Cortes C,VapniK V.Support vector networks[J].Machine Learning,1995,20(3):273-297.
  • 4Guo G,Li S Z,Chan K.Face Recognition by Support Vector Machines[A].Proceedings of the 4th IEEE Intern.Conf.on Automatic Face and Gesture Recognition[C].[s.l.]:[s.n.],2000:196-201.

共引文献8

同被引文献5

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部