Domain Adaptation and Adaptive Information Fusion for Object Detection on Foggy Days

Sensors

Huibin Wang¹  Zhe Chen¹  Hongmin Gao¹  Hao Zheng²  Xiaofang Li³ 

[1] College of Computer and Information, Hohai University, Nanjing 210098, China;College of Information Engineering, Nanjing Xiaozhuang University, Nanjing 210017, China;Jiangsu Collaborative Innovation Center for Cultural Creativity, Changzhou 213000, China;

关键词: object detection;    foggy day;    haze effect;    depth information;   

DOI  :  10.3390/s18103286

来源: DOAJ

【 摘 要 】

Foggy days pose many difficulties for outdoor camera surveillance systems. On foggy days, the optical attenuation and scattering effects of the medium significantly distort and degenerate the scene radiation, making it noisy and indistinguishable. Aiming to solve this problem, in this paper we propose a novel object detection method that has the ability to exploit the information in the color and depth domains. To prevent the error propagation problem, we clean the depth information before the training process and remove false samples from the database. A domain adaptation strategy is employed to adaptively fuse the decisions obtained in the color and depth domains. In the experiments, we evaluate the contribution of the depth information for object detection on foggy days. Moreover, the advantages of the multiple-domain adaptation strategy are experimentally demonstrated via comparison with other methods.

【 授权许可】

Unknown