Int J Performability Eng ›› 2020, Vol. 16 ›› Issue (4): 537-548.

• Orginal Article •

### Ship Target Recognition Technology of Radar High Resolution Range Profile based on Machine Learning

Dan Bo*, Shan Gao, and Zhihong Ji

1. Naval Aviation University, Yantai, 264001, China
• Submitted on  ;  Revised on  ; Accepted on
• Contact: Dan Bo
• About author:Shan Gao is an associate research fellow at the Naval Aviation University of China. His research interests include semi-supervised machine learning and information analysis.
Zhihong Ji is an associate research fellow at the Naval Aviation University of China. His research interests include signal processing and target recognition.

Abstract: The traditional equal interval framing method for high resolution range profiles (HRRP) of ship targets has the problems of large computation and low recognition rate. Therefore, an adaptive framing method based on the power spectrum correlation coefficient is proposed, which can be used for non-equal framing according to the characteristics of the ship. Using this method can effectively reduce recognition and matching templates, greatly reduce the recognition time, and can achieve high efficiency in combat. On this basis, this paper further discusses the features in the time domain and frequency domain extracted from the one-dimensional range profiles of ten typical ships, and it verifies the effectiveness by support vector machine (SVM) and other typical classifiers. The conclusion shows that the recognition rate can be effectively improved by selecting appropriate features and classifiers for different targets.