Image-based correction of radial distortion and vignetting via phase and entropy optimization

Radial distortion and vignetting degrade image geometry and brightness uniformity in camera systems. This paper presents two correction methods: a phase-shift-based algorithm for radial distortion and an entropy-minimizing simulated annealing approach for vignetting. The latter models radial brightness attenuation using a polynomial function and optimizes parameters by minimizing image intensity entropy. Both simulation and experimental results verify the effectiveness of the proposed methods. Compared to Zhang's method [IEEE Trans. Pattern Anal. Mach. Intell.22, 1330 (2000)ITPIDJ0162-882810.1109/34.888718], our radial correction reduces peripheral distortion, and vignetting correction achieves the highest PSNR (50.24 dB) and SSIM (0.9978), reducing measurement error from 0.003 to 0.001 mm in bearing inspection. The proposed approach improves image quality and supports more reliable measurements in industrial imaging tasks.