Research on coding angle self-calibration based on compound eye array

Le résumé fourni par la source

Coded aperture super-resolution imaging technology is an effective method to improve imaging resolution. By leveraging the anisotropic properties of coding, this technology captures high-frequency information in different directions and generates high-resolution images using iterative algorithms. For array systems adopting parallel coding, accurately establishing the forward model of coding is crucial. In this study, a self-calibration technology based on the coding angle of compound eye arrays is proposed. This technology transforms the spectrum of the modulated image from the Cartesian coordinate system to the polar coordinate system, thereby converting the calibration of coding rotation angle into translation calibration. Using the spectral cross-correlation method in the polar coordinate system, this study can accurately measure the displacement between the image to be measured and the reference image, and further determine the actual rotation angle of the coding. This self-calibration technology achieves precise calibration of the coding angle of multi-aperture arrays, with a calibration accuracy of 91.3%. It provides prior information for subsequent high-resolution image reconstruction and simplifies the calibration process of camera arrays. In the USAF calibration and reconstruction experiment, compared with the uncalibrated results, the calibrated results show that the PSNR and SSIM indices of the reconstructed images are increased by 4.22

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.