基于通道注意递归残差网络的图像超分辨率重建

Image Super-resolution Reconstruction Using Recursive Residual Network Based on Channel Attention

近年来,深度学习被广泛应用于图像超分辨率重建。针对基于深度学习的超分辨率重建方法存在的特征提取不充分、细节丢失和梯度消失等问题,提出一种基于通道注意的递归残差深度神经网络模型,用于单幅图像的超分辨率重建。该模型采用残差嵌套网络和跳跃连接构成一种简洁的递归残差网络结构,能够加快深层网络的收敛,同时避免网络退化和梯度问题。在特征提取部分,引入注意力机制来提升网络的判别性学习能力,以提取到更准确、有效的深层残差特征;随后结合并行映射重建网络,最终实现超分辨率重建。在数据集Set5,Set14,B100和Urban100上进行放大2倍、3倍和4倍的重建测试实验,并从客观指标和主观视觉效果上将所提方法与主流方法进行比较。实验结果显示,所提方法在全部4个测试数据集上的客观指标较对比方法均有明显提升,其中,相比插值法和SRCNN算法,在放大2倍、3倍、4倍时所提方法的平均PSNR值分别提升了3.965 dB和1.56 dB、3.19dB和1.42 dB、2.79 dB和1.32 dB。视觉效果对比也表明所提方法能更好地恢复图像细节。

In recent years,deep learning has been widely used in image super-resolution reconstruction.To solve the problems of inadequate feature extraction,loss of details and gradient disappearance in super-resolution reconstruction methods based on deep learning,a deep recursive residual neural network model based on channel attention is proposed for single image super-resolution reconstruction.The proposed model constructs a simple recursive residual network structure by residual nested networks and jump connections to deepen the network and speed up its convergence while avoiding network degradation and gradient problems.An attention mechanism is introduced into the feature extraction part to improve the discriminant learning ability of the network for more accurate and more effective extraction of deep residual features,which is combined with the subsequent reconstruction network with parallel mapping structure to ensure the final accurate reconstruction.Quantitative and qualitative assessments are performed on benchmark dataset Set5,Set14,B100 and Urban100 at the magnification of 2,3 and 4 times by comparison with the mainstream methods.Experimental results show that the objective index values of the proposed method increase significantly compared to the comparative methods on all four test data sets.Among them,compared with the interpolation method and the SRCNN algorithm,the average PSNR improves 3.965 dB and 1.56 dB,3.19 dB and 1.42 dB,2.79 dB and 1.32 dB,respectively,at the magnification of 2,3 and 4 times.Visual effects show that the proposed method can recover image details better.