Graph Data and GCN Based Maritime Target Detection of Multi-frame Scanning Radar

The research on sea target signal detection methods is valuable in many fields. Affected by the complex sea clutter environment and target features, it is difficult for conventional target detection methods to achieve high-performance detection. When deep learning methods such as convolutional neural networks are applied for radar signal processing, each signal sample is independent of each other, and the spatial-temporal relationship information cannot be effectively utilized. In recent years, Graph Convolutional Networks (GCN) have shown great advantages in graph data processing and have been applied in the field of signal processing. This paper proposes a maritime target detection method based on radar signal graph data and GCN, which converts radar multi-frame scanning signals into data with a graph structure, and uses GCN to perform binary classification on the nodes in the graph data to realize maritime target detection. Experimental results show that the proposed method has better performances that conventional 2D-CACFAR method with lower false alarm rate and higher detection probability.