Abstract: Target recognition is one of the key techniques in underwater acoustic area.This article uses high-order cu-mulant and Hilbert transform for feature extraction,firstly gets the ship radiated noise from target ships,and then ex-tracts the ratio of average instantaneous frequency between neighboring IMFs,relative standard deviation,center fre-quency,average intensity,high-order moment and high-order cumulant of different orders of IMFn(n=1-8),finally re-cognizes and classifies two types of ship targets through BP neural network.Good recognition effect of this method has been verified through the classification tests for the actual ship radiated noise.