A novel blind approach for image restoration using adaptive kurtosis based deconvolution
Denish Kheni, Truptesh Italiya, Dippal Isarani, D Karthick · 2017
In the process of image formation, the quality of original image diminishes due to atmospheric conditions, imaging system, transmission medium or faulty devices. Major affecting areas like surveillance, drones, medical imaging, astronomy, microscopy, remote sensing, rover etc., are the key motivation behind blind image restoration process. In restoration through blind approach, a prominent challenge is to estimate the Point Spread Function (PSF) of blur image. Hence, an effective technique requires to solve degradation problem and to restrain image obscureness as minimum as possible. A novel technique is aimed in this paper which works effectively on image corrupted by atmospheric turbulence blur. Corrupted image is initially pre-processed to mitigate heat scintillation. This paper also proposes an efficient adaptive kurtosis based blind deconvolution algorithm to restore degraded image sequences.