Open AccessPlane dimension detection with a single camera based on a flat refractive camera model
Author(s) -
Xu Dong,
Liu Kedong,
Yang Quan,
Dong Feng,
Wang Xiaochen,
Dai Zhenyang
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2018.5124
Subject(s) - pinhole camera model , pinhole (optics) , optics , camera resectioning , refraction , pinhole camera , computer vision , projection (relational algebra) , dimension (graph theory) , artificial intelligence , plane (geometry) , planar , computer science , image plane , camera auto calibration , physics , computer graphics (images) , mathematics , image (mathematics) , algorithm , geometry , pure mathematics
It is common that a pinhole camera captures object images through a transparent plate with two parallel planar faces. In this case, the refraction leads to a non‐linear error using the simple pinhole model. To address this issue, a modified pinhole camera model, which is called flat refractive camera model, is established based on the analysis of light refraction in the transparent plate placed between the scene and the camera. Then, an effective backward‐projection algorithm is proposed to realise plane dimension detection and eliminate the plate refraction effect. Unlike other models, this proposed model has a form similar to the pinhole camera model. Therefore, it is easy to employ conventional vision techniques in the plate imaging system. At the same time, experiments are designed to demonstrate the plane dimension detection based on the proposed method. Experimental results show that the proposed method can obtain higher accuracy than the traditional method without considering the plate refraction effect.