Noise-robust boundary recursive algorithm for super-resolution reconstruction of staring focal plane array micro-scanning imaging

A boundary recursive algorithm for the super-resolution reconstruction of staring focal plane array (FPA) micro-scanning imaging with consideration of a fill ratio-based sampling model is presented. The reconstruction errors of the algorithm introduced by image noise and boundary approximation are analyzed. Then, a modified bilateral filter and gray statistical principle are used in the algorithm to reduce these errors. Simulation and actual imaging experiments confirm that the proposed algorithm has effective noise robustness and can achieve superior results compared to an over-sampled reconstruction in the presence of low noise. This algorithm can achieve ideal sub-pixel imaging and has excellent immunity to noise. Its application will enhance the performance of optoelectronic-imaging systems.