Blind Remote Sensing Image Deblurring Based on Overlapped Patches’ Non-Linear Prior

• DOI: 10.3390/s22207858
• Publication Date: 2022-10-17T07:43:58Z
• Authors: Ziyu Zhang; Liangliang Zheng; Wei Xu; Tan Gao; Xiaobin Wu; Biao Yang
• Keywords: Deblurring

The remote sensing imaging environment is complex, in which many factors cause image blur. Thus, without prior knowledge, the restoration model established to obtain clear images can only rely on observed blurry images. We still build with extreme pixels but no longer traverse all pixels, such as channels. features are extracted units of patches, segmented from an and partially overlap each other. In this paper, we design a new prior, i.e., overlapped patches’ non-linear (OPNL) derived ratio affected by blurring patches. analysis more than 5000 confirms that OPNL prefers rather process. complexity optimization problem increased due introduction makes it impossible solve directly. A related solving algorithm based projected alternating minimization (PAM) combined half-quadratic splitting method, fast iterative shrinkage-thresholding (FISTA), Fourier transform (FFT), etc. Numerous experiments prove has excellent stability effectiveness obtained competitive processing results restoring