Citation

Watanabe K, Nomoto K, Kato S. Trans. Soc. Automot. Eng. Jpn. 2016; 47(6): 1405-1409.

Copyright

(Copyright © 2016, Society of Automotive Engineers of Japan)

DOI

10.11351/jsaeronbun.47.1405

PMID

unavailable

Abstract

For object (e.g. pedestrian) detection, large scale database measured via an in-vehicle camera have been proposed. In this report, we visualized statistical characteristics for the databases by using non-linear discriminant analysis (NDA) and presented the pedestrian recognition accuracies based on the estimated posterior probabilities in the visualization by using NDA

Language: ja

Keywords

Accident avoidance; Computer application; Safety