Low-Light Image Enhancement Algorithm for Unmanned Aerial Vehicle Imagery via Color Distribution

Unmanned aerial vehicle (UAV) imagery captured under low-light conditions are often suffering from uneven illumination and lack of detail. To address these issues, this paper proposes a low-light image enhancement algorithm for UAV imagery, which locate different illumination regions with local color distribution. Firstly, a bright channel prior(BCP) is utilized to train the enhancement network by unsupervised learning approach. Secondly, a color distribution module is constructed to locate and enhance different illumination regions. Experimental results show that the proposed method can preserve the detailed information in images while enhancing image brightness, and enhance different illumination regions in low-light UAV images.