Non-cooperative underwater target automatic identification method based on Gaussian mixture model

Abstract

The invention discloses a non-cooperative underwater target automatic identification method based on a Gaussian mixture model. The method is used for solving the technical problem that an existing non-cooperative underwater target identification method is poor in practicability. According to the technical scheme, when a target of an unknown category is faced, starting from the perspective of a setinternal and external identification mode, a Mel frequency cepstrum coefficient capable of describing nonlinear characteristics of human auditory sense is extracted by utilizing a known intra-set data structure, namely, the MFCC coefficient, a Gaussian mixture model is trained, appropriate thresholds are selected, and a target identification system is constructed. According to the method, the samples with the label information difficult to obtain are substituted into the target recognition system to distinguish whether the samples belong to the samples in the known sample set or not, the categories of the samples are judged, the categories of the non-cooperative targets are preliminarily judged, meanwhile, the requirements for the label information and the data size of the sample data arereduced, and the practicability is good.

Description

technical field [0001] The invention relates to an underwater target recognition method, in particular to a non-cooperative underwater target automatic recognition method based on a Gaussian mixture model. Background technique [0002] The document "Application Progress of Deep Learning in Passive Recognition of Underwater Targets, Signal Processing, 2019, Vol35(9), p1460-1475" discloses a method for passive recognition of underwater targets based on deep learning. After performing the steps of typical pattern classification and recognition systems such as preprocessing and feature extraction, a specific deep neural network structure is used to realize the classifier design and classifier link, and a certain number of samples are used to complete the model training of the deep neural network, or directly use the depth The neural network has the characteristics of good feature learning ability, and the use of deep neural network weakens or completely replaces the key feature ...

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the shortcomings of poor practicability of existing non-cooperative underwater target recognition methods, the present invention provides a non-cooperative underwater target automatic recognition method based on Gaussian mixture model

Images