Side-scan sonar image target classification method based on style migration

Abstract

The invention belongs to the technical field of side-scan sonar image recognition, and particularly relates to a side-scan sonar image target classification method based on style migration. Accordingto the method, a saliency detection method and a style migration network are used, a conventional optical image is converted into a side-scan-imitating sonar image, the distance between a source fieldand a target field is shortened, the number of basic features capable of being directly migrated is increased, and migration learning efficiency can be effectively improved; meanwhile, a migration learning method is used for migrating a fully-trained deep learning network, and the characteristic that the basic characteristics of the image have universality is utilized, so the number of optimization parameters can be effectively reduced, and the problem that a deep learning technology cannot be applied due to insufficient training samples is avoided. According to the method, a style migrationand transfer learning method is used for migrating a convolutional neural network trained by using the artificially generated side-scan-imitating sonar image, so transfer learning efficiency is improved, and the phenomenon of negative migration is prevented.

Description

Technical field [0001] The invention belongs to the technical field of side scan sonar image recognition, and specifically relates to a method for classifying a side scan sonar image target based on style transfer. Background technique [0002] The automatic classification of targets in side scan sonar images is of great significance for ocean detection and underwater search, especially for the search for crashed aircraft and sunken ships. The current common search method is to use the sonar carried by the AUV to scan a large area of ​​the target seabed, and then after a complete scan of the sea area, the data is copied out, and the existence of the target is judged manually. Because it does not have the ability to detect and recognize the target independently, this The efficiency of this search process is low. Therefore, how to improve the ability of autonomous detection and recognition of sonar images has become increasingly important. [0003] The current sonar image target det...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a transfer learning method based on style transfer using the powerful feature extraction and feature combination capabilities of the deep learning network to overcome the problem that the deep learning technology cannot be applied due to too few training samples of side-scan sonar images. Object Classification Method in Side Scan Sonar Image

Images