Comparison of two target detection algorithms based on remote sensing images

Theoretically, the detection accuracy of two-stage target detection algorithm is higher than that of single-stage target detection algorithm.The single-stage target detection algorithm realizes end to end target detection and greatly improves detection speed while losing part of accuracy.However, different networks have different extraction capabilities for remote sensing image features. In this paper, two algorithms, Faster RCNN and YOLOv3, were respectively used to train and test remote sensing image data sets. The results show that YOLO V3 has higher detection accuracy and Faster speed for remote sensing image.