基于Retinex对UAV单幅侦察图像的去雾算法研究

针对现有的暗原色先验算法和Retinex算法在处理无人机侦察影像时存在的运算速度慢、光晕效应、细节丢失的问题,提出一种改进后的Retinex算法。首先将RGB颜色空间转化到HIS颜色空间,分离出亮度Ⅰ,对大中小三个尺度赋值,利用高斯环绕求得三种尺度亮度图像后加权,再将亮度Ⅰ转化回RGB颜色空间,以求得到细节信息更加清晰的复原图像。最后在MATLAB中进行仿真实验,并通过改变参数调整图像复原效果。

Aiming at the problems such as slow calculation speed, halo effect and loss of details of the existing Dark Channel Prior algorithm and Retinex algorithm in processing UAV reconnaissance images, an improved Retinex algorithm is proposed. Firstly, the RGB color space is transformed into the HIS color space, the Brightness Ⅰ is separated, and the values are assigned to the large, medium and small scales. The three scale Brightness Images are weighted by Gaussian encirclement,and then the Brightness Ⅰ is converted back to the RGB color space, in order to obtain the restored images with clearer and more detailed information. Finally, the simulation experiment is carried out in MATLAB, and the image restoration effect is adjusted by changing the parameters.