Enhancing Real-World Image Deblurring: Algorithm Selection and Performance Evaluation for CCTV
Venu Dattathreya Vemuru, Nalluru Mourya Sai Eatesh, Varshith Yadavalli, Nellore Dhanush, C M Varun, S.V Suguna Sekhar · 2023
To restore clear photos from blurred photographs due to camera movement, several image deblurring techniques have been developed. Deblurring is th process of taking off blurs and bringing back the sharp image. Defocus, motion, Gaussian, average, and other types of blur are only a few examples. We want to identify the bes algorithms for real-world problems (such as CCTV and traffic camera images) in this project. We look at the algorithms used to solve a certain issue. The PSNR and SSIN values will be used as benchmark scores as these chosen methods are tested on actual blurred photographs. Existing datasets can be used to evaluate the performance of algorithms; however, actual data may differ slightly from these. Three major categories can be used to categorize deblurring techniques: blind and non-blind, regularization, deep learning, and hybrid methods. We are focusing on applying the best algorithm for deblurring photos in this to produce incredibly effective results.