In Press                   Back to the articles list | Back to browse issues page

XML Print

Modified Richardson-Lucy Deconvolution Algorithm for Image Restoration
Jaliba Sherin , SHAJEE MOHAN B S * , Ahammed Muneer K V
Abstract:   (45 Views)
Effective image processing involves image restoration as a major task. When image is being acquired or being sent, they frequently suffer from degradation. Blurring, data loss as a result of sampling, quantization effects, and other noise sources are all examples of degradation. Trying to recreate or prepare an estimate of the original image from a degraded image is the goal of image restoration. For image restoration from degraded and damaged images, the Richardson-Lucy deconvolution technique is frequently used. We suggest a modified version of the original Richardson-Lucy deconvolution technique in this paper, to find the best deconvolution transfer function. We also propose a method to obtain the optimal deconvolution transfer function in a class-specific way. We considered performance metrics such as PSNR, MSE, SSIM and KL Divergence for evaluating the performance of the proposed restoration processes and evaluated the performances of different implementations of the Richardson-Lucy deconvolution algorithm. It is seen that the results are at par and in some cases superior to the existing techniques.  A close examination of the proposed methodology, it is seen that the proposed class-specific Richardson-Lucy deconvolution method outperforms all other techniques.
Keywords: Richardson-Lucy deconvolution algorithm, Optimal transfer function, Support Vector Machine, MSE, KL Divergence, SSIM, PSNR
Full-Text [PDF 1403 kb]   (10 Downloads)    
Type of Study: Research Paper | Subject: Image Processing
Received: 2025/08/24 | Revised: 2026/10/27 | Accepted: 2026/08/12
Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Creative Commons License
© 2022 by the authors. Licensee IUST, Tehran, Iran. This is an open access journal distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.