摘要
无人机航拍、测绘等技术以其诸多优点,得到越来越广泛的应用。但在部分地区因天气条件或大气污染等因素雾、霾天较多,使得采集的图像严重降质。针对该问题,提出一种基于物理模型的无人机遥感图像去雾算法。首先,通过比较无人机遥感图像和普通景物图像,归纳无人机遥感图像的主要特点,根据这些特点并结合现有一些去雾算法的优点,给出了大气光矢量的模、大气光矢量的方向、全局透射率等去雾参数的计算方法。实验结果显示,该算法能够在多个指标上取得较好的效果。这表明该算法有效可行。
UAV aerial photography, mapping and other technologies, are being more and more widely used for their specific advantages. But in some areas, more fog or haze days severely degrade the images because of weather conditions or atmospheric pollution or other factors. For this problem, proposes the UAV remote sensing image dehazing algorithm based on the dark-channel. First, by compared the UAV remote sensing image with the general scene image, summarizes the main features of the UAV remote sensing image. Then, according to these features and combining the advantages of these algorithms, gets the methods of computing the airlight orientation, airlight magnitude and global transmission. The results show that, the proposed algorithm can achieve nice effect on multiple indicators. It means that our algorithm is effective and feasible.
基金
国家自然科学基金项目(No.41161065)
贵州省科学技术厅
贵州师范大学联合科技基金资助项目(黔科合J字LKS[2013]28号)
