A novel camera calibration technique based on differential evolution particle swarm optimization algorithm

NEUROCOMPUTING

卷:174

Article

Deng, Li^1,3  Lu, Gen^1,3  Shao, Yuying²  Fei, Minrui^1,3  Hu, Huosheng⁴ 

[1] Shanghai Univ, Sch Mechatron Engn & Automat, Shanghai 200072, Peoples R China

[2] State Grid Shanghai Municipal Elect Power Co, Shanghai 200025, Peoples R China

[3] Shanghai Key Lab Power Stn Automat Technol, Shanghai 200072, Peoples R China

[4] Univ Essex, Dept Comp Sci, Colchester CO4 3SQ, Essex, England

关键词: Camera calibration;    Internal parameter;    External parameter;    Differential evolution;    Particle swarm algorithm;    Visual identification;   

DOI  :  10.1016/j.neucom.2015.03.119

来源: Elsevier

【 摘 要 】

Camera calibration is one of the fundamental issues in computer vision and aims at determining the intrinsic and exterior camera parameters by using image features and the corresponding 3D features. This paper proposes a relationship model for camera calibration in which the geometric parameter and the lens distortion effect of camera are taken into account in order to unify the world coordinate system (WCS), the camera coordinate system (CCS) and the image coordinate system (ICS). Differential evolution is combined with particle swarm optimization algorithm to calibrate the camera parameters effectively. Experimental results show that the proposed algorithm has a good optimization ability to avoid local optimum and can complete the visual identification tasks accurately. (C) 2015 Elsevier B.V. All rights reserved.

【 授权许可】

Free