SAR image ship target identification method and system

Abstract

The invention relates to an SAR image ship target identification method. According to the SAR image ship target identification method, training data are preprocessed through a proposed inter-class sample imbalance processing technology (including up-sampling processing based on data enhancement and a method for generating batches in proportion), and the training data sent into a network are kept in inter-class balance while the diversity of the training data is improved; through the provided dense residual network used for SAR image ship target identification, reutilization of original features can be realized while more new features are learned; through a proposed loss function based on center loss, network model parameters are adjusted, and simultaneous optimization of intra-class compactness and inter-class separability is realized. A recognition result on OpenSARShip shows that the designed dense residual network has higher accuracy, smaller model size and smaller calculation amount compared with a conventional neural network.

Description

technical field [0001] The invention relates to the technical fields of target recognition and radar remote sensing, in particular to a method and system for recognizing a ship target in a SAR image. Background technique [0002] Synthetic Aperture Radar (SAR) uses the movement of the remote sensing platform to install a small-aperture antenna on the side of the platform to form an equivalent large-aperture antenna through movement to improve the azimuth resolution. It is an all-weather and all-weather sensor that can generate high-resolution SAR images and is widely used in military and civilian fields. With the increase of available SAR images, the automatic and intelligent interpretation of SAR images becomes more and more important, and ships, as important military and civilian targets, need to be focused on. Object recognition has been extensively studied, and many achievements have emerged. At present, ship target recognition methods in SAR images mostly refer to gen...

AI Technical Summary

Problems solved by technology

The currently public dataset for SAR image target recognition is only available in OpenSARShip, but this dataset has the problem of unbalanced samples between categories, which leads to a serious decline in recognition performance

Moreover, the existing classic CNN is specially designed for three-channel natural images, but when applied to SAR images, there are problems of model parameter redundancy and large amount of calculation.

Images