Research on Fast Target Detection for X-band Marine Radar Image Based on RBF Neural Network

Currently, X-band marine radar images are usually used to detect ships, islands and other targets on the sea. The Constant False Alarm Rate (CFAR) detection algorithm, as a classic target detection algorithm, is widely used in marine radar target detection. However, traditional CFAR detection algorithms have a long time for target detection, high false alarm rate, and are difficult to detect targets in sea clutter background. In response to the above issues, this article proposes a fast detection algorithm for maritime radar targets based on a combination of radial basis function (RBF) neural network and CFAR. This algorithm can improve the efficiency and accuracy of target detection in sea clutter background, so that ships can quickly and accurately discover targets in the marine military battlefield to obtain battlefield initiative. At the same time, it can ensure rapid and accurate detection of targets under extreme conditions of high waves, effectively avoiding ship collisions with rocks, and ensuring safe navigation. This algorithm first obtains an RBF neural network model and uses it for coarse target detection. Then, under the assumption that the sea clutter follows a Gaussian distribution, CFAR is used for precise target detection of potential target data. Improve the efficiency and accuracy of target detection by combining coarse detection and fine detection. In addition, to ensure the accuracy of target detection, it is necessary to preprocess the image. This article uses a mean filtering method to denoise the original radar image. The experimental results in this paper show that the method is feasible and can effectively detect targets from X-band marine radar images.