International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Block-Matching and 3D Collaborative Filtering for Image Denoising |
|---|---|
| ABSTRACT | Image denoising is a fundamental problem in signal processing and computer vision, with wide applications in medical imaging, remote sensing, and digital photography. This paper presents a comprehensive analysis of the Block-Matching and 3D Filtering (BM3D) algorithm, a state-of-the-art approach for Gaussian noise reduction. BM3D utilizes block matching to identify similar image patches, applies collaborative filtering in the transform domain to suppress noise, and performs aggregation to reconstruct the final image. Comparative experiments with Non-Local Means (NLM) and wavelet-based methods on standard test images under varying noise levels (σ = 10–50) demonstrate BM3D’s superior performance, achieving PSNR improvements of 3–5 dB and SSIM values above 0.90. The results highlight BM3D’s strong ability to preserve fine details, textures, and edges even under heavy noise. Although BM3D is computationally intensive, its robustness, high perceptual quality, and consistent results across diverse images reaffirm its value as a benchmark denoising method. Furthermore, this study provides insight into integrating BM3D principles with deep learning frameworks for improved efficiency and visual fidelity in future denoising systems. |
| AUTHOR | TCHEPSEU PATENG URICHE CABREL, GANBOURI ABDELLATIF, GIFTY ADWOA BEMPONG, LANGAT NIXON KIPKIRUI, ABDERRAZAK HAMRICH, CHOUACHI WASSIM |
| PUBLICATION DATE | 2025-10-29 |
| VOLUME | 175 |
| DOI | DOI: 10.15680/IJIRCCE.2025.1310023 |
| pdf/23_Block-Matching and 3D Collaborative Filtering for Image Denoising.pdf | |
| KEYWORDS |
