Journal of Computer Applications ›› 0, Vol. ›› Issue (): 0-0.DOI: 10.11772/j.issn.1001-9081.2023081131

    Next Articles

Image denoising algorithm based on local and global feature decoupling

  

  • Received:2023-08-23 Revised:2023-11-01 Online:2023-12-18 Published:2023-12-18
  • Contact: SHI Hongbo
  • Supported by:
    Central Guiding Local Science and Technology Development Fund Projects;Key Research and Development Program of Shanxi Province;Natural Science Foundation of Shanxi Province;Graduate Research Innovation Project of Shanxi Province

基于局部和全局特征解耦的图像去噪算法

丁宇伟1,石洪波2,李杰3,梁敏4   

  1. 1. 山西财经大学信息学院
    2. 山西财经大学 信息管理学院,太原 030031
    3. 山西财经大学
    4. 山西财经大学 信息学院,太原 030006;
  • 通讯作者: 石洪波
  • 基金资助:
    中央引导地方科技发展资金项目;山西省重点研发计划项目;山西省自然科学基金资助项目;山西省研究生科研创新项目

Abstract: Abstract: Current Transformer-based algorithms focus on capturing the global features of images, while ignoring the crucial role of local features in restoring image detail information. To address this issue, an image denoising algorithm based on local and global feature decoupling is proposed, which includes two multi-scale branches based on hybrid Transformer block(HTB) and a single-scale branch based on convolutional neural network(CNN), aiming at combining HTB's powerful global modeling capabilities with CNN's local modeling advantages, and yielding outputs with enriched contextual information and precise spatial details. HTB employs self-attention mechanisms to adaptively model spatial and channel-dimensional dependencies to activate a wider range of input pixels for reconstruction. Given the potential information conflicts across different branches, a feature transfer block is designed to facilitate cross-branch propagation of global features and suppress low-frequency information, thereby ensuring collaborative interactions among the branches. Compared to the Transformer-based denoising algorithm (Uformer) on the real-world image dataset DND, the proposed algorithm achieves an improvement of 0.14dB in peak signal-to-noise ratio (PSNR). For the synthetic image dataset Urban100, the proposed algorithm achieves an average PSNR enhancement of 0.41dB compared to the multi-stage denoising algorithm (MSPNet). Experimental results demonstrate that the proposed algorithm effectively removes image noise and reconstructs finer texture details.

Key words: Keywords: Transformer, image denoising, global feature, local feature, feature decoupling

摘要: 摘 要: 当前基于Transformer的图像去噪算法侧重于捕获图像的全局特征,而忽视了局部特征对于恢复图像细节信息的关键作用。针对该问题,提出一种基于局部和全局特征解耦的图像去噪算法,它包含两个基于混合Transformer模块(HTB)的多尺度分支和一个基于卷积神经网络(CNN)的单尺度分支,旨在将HTB强大的全局建模能力与CNN的局部建模优势有机结合,生成上下文信息丰富且空间细节准确的输出。HTB采用自注意力机制自适应地对空间和通道维度的依赖关系进行建模,以激活范围更广的输入像素进行重建。鉴于不同分支间可能存在的信息冲突,设计了特征传递模块,通过跨分支传递全局特征并抑制低频信息,从而确保各分支间的协同作用。在真实世界图像数据集DND上,与基于Transformer的去噪算法(Uformer)相比,所提算法的峰值信噪比(PSNR)提高了0.14dB。在合成图像数据集Urban100上,与多阶段去噪算法(MSPNet)相比,所提算法的平均PSNR提高了0.41dB。实验结果表明,所提算法能够有效去除图像噪声,并重建出更精细的纹理细节。

关键词: Transformer, 图像去噪, 全局特征, 局部特征, 特征解耦

CLC Number: