Synthetic aperture radar image target identification method based on depth model

Abstract

The invention discloses a synthetic aperture radar image target identification method based on a depth model. The method comprises steps of image cutting; depth model level structure design, characteristic extraction filter design, parameter quantity control and overfitting prevention, function activation and non-linear lifting, identification classification and autonomous parameter correction update; depth model training; target identification. The method is advantaged in that filter parameters can realize autonomous iteration update in a training process, characteristic selection and extraction cost is greatly reduced, moreover, target different-level characteristics can be extracted through the depth model, the characteristics can be acquired through high-degree matching and training, so high-degree target representation can be realized, and target identification accuracy of SAR images is improved.

Description

technical field [0001] The invention relates to machine learning and deep learning application technology, in particular to the application of deep learning methods in synthetic aperture radar image target recognition. Background technique [0002] Synthetic Aperture Radar (SAR for short) can provide high-resolution images all day long under all weather conditions. The main method of the current SAR image target recognition system is to use the characteristics of the image target to train the classifier, and realize the SAR image target recognition through the classifier, so the performance of the classifier determines the recognition ability of the recognition system. [0003] The selection and extraction of features has a great influence on the design and performance of classifiers. Pattern recognition is the process of classifying specific things into a specific category, that is, first use a certain number of samples, design a classifier according to the similarity betw...

AI Technical Summary

Problems solved by technology

At the same time, the depth model has good parallel processing ability and learning ability, which can deal with the complex problem of SAR image environment information, and even allows the sample to have a large displacement, stretching, rotation, etc.

Images