Recognizing pedestrian’s unsafe behaviors in far-infrared imagery at night
详细信息 查看全文
- 作者：Eun Ju Lee frigia@kmu.ac.kr" class="auth_mail" title="E-mail the corresponding author ; Byoung Chul Ko ; niceko@kmu.ac.kr" class="auth_mail" title="E-mail the corresponding author ; Jae-Yeal Nam jynam@kmu.ac.kr" class="auth_mail" title="E-mail the corresponding author
- 关键词：Unsafe behavior recognition ; Advance driver assistance system ; Convolution neural network ; Genetic algorithm ; Spatial pyramid pooling ; Boosted random forest
- 刊名：Infrared Physics & Technology
- 出版年：2016
- 出版时间：May 2016
- 年：2016
- 卷：76
- 期：Complete
- 页码：261-270
- 全文大小：1658 K
文摘
- •
Recognizing pedestrian’s behaviors using thermal image from moving vehicle at night.
- •
Designing the light convolutional neural networks with random forest classifier.
- •
Generating the pedestrian’s unsafe behavior (PUB) dataset using thermal camera.
- •
Behavior recognition accuracy is higher than that of related algorithms.