宽型自注意力融合密集型残差网络的图像去雾  

Wide Self-attention Mechanism Fusion Dense Residual Network Image Dehazing

作　　者：邬开俊[1] 丁元 WU Kaijun;DING Yuan(School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730073,China)

机构地区：[1]兰州交通大学电子与信息工程学院,甘肃兰州730070

出　　处：《湖南大学学报（自然科学版）》2023年第8期13-22,共10页Journal of Hunan University:Natural Sciences

基　　金：国家自然科学基金资助项目(61966022);甘肃省自然科学基金资助项目(21JR7RA300);甘肃省敦煌文物保护研究中心开放课题(GDW2021YB15)。

摘　　要：当前去雾算法无法很好解决不均匀雾霾图像去雾的问题,为此提出了一种宽型自注意力融合的条件生成对抗网络图像去雾算法.在算法中加入了宽型自注意力机制,使得算法可以为不同雾度区域特征自动分配不同权重;算法特征提取部分采用DenseNet融合自注意力网络架构,DenseNet网络在保证网络中层与层之间最大程度的信息传输的前提下,直接将所有层连接起来,获取更多的上下文信息,更有效利用提取的特征;融合自注意力可以从编码器部分提取的特征中学习复杂的非线性,提高网络准确估计不同雾度的能力.算法采用Patch判别器,增强去雾图像的局部和全局一致性.实验结果证明,算法网络在NTIRE 2020、NTIRE2021和O-Haze数据集上的定性比较,相比于其他先进算法得到更好的视觉效果;定量比较中,相较于所选择先进算法的最好成绩,峰值信噪比和结构相似性指数分别提高了0.4和0.02.The current defogging algorithm cannot solve the problem of uneven haze image defogging,so this paper proposes a wide self-attention fusion conditional generation against network image defogging algorithm.The wide self-attention mechanism is added to the algorithm,so that the algorithm can automatically assign different weights to the features of different haze regions.The feature extraction part of the algorithm adopts the DenseNet fusion self-attention network architecture.Under the premise of ensuring the maximum information transmission between the middle layers of the network,the DenseNet network directly connects all layers to obtain more context information and make more effective use of the extracted features.Fusion of self-attention can learn complex nonlinearity from the features extracted from the encoder part,and improve the ability of the network to accurately estimate different haze.The algorithm uses Patch discriminator to enhance local and global consistency of defogging images.The experimental results show that the qualitative comparison of the algorithm network on NTIRE 2020,NTIRE 2021 and O-Haze datasets has better visual effects than other advanced algorithms. In the quantitativecomparison, compared with the best performance of the selected advanced algorithms, the peak signal-to-noiseratio and the structural similarity index increases by 0.4 and 0.02, respectively.

关 键 词：图像去噪 图像去雾 生成对抗网络 宽型自注意力机制 马尔科夫判别器 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术]