Aircraft targets in remote sensing images are important research objects. Accurately identifying different types of aircraft will bring the greatest value of intelligence into play. In this paper, we combine the YOLOv5 detection algorithm and the MMAL-Net classification algorithm to detect the aircraft target and identify the aircraft type in the remote sensing image. Firstly, we use the YOLOv5 algorithm to determine the area where the aircraft target is located in the complex scene. Then the MMAL-Net algorithm is used to classify the aircraft in the area at the fine-grained level, and the type of aircraft target is identified. Through the experiment, the method proposed in this paper can accurately detect the aircraft target and identify the aircraft type, and the accuracy is 73. 2%.
Research on Aircraft Type Recognition from Remote Sensing Images in Complex Scenes
2021-09-01
1515867 byte
Conference paper
Electronic Resource
English