Research on Photon-Integrated Interferometric Remote Sensing Image Reconstruction Based on Compressed Sensing

Abstract

Achieving high-resolution remote sensing images is an important goal in the field of space exploration. However, the quality of remote sensing images is low after the use of traditional compressed sensing with the orthogonal matching pursuit (OMP) algorithm. This involves the reconstruction of the sparse signals collected by photon-integrated interferometric imaging detectors, which limits the development of detection and imaging technology for photon-integrated interferometric remote sensing. We improved the OMP algorithm and proposed a threshold limited-generalized orthogonal matching pursuit (TL-GOMP) algorithm. In the comparison simulation involving the TL-GOMP and OMP algorithms of the same series, the peak signal-to-noise ratio value (PSNR) of the reconstructed image increased by 18.02%, while the mean square error (MSE) decreased the most by 53.62%. The TL-GOMP algorithm can achieve high-quality image reconstruction and has great application potential in photonic integrated interferometric remote sensing detection and imaging.