1 November 2016 Scene recognition and colorization for vehicle infrared images
Author Affiliations +
Proceedings Volume 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control; 101572V (2016) https://doi.org/10.1117/12.2247138
Event: International Symposium on Optoelectronic Technology and Application 2016, 2016, Beijing, China
Abstract
In order to make better use of infrared technology for driving assistance system, a scene recognition and colorization method is proposed in this paper. Various objects in a queried infrared image are detected and labelled with proper categories by a combination of SIFT-Flow and MRF model. The queried image is then colorized by assigning corresponding colors according to the categories of the objects appeared. The results show that the strategy here emphasizes important information of the IR images for human vision and could be used to broaden the application of IR images for vehicle driving.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Junjie Hou, Junjie Hou, Shaoyuan Sun, Shaoyuan Sun, Zhenyi Shen, Zhenyi Shen, Zhen Huang, Zhen Huang, Haitao Zhao, Haitao Zhao, } "Scene recognition and colorization for vehicle infrared images", Proc. SPIE 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control, 101572V (1 November 2016); doi: 10.1117/12.2247138; https://doi.org/10.1117/12.2247138
PROCEEDINGS
5 PAGES


SHARE
Back to Top