Arrested Landing State Recognition of Carrier-Based Aircraft Based on Coordinate Attention and Weighted Bi-directional Feature Pyramid Network

Objectives The successful locking of the tailhook and arresting wires of the carrier-based aircraft is the key to determining the safe landing of the carrier-based aircraft. Therefore, precise recognition of the tailhook status and the arrested landing state recognition of carrier-based aircraft is fundamental for the Landing Signal Officer(LSO) to execute effective commands. To this end, a arrested landing state recognition model based on Coordinate Attention(CA) and weighted Bi-directional Feature Pyramid Network(BiFPN) is proposed, aiming to provide intelligent auxiliary recognition tools for LSO. Methods Firstly, the CA is used to enhance the network's feature extraction ability from both spatial and channel dimensions. Then, BiFPN introduces learnable weights to learn the weights of different input features, repeatedly using top-down and bottom-up multi-scale feature fusion, and finally adopting a C2F lightweight model structure, reduce parameters and computational complexity. Results Through simulation experiments, the proposed model was compared with five baseline models. The experimental finding revealed that our model outperformed the baseline model in detecting the tailhook and arresting wires of the carrier-based aircraft. Conclusions It helps to improve the accuracy and robustness of the tailhook and arresting wires of the carrier-based aircraft detection, and is of great significance for improving the efficiency and safety of carrier-based aircraft landing operations, preventing potential personnel injuries and equipment losses.