An Adaptive Thresholding Method for Image Quality Improvement

Remote sensing images can degrade in quality due to various circumstances, both internal to the imaging process and external to it. Overall, image restoration can assist improve image quality. The Maximum a Posteriori (MAP) method is the most widely used strategy for reconstructing remote sensing images. During the image reconstruction process, the algorithm integrates a priori knowledge and observational data to improve the clarity of the resultant image while keeping its authenticity. This research describes a new image reconstruction method that blends adaptive Huber-function-based thresholding with Local Binary Patterns (LBP) features for remote sensing image reconstruction. The magnitudes of image gray values (DN values) inform the adaptive selection of threshold values for the Huber regularization term. The results show that the suggested technique outperforms existing state-of-the-art methods both subjectively and objectively.