Improved YOLOv3 Algorithm for Airport Detection under Complex Conditions

Airport target detection in optical remote sensing images is a hot topic in image processing and machine vision in recent years. In this paper, an improved YOLOv3 target detection algorithm is proposed for the problem that the traditional target detection method is insufficient for airport detection under complex conditions. Firstly, the remote sensing image data set of complex background, multi-scale target, multi-objective, multi-category and different perspectives are constructed independently, which lays a foundation for the training of the model. Then the YOLv3 algorithm is improved for the target characteristics in the data set, so that The model can extract more deep-separated features of the target and play a better training effect. Finally, the effectiveness and significance of the algorithm are verified by comparison with other algorithms. The results show that the algorithm can achieve Real-time detection and gets a high detection rate.