基于注意力机制的多尺度特征融合图像去雨方法  

Image rain removal via multi-scale feature fusion based on attention mechanism

作　　者：刘忠洋 周杰[1] 陆加新 缪则林 邵根富[3] 江凯强 高伟 LIU Zhongyang;ZHOU Jie;LU Jiaxin;MIAO Zelin;SHAO Genfu;JIANG Kaiqiang;GAO Wei(School of Artificial Intelligence(School of Future Technology),Nanjing University of Information Science&Technology,Nanjing 210044,China;School of Electronics&Information Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,China;School of Automation,Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]南京信息工程大学人工智能学院(未来技术学院),南京210044 [2]南京信息工程大学电子与信息工程学院,南京210044 [3]杭州电子科技大学自动化学院,杭州310018

出　　处：《南京信息工程大学学报（自然科学版）》2023年第5期505-513,共9页Journal of Nanjing University of Information Science & Technology（Natural Science Edition）

基　　金：国家自然科学基金(61971167,62101275,62101274);江苏省信息与通信工程优势学科建设项目。

摘　　要：雨纹分布和形状具有多样性,现有去雨算法在去雨的同时会产生图像背景模糊、泛化性能差等问题.因此,本文提出一种基于注意力机制的多尺度特征融合图像去雨方法.特征提取阶段由多个包含两个多尺度注意力残差块的残差组构成,多尺度注意力残差块利用多尺度特征提取模块提取及聚合不同尺度的特征信息,并通过坐标注意力进一步提高网络的特征提取能力.在组内进行局部特征融合,组间利用全局特征融合注意力模块更好地融合不同层次的特征,通过像素注意力使网络重点关注于雨纹区域.在仿真和真实雨像数据集上与其他现有图像去雨算法相比,本文方法的定量指标有着明显提高,去雨后的图像视觉效果较好且具有良好的泛化性.Due to the diversity of the distribution and shape of rain streaks,existing rain removal algorithms produce problems such as blurred image background and poor generalization performance while removing rain.A multi-scale feature fusion image rain removal approach based on attention mechanism is proposed to address these problems.The feature extraction consists of multiple residual groups containing two multi-scale attention residual blocks,which use the multi-scale feature extraction module to extract and aggregate feature information at different scales and further improve the feature extraction capability of the network through coordinate attention.Local feature fusion is performed within groups,and the global feature fusion attention module is used between groups to better fuse features at different levels and to focus the network on rain streak regions through pixel attention.The quantitative metrics of the proposed approach are significantly improved compared with other existing image rain removal algorithms on both simulated and real rain image datasets,and the rain removal images are greatly improved in both visual effects and generalization performance.

关 键 词：图像去雨 多尺度 特征融合 残差网络 坐标注意力 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]