摘要
在利用深度学习进行偏振图像计算成像过程中,图像映射函数的解空间极大、空间分辨率一般较低,难以生成清晰的纹理细节且存在高频信息缺失等问题。为解决该问题,提出一种结合双注意力机制的深度残差偏振图像超分辨率网络。该网络由一个具有全局跳跃连接的残差网络组成,包含10个残差组,每个残差组包含20个具有局部跳跃连接的双重注意力块级联的残差块;同时考虑通道间的相互依赖性,设计自适应通道特征调整机制;引入级联的空间注意力块,将残差的特征更集中于关键的空间内容。将所提方法与Bicubic、SRCNN、FSRCNN、EDSR等方法进行对照实验与成像系统对比校正实验,结果表明该方法重建图像纹理细节更加丰富,亮度均匀,较为接近成像系统的高清图像,同时峰值信噪比和结构相似性指标优于其他方法但参数量仅约为EDSR的2/5。
In polarization imaging detection,the difference and change of physical properties of aerosols or detection targets are reflected by polarization characteristics. The high-dimensional polarization characteristics effectively improve the contrast between the target and the background,thereby laying the foundation for realizing the inversion of the target′s spatial structure. This feature can enhance the recognition effect of the target in the cluttered background. Affected by the imaging distance and atmospheric interference,the limit resolution of the image projected on the focal plane is greatly reduced(much smaller than the optical system diffraction limit resolution),resulting in a lower spatial resolution of the polarized image. On the other hand,the spatial resolution of the polarized image is limited by the number of detector pixel. High resolution images are of great significance and value to the accuracy of target detection. For this reason, without replacing the hardware imaging system, the super-resolution reconstruction method is usually adopted. This method is a common technical means in image processing and practical engineering applications,and it is also a hot research issue of underlying computer vision. The existing image super-resolution algorithms have some problems, such as low utilization of feature information,large amount of parameters,blurred image reconstruction details and so on. The input features of low-resolution images contain rich low-frequency information,which is treated equally in different channels. In computational imaging process using deep learning,image mapping function solution space is very large,it is difficult to generate detailed texture and high-frequency information lack,which hinders convolutional neural network representation ability in image super-resolution. In order to solve this problem,a depth residual polarization image super-resolution network combined with double attention mechanism is proposed. This paper proposes a dual attention residual network model.
作者
徐国明
王杰
马健
王勇
刘佳庆
李毅
XU Guoming;WANG Jie;MA Jian;WANG Yong;LIU Jiaqing;LI Yi(School of Internet,Anhui University,Hefei 230039,China;National Engineering Research Center for Agro-Ecological Big Data Analysis&Application,Anhui University,Hefei 230601,China;Anhui Province Key Laboratory of Polarized Imaging Detecting Technology,Army Artillery and Air Defense Forces Academy of PLA,Hefei 230031,China;Institute of Intelligent Technology,Anhui Wenda University of Information Engineering,Hefei 231201,China)
出处
《光子学报》
EI
CAS
CSCD
北大核心
2022年第4期295-309,共15页
Acta Photonica Sinica
基金
国家自然科学基金(No.61906118)
安徽省自然科学基金(Nos.1908085MF208,2108085MF230)
陆军装备部十三五预研子课题,安徽省高校自然科学研究重点项目(No.KJ2019A0906)。
关键词
计算成像
超分辨率
深度残差网络
偏振图像
双重注意力块
Computational imaging
Super-resolution
Depth residual network
Polarization images
Dual attention block
