Confidence Guided Network For Atmospheric Turbulence Mitigation
Confidence Guided Network For Atmospheric Turbulence Mitigation
复制标题
用于缓解大气湍流的置信引导网络
DOI：
10.1109/icip42928.2021.9506125
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Vishal M. Patel
中科院分区：
文献类型：
--
作者：
Nithin Gopalakrishnan Nair;Vishal M. Patel
大气湍流会对远程成像系统捕获的图像或视频的质量产生不利影响。湍流会导致图像中的几何和模糊失真，进而导致后续计算机视觉算法（例如识别和检测）的性能不佳。现有的大气湍流缓解方法使用配准和反卷积方案来消除退化。在本文中，我们提出了一种基于深度学习的解决方案，其中使用基于有效最近邻（ENN）的方法进行配准，并使用基于不确定性的网络进行恢复。我们使用合成数据集和真实数据集进行定性和定量比较，以显示我们工作的重要性。
Atmospheric turbulence can adversely affect the quality of images or videos captured by long range imaging systems. Turbulence causes both geometric and blur distortions in images which in turn results in poor performance of the subsequent computer vision algorithms like recognition and detection. Existing methods for atmospheric turbulence mitigation use registration and deconvolution schemes to remove degradations. In this paper, we present a deep learning-based solution in which Effective Nearest Neighbors (ENN) based method is used for registration and an uncertainty-based network is used for restoration. We perform qualitative and quantitative comparisons using synthetic and real-world datasets to show the significance of our work.