激光与光电子学进展, 2020, 57 (4): 041504, 网络出版: 2020-02-20
基于残差通道注意力和多级特征融合的图像超分辨率重建 下载: 1219次
Super-Resolution Image Reconstruction Based on Residual Channel Attention and Multilevel Feature Fusion
机器视觉 超分辨率 深度学习 递归结构 分组卷积 残差通道注意力 多级特征融合 machine vision super-resolution deep learning recursive structure group convolution residual channel attention multilevel feature fusion
摘要
针对模型VDSR(very deep super resolution)中存在的忽略特征通道间的相互联系,不能充分利用各层特征,以及参数量过大,计算复杂度过高等问题,本文提出了一种基于残差通道注意力和多级特征融合的图像超分辨率重建网络结构,通过引入残差通道注意力,自适应校正信道的特征响应,提高了网络的表征能力。网络整体使用递归结构,在每个递归块内实现参数共享,减少了参数数量;多级特征融合的方式可以充分提取图像特征;用分组卷积代替传统卷积,进一步减少了参数数量,并降低了计算复杂度。所提算法在保证图像重建质量的同时,减少了模型的参数量并降低了计算复杂度,在图片放大4倍时,参数量和计算复杂度分别约为VDSR的0.33和0.02。
Abstract
The VDSR (very deep super resolution) model has some problems such as neglecting the interconnection between feature channels, inability to fully utilize the features of each layer, excessive parameter quantity, and computational complexity. To solve these problems, this paper proposes a network structure based on a residual channel attention mechanism and multilevel feature fusion. By introducing residual channel attention, the channel's characteristic response is adaptively corrected to improve network representation ability. A recursive structure is adopted in the network and parameter sharing is implemented in each recursive block, which reduces the number of parameters. The proposed multilevel feature fusion method can fully extract image features; traditional convolution is replaced by group convolution to further reduce the number of parameters and computational complexity. The algorithm reduces the number of parameters and complexity of the model while ensuring the quality of image reconstruction. When an image is enlarged four times, parameter quantity and computational complexity are approximately 0.33 and 0.02 times, respectively, those of VDSR.
席志红, 袁昆鹏. 基于残差通道注意力和多级特征融合的图像超分辨率重建[J]. 激光与光电子学进展, 2020, 57(4): 041504. Zhihong Xi, Kunpeng Yuan. Super-Resolution Image Reconstruction Based on Residual Channel Attention and Multilevel Feature Fusion[J]. Laser & Optoelectronics Progress, 2020, 57(4): 041504.