Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.14 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
Abstract
In many image processing applications, image deblurring is an essential preprocessing step that addresses degradation caused by motion blur, defocus, or defective sensors. The importance of image deblurring resides in its capacity to restore clarity, improve image quality, and increase the accuracy of following analyses, including object detection, recognition, and medical imaging. Effective deblurring techniques can have a major impact on the accuracy and dependability of results in domains such as digital forensics, autonomous driving, security surveillance, and satellite imaging. Recent advances in deep learning and computational algorithms have enabled more effective and superior deblurring techniques, offering significant improvements over conventional methods.This study proposes the development and evaluation of an effective image deblurring model, MSREDNet, aimed at producing high-quality images from blurred inputs. The model utilizes an advanced deep learning approach to proficiently resolve the issues related to image blur. The proposed MSREDNet comprises a hierarchical multi-scale feature extraction block, an encoder-decoder with a residual block, and a multi-scale feature integration block. The suggested MSRED Net framework is evaluated on the GoPro dataset. The findings show that MSRED Net achieves an impressive PSNR of 36.89 dB and an SSIM of 0.972, indicating its remarkable ability to restore fine details, lower noise, and keep the structural integrity of deblurred images. These findings demonstrate how MSRED Net can be a strong and dependable option for image deblurring in practical applications where quality and visual impact are crucial.
Keywords:
Keywords:Digital Image Processing, Motion Blur, Deblurring, Deep Learning, Image Restoration, Peak Signal to Noise Ratio, Structural Similarity Index Measure.
Cite Article:
"Multi-Scale Residual Encoder-Decoder Network for Robust Image Deblurring: A Deep Learning-Based Digital Image Processing Solution", International Journal for Research Trends and Innovation (www.ijrti.org), ISSN:2455-2631, Vol.9, Issue 12, page no.a116-a133, December-2024, Available :http://www.ijrti.org/papers/IJRTI2412013.pdf
Downloads:
000818
ISSN:
2456-3315 | IMPACT FACTOR: 8.14 Calculated By Google Scholar| ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.14 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator