Research on a UAV Image Defogging Algorithm with Optimized Transmittance
The inhomogeneous fog in the images obtained by unmanned aerial vehicle(UAV)causes problems such as loss of detail and color distortion in the dehazing results of traditional algorithms,which greatly limits the application value of images.In order to address this problem,this paper proposes an inhomogeneous fog algorithm for UAV image defogging with improved transmittance.Firstly,the transmittance is calculated by combining the color attenuation model and the atmospheric scattering model,and then the transmittance is refined by improving the guided filtering,which in turn enhances the image edge information.Finally,the image after defogging is enhanced to improve the clarity and contrast of the image.The experimental results show that the algorithm proposed in this paper performs well in removing inhomogeneous fog.The method of removing inhomogeneous fog from UAV images by calculating the transmittance of foggy and fog-free regions of the same image not only promotes the theoretical development of de-fogging algorithms,but also facilitates the application of UAVs to the situation of rapid monitoring and rescue of disasters in foggy days.
NETL
NSTL
万方数据
