城市级无人机影像匀色算法

无人机在获取影像时，受到天气、光照条件、相机传感器、主观人为等多种因素的影响，导致获取的影像内部或影像之间存在不同程度的亮度与色调分布不均匀，因此需要对无人机影像进行匀光匀色处理。本文提出了灰度值均值算法、基于相机响应函数的灰度值误差参数算法两种匀光匀色算法，并将其应用于镇江市实景三维数据建设无人机影像处理。试验结果表明，两种算法均能较好地去除色差，使色调过渡更缓和，可改善后期的实景三维整体效果，对城市级实景三维建设相关项目具有较好的借鉴作用。

Uniform light and color algorithm for urban-level UAV remote sensing images

Due to the influence of various factors such as sensor factors, weather factors, human factors, et al, the images obtained by UAVs have uneven distribution of brightness and hue in different degrees during acquisition. The image is processed with uniform light and color. In this study, two color-leveling algorithms are proposed, the mean algorithm and the parameter algorithm based on the camera response function. The algorithms are implemented through Python language programming, which are applied to the UAV images of the real 3D data construction area in Zhenjiang city, and the experimental results are evaluated and analyzed. Experiments show that both algorithms can better remove chromatic aberration, make the tone transition more gentle, and optimize the effect of the 3D scene in the later stage.