北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2024, Vol. 47 ›› Issue (4): 63-70.

• 体系化人工智能专题 • 上一篇    下一篇

基于雅可比动态逼近的图像去噪网络算法

刘美琴1,姬厚国1,白羽1,姚超2,赵耀1   

  1. 1. 北京交通大学
    2. 北京科技大学

  • 收稿日期:2023-12-12 修回日期:2024-01-29 出版日期:2024-08-28 发布日期:2024-08-26
  • 通讯作者: 姚超 E-mail:yaochao@ustb.edu.cn
  • 基金资助:
    国家自然科学基金项目; 中央高校基本科研业务费专项项目

Image Denoising Network Algorithm Based on Jacobian Dynamic Approximation

LIU Meiqin1, JI Houguo1, BAI Yu1, YAO Chao2, ZHAO Yao1   

  • Received:2023-12-12 Revised:2024-01-29 Online:2024-08-28 Published:2024-08-26

摘要: 受到环境因素和采集设备性能的限制,图像采集易受噪声干扰从而降低用户的视觉体验。为了有效去除图像噪声,提出了基于雅可比动态逼近的端到端图像去噪网络算法。利用常微分方程的思想构建前向微分结构,动态拟合噪声分布,并设计雅可比矩阵的求解模块实现前向求导,降低去噪网络的复杂度。为了提高对复杂噪声的特征表示能力,应用多尺度特征提取模块获取图像中非均匀噪声的特征,并进一步获取图像的上下文语义信息。此外,采用双注意力结构增强重建特征,提高重建图像的质量。大量实验结果表明,所提算法可以有效地去除合成噪声和真实噪声,重建图像在客观评价指标与主观视觉效果上均取得较好的效果。

关键词: 常微分方程 , 图像去噪, 雅可比求解模块, 多尺度特征提取模块, 卷积神经网络

Abstract: Limited by the environment and the acquisition device, the captured images are susceptible to the noise and this leads to the poor visual perception for the users. To systematically remove noise from the images, an end-to-end Jacobian approximation denoising network algorithm is proposed. Specifically, an ordinary differential equation is utilized to construct a forward differential structure for dynamically simulating the noise distribution in the image. A Jacobian matrix-based solution module is designed to implement the forward derivative and reduce the complexity of the denoising network. To enhance the feature representation capability for complex noise, a multi-scale feature extraction module is designed to capture the features of non-uniform noise. Besides, a dual attention structure is used to enhance the reconstructed features and improve the quality of the reconstructed images. Extensive experimental results demonstrate the effectiveness of the proposed algorithm on eliminating synthetic and real noise from images, and the reconstructed image achieves better results in both subjective visual effect and objective evaluation metrics.

Key words: ordinary differential equation ,  image denoising ,   Jacobian solution module ,   multi-scale feature extraction module ,   convolutional neural network

中图分类号: