Radar Autonomous Recognition Technology of Sea Target

With the development of artificial intelligence, radar technology and the increasingly demand for unmanned combat, the research on radar target autonomous recognition (RATR) has been developing continuously. This paper summarizes the key problems and research status in the field of radar recognition of sea target, summarizes the research progress of various common RATR methods, and puts forward three RATR schemes based on multi information radar recognition result fusion, reliable pseudo label and transfer learning, so as to solve the problem of few training samples in the process of radar autonomous recognition of sea target. Carry out the simulation verification of radar target recognition based on transfer learning, and the simulation results show that the recognition accuracy can reach 97%. It can be concluded that using transfer learning can effectively overcome the problem of small number of samples and improve the accuracy of marine target radar recognition.