Lenslets combination optimal design of the segmented planar image detector for electro-optical reconnaissance

Abstract

Segmented planar image detector for electro-optical reconnaissance (SPIDER) is a new type of lightweight and high-resolution computational imaging system that has applications in fields such as remote sensing terrain exploration, high precision military detection, and remote environmental monitoring. Currently, the lens combination methods used in the SPIDER system misses a large amount of the spatial-frequency information detected by each spoke’s lens array. The SPIDER imaging principle was analyzed with a particular focus on understanding the effect of the lenslets combination method on image quality. To optimize the quality of reconstructed images, we have proposed a pseudo two-layer parity configuration. To compare the quality of image restoration using the pseudo two-layer parity combination methods with previously existing methods, the entire imaging process was numerically simulated and the peak signal-to-noise ratio of each reconstructed image was estimated. The simulation demonstrated that the reconstructed image obtained using the pseudo two-layer parity combination was more similar to the original image and exhibited a higher image quality than the images reconstructed using other methods. These results indicate that the optical structure of the SPIDER system can be optimized by implementing the pseudo two-layer parity aperture combination method and this provides theoretical support for the further development of the SPIDER system.