Title: Camera calibration based on homogenous fog weather condition

Authors: Hongjun Song; Yuanyuan Gao; Yangzhou Chen

Addresses: Center for Autonomous Technology and Intelligent Control, Beijing University of Technology, Beijing, 100124, China ' School of Information Engineering, Zhejiang A&F University, Lin'an, 311300, China ' Center for Autonomous Technology and Intelligent Control, Beijing University of Technology, Beijing 100124, China

Abstract: A novel algorithm for automatic and dynamic camera calibration based on dark channel prior in homogenous fog weather condition is presented in this paper. Three main steps are included in the algorithm. Firstly, current weather condition will be judged to justify whether it is homogenous weather condition or not based on area search method (ASE). Secondly, traffic road plane will be found by activity map created by calculating the expected value of the absolute intensity difference between two adjacent frames. Finally, transmission image will be obtained by dark channel prior. Intrinsic and extrinsic parameters of the camera will be calculated based on the parameter calibration formula deduced from camera model and light transition model in fog. In this step eight key points for generating necessary calculation equations in transmission image in road plane will be selected to calibrate the camera. Experiments verify the effectiveness and the robustness of the proposed algorithm.

Keywords: homogenous fog weather conditions; dark channel prior; dynamic camera calibration; automatic camera calibration; light transition models; modelling.

DOI: 10.1504/IJMIC.2013.054039

International Journal of Modelling, Identification and Control, 2013 Vol.19 No.1, pp.75 - 88

Published online: 27 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article