A Novel Method for Extracting Frequency Line on Lofargram based on Feature Function

Passive sonar signals are generated by receiving and processing the complex noise of the surrounding environment, so it is very difficult to extract useful information from them. In this paper, a novel algorithm for extracting frequency lines on lofargram based on feature function is proposed. Firstly, the lofargram is preprocessed before extracting the frequency line, which makes the frequency line in the lofargram more obvious. Secondly, according to the high brightness, continuity and smoothness of frequency line information, density based spatial clustering of applications with noise (DBSCAN) and polynomial fitting are used to set the detection window for feature function. Finally, simulation experiments and real data experiments verify that the proposed algorithm also has good performance in low signal-to-noise ratio lofargram.