电子学报Issue(2)：217-224,8.DOI:10.3969/j.issn.0372-2112.2015.02.002
基于后验 HOG 特征的多姿态行人检测
Multi-Pose Pedestrian Detection Based on Posterior HOG Feature
摘要
Abstract
Pedestrian detection remains one of the challenging tasks in the area of computer vision .A multi-pose pedestrian detection method based on posterior HOG feature is proposed .Firstly ,the generality information of gradient feature energy is com-puted with all pedestrian samples .The posterior HOG feautre is obtained by weighting the HOG feature of individual pedestrian sam-ple with the computed gradient feature energy .The posterior HOG feature can capture the contours and edges of pedesrtians ,and significantly reduce the influence of complex and cluttered background .Secondly ,pedestrians of different poses and views are divid-ed into subclasses with S-Isomap and K-means algorithm .A classifier is trained for each subclass .Finally ,a multi-pose-view ensem-ble classifier is trained to combine the output values of different subclass classifiers with an equally weighted sum rule .Experimental results on different datasets suggest that the proposed posterior feature outperforms the classic HOG feature and other typical fea-tures .Compared with the existing methods ,by combining the posterior feature and the multi-pose-view ensemble classifier ,the pro-posed method boosts the detection accuracy effectively .关键词
后验HOG特征/梯度能量图/S-Isomap/支持向量机/行人检测Key words
posterior HOG feature/gradient energy map/S-Isomap/support vector machine/pedestrian detection分类
信息技术与安全科学引用本文复制引用
刘威,段成伟,遇冰,柴丽颖,袁淮,赵宏..基于后验 HOG 特征的多姿态行人检测[J].电子学报,2015,(2):217-224,8.基金项目
国家自然科学基金（No ．61273239）；中央高校基本业务费 （）
