Fine-grained Classification of YOLOv5 Remote Sensing Aircraft Targets Incorporating Broad Learning System
Hong Xue, Xiyuan Wang, Meng Yuan, Xueqin Wang · 2023
In the paper, a method of fusing broad learning image classification network and YOLOv5 classification algorithm to improve the fine-grained classification ability of the model was proposed. Firstly, a broad learning hybrid stack model was constructed using the broad learning system and its variants as well as the stack structure. Then, the aircraft model features extracted from the YOLOv5 classification algorithm network were used for training, and the precision of fine-grained classification was improved by enhancing the ability of feature extraction and nonlinear expression of the algorithm. The experimental results show that the proposed method improves the accuracy of YOLOv5 algorithm’s aircraft type classification by 0.8%. Compared with other algorithms, the proposed algorithm is more suitable for deployment on the mobile devices.