Adaptive framework for robust high-resolution image reconstruction in multiplexed computational imaging architectures

摘要: In multiplexed computational imaging schemes, high-resolution images are reconstructed by fusing the information in multiple low-resolution detected a two-dimensional array of image sensors. The reconstruction procedure assumes mathematical model for process that could have generated observations from an unknown image. practical settings, parameters known only approximately and typically estimated before takes place. Violations to assumed model, such as inaccurate knowledge field view imagers, erroneous estimation parameters, and/or accidental scene or environmental changes can be detrimental quality, even if they small number. We present adaptive algorithm robust architectures. Using M-estimators incorporating similarity measure, proposed scheme adopts strategy effectively deals with violations model. Comparisons nonadaptive techniques demonstrate superior performance terms quality robustness.