Pedestrian protection system (PPS) in the advanced driver assistance system (ADAS) to improve traffic safety, has become an important research area. The major challenge in PPS is to develop a reliable on-board pedestrian detection system. Compared to the pedestrian detection on static images, on-board pedestrian detection is facing some new difficulties, such as high real-time demand, wide range of illumination variation and so on. In order to deal with these challenges, we presented a method in this paper by combining the road surface extraction technique and the histogram of oriented gradient (HOG) feature based classification, so that the search regions are only limited within the extracted road surface. Experiment results show that this method can remarkably reduce the false alarm rate, improve the detection speed, and significantly improve the small pedestrian detection rate.
Pedestrian Detection Based on Road Surface Extraction in Pedestrian Protection System
Lect. Notes Electrical Eng.
2013-10-04
8 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
PEDESTRIAN PROTECTION SYSTEM AND PEDESTRIAN PROTECTION PROGRAM
Europäisches Patentamt | 2015
|Body Parts Features-Based Pedestrian Detection for Active Pedestrian Protection System
DOAJ | 2016
|PEDESTRIAN DETECTION DEVICE, PEDESTRIAN DETECTION SYSTEM, AND PEDESTRIAN DETECTION METHOD
Europäisches Patentamt | 2018
|Europäisches Patentamt | 2022
|