Context aided pedestrian detection for danger estimation based on laser scanner and computer vision

作者：

Highlights：

• A novel pedestrian detection in road safety application based on JPDA and KF.

• Results have proved that, both performance and trustability are increased.

• The presented work gives a solution based on processes performed at all JDL levels.

• The tests provided compare JPDA with classic GNN.

• Context aids at levels 2 and 3 are presented (danger estimation and threat detections).

摘要

•A novel pedestrian detection in road safety application based on JPDA and KF.•Results have proved that, both performance and trustability are increased.•The presented work gives a solution based on processes performed at all JDL levels.•The tests provided compare JPDA with classic GNN.•Context aids at levels 2 and 3 are presented (danger estimation and threat detections).

论文关键词：Data fusion,Context,Laser scanner,Computer vision,Pedestrian detection

论文评审过程：Available online 14 May 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.04.034