一种基于深度估计模型的图像加雾软件  

An Image Fogging Software Based on Depth Estimation Model

作　　者：王旭光 张崇 田珊珊 白康 WANG Xu-guang;ZHANG Chong;TIAN Shan-shan;BAI Kang(Hebei Technology Innovation Center of Simulation&Optimized Control for Power Generation,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学河北省发电过程仿真与优化控制工程技术研究中心,河北保定071003

出　　处：《计算机仿真》2024年第1期232-237,242,共7页Computer Simulation

基　　金：国家科学自然基金项目(62076093)。

摘　　要：随着深度学习方面的研究不断深入,数据的重要性也上升到新的高度。在较为火热的图像去雾领域,去雾模型性能强弱很大程度上取决于雾天图像数据集的好坏。而现有数据集难以满足科研的需求,同时制作这样的数据集代价高昂,因此利用无雾图像合成真实雾天图像的研究显得尤为重要。在大气散射模型的基础上,引入单目图像深度估计的方法,实现无雾图像加雾的处理,同时比较了深度估计方法的鲁棒性,推荐了适合用于图像加雾的深度估计方法,实现了操作简单的图像加雾软件。With the deepening of deep learning research,the importance of data has risen to a new height.In the hot field of image defogging,the performance of the defogging model largely depends on the quality of the image dataset on foggy days.However,the existing data sets are difficult to meet the needs of scientific research,and the cost of making such data sets is high.Therefore,it is particularly important to synthesize real foggy images using fog-free images.Based on the atmospheric scattering model,the method of monocular image depth estimation is introduced to realize the fogging process of fogless images.Meanwhile,the robustness of the depth estimation method is compared,and the depth estimation method suitable for image fogging is recommended.Finally,the image fogging software with simple operation is realized.

关 键 词：大气散射模型 单目图像深度估计 图像合成雾 图像加雾软件 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]