期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2015
卷号：6
期号：2
页码：1429-1432
出版社：TechScience Publications
摘要：This work is concerned with the challenging task of pedestrian detection in real world environments. That is, the aim is to successfully localize pedestrian in video surveillance system despite the presence of background clutter or partial occlusions. Even though pedestrian has many practical applications and has been an active area of research for many years, it has not been until recently that pedestrian recognition algorithms have become robust enough to deal with scene of realistic situations. This work proposes algorithms and algorithmic extensions, which further enhance detection accuracy compared to existing state- of-the –art approaches. Persons and pedestrian however are not rigid and their appearance changes greatly depending on the body articulation or pose. The variations of colors in clothing and textures add further difficulties. The proposed system has two phases learning phase and detection phase. In the learning phase create robust feature set that allows human form to be discriminated clearly even in cluttered background under different illumination. Object finding technique is based on Histogram of oriented gradients(HOG). Here shape based analysis is carried out. This thesis puts particular emphasis on pedestrian on detection in improved detection accuracy with decreased computational cost.
关键词：Visual features; object detection;Histogram of oriented gradients (HOG); pedestrian;detection; real-time systems
