Lens Distortion Correction Method Based on Cross-Ratio Invariability and Compressed Sensing

Dingfei Jin,Yue Yang,Yiping Luo,Sisi Liu
DOI: https://doi.org/10.1117/1.oe.57.5.054103
IF: 1.3
2018-01-01
Optical Engineering
Abstract:Optical lens distortions will seriously affect the measurement accuracy of the vision system. An independent lens distortion correction method based on cross-ratio invariability and compressed sensing is, thus, proposed to improve the measurement accuracy of the system. A virtual distortion correction template is designed according to the characteristics of the lens distortion. The ideal projection coordinate in the imaging plane of the camera and the distortion correction value of the characteristic points in the virtual correction template can be calculated based on the four-point cross-ratio invariance principle. The ideal projection coordinate of the other pixels in the image can then be reconstructed based on compressed sensing theory and, thus, correcting the distortion of all pixels. The experimental results show that for the same vision measurement system, when the independent distortion correction algorithm is used, its measurement accuracy is significantly improved, and the maximum error is only 0.069 mm. This method can correct the distortion effectively, thus improving the precision of the system. (C) 2018 Society of Photo-Optical Instrumentation Engineers (SPIE)
What problem does this paper attempt to address?