DDAN: A Dense Dual-Attention Network with Physics-Informed Loss for Robust Multi-Type Optical Image Denoising
Abstract
Abstract Image degradation caused by additive Gaussian noise, multiplicative speckle, and their hybrid combination is a fundamental challenge in optical imaging. Existing methods predominantly specialise for a single noise type, leaving a critical performance gap when multiple noise sources co-exist. This paper proposes the Dense Dual-Attention Network (DDAN), an end-to-end encoder-decoder framework for robust multi-type image denoising. DDAN incorporates Residual Dense Blocks with Dual Attention (RDBA) at every encoder and decoder stage, combining progressive dense feature concatenation maximising gradient flow and feature reuse with a sequential Dual Attention Module (DAM) that performs channel recalibration via global average pooling and spatial gating via a 7\((\times)\)7 depthwise convolution. Training is governed by a physics-informed composite loss integrating \((L_{1})\) pixel fidelity, structural similarity (SSIM), and a log-ratio speckle constraint (\((\alpha=0.7)\), \((\beta=0.2)\), \((\gamma=0.1)\)). Evaluated on Classic5, Kodak24, McMaster, and Set12 against seven competing methods, DDAN achieves state-of-the-art performance in all 96 evaluated conditions without exception surpassing DATNet by 7.94-12.51 dB PSNR under additive noise, by 0.58-1.44 dB across all multiplicative look counts including single-look (L=1), and by 0.90-7.41 dB under hybrid noise, establishing a new universal benchmark for multi-type optical image denoising.
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.