Polarimetric SAR image classification method and system, medium and program product

By constructing a residual neural network model and performing two-stage training, combining anchor samples and positive sample pair data sets, the Hungarian matching algorithm is used to generate polarized SAR image classification results, which solves the classification accuracy and label dependence problems in the existing technology, and realizes efficient classification under labelless data.

CN120388283AActive Publication Date: 2025-07-29ANHUI UNIV
6 Cites 0 Cited by

Patent Information

Application Number
CN202510453982.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-11
Publication Date
2025-07-29
Estimated Expiration
2045-04-11

AI Technical Summary

Technical Problem

The existing polarized SAR image classification method relies on a simple scattering model, and the classification accuracy is difficult to meet the actual needs. The deep unsupervised learning method is rarely used on PolSAR data and has discontinuous semantics. The traditional unsupervised method has fast computing speed but not ideal accuracy.

Method used

The residual neural network model is constructed, and the data set is trained through a two-stage training process using anchor sample training set and positive samples. The polarized SAR image classification results are generated in combination with the Hungarian matching algorithm to avoid dependence on the annotated data.

Benefits of technology

It realizes accurate classification and recognition of polarized SAR images without labeled data, improves classification accuracy and solves the accuracy and dependence problems of traditional methods.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a polarized SAR image classification method and system, a medium and a program product, and relates to the field of polarized SAR image classification, and the method comprises the steps: obtaining a polarized data set; constructing an anchoring sample training set and a positive sample pair data set according to the polarization data set; constructing a residual neural network model; performing first-stage training on the residual neural network model by using the positive sample pair data set to obtain a middle residual neural network model which passes through the first-stage training; performing second-stage training on the intermediate residual neural network model by using the polarization data set and the anchoring sample training set to obtain a residual neural network model through the second-stage training; and clustering the to-be-classified polarized SAR images by using the residual neural network model trained in the second stage and generating a classification result graph. According to the method, the polarimetric SAR image can be effectively and accurately classified and identified under the condition that any marking cost is not needed.
Need to check novelty before this filing date? Find Prior Art