Polarimetric SAR image classification method, device and equipment based on complex and real numbers

By extracting the real and complex polarized features of polarized SAR images and building an image classification composite network for training, the problems of incomplete information utilization and feature loss in the prior art are solved, and higher image classification accuracy and adaptability are achieved.

CN119785124BActive Publication Date: 2025-05-23NAT UNIV OF DEFENSE TECH
1 Cites 0 Cited by

Patent Information

Application Number
CN202510284429.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-03-11
Publication Date
2025-05-23
Estimated Expiration
2045-03-11

AI Technical Summary

Technical Problem

The existing polarized SAR image classification methods mainly use complex neural networks, which only consider complex scenarios, resulting in incomplete information utilization and loss of features carried by real-number parts, reducing adaptability to actual scenarios.

Method used

A polarized SAR image classification method based on complex and real numbers is proposed. By obtaining the fully polarized SAR sample data for pre-processing, multiple polarized features are extracted by reflective symmetric decomposition method, and divided into real and complex polarized features, corresponding training data is constructed, and the image classification complex network is input for training, and finally classified by fusing the output features of real and complex branch networks.

Benefits of technology

It effectively improves the classification accuracy of polarized SAR images, fully explores the high-order polarization information of land objects, and improves the adaptability to actual scenes.

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

Abstract

The present application relates to a method, device and equipment for polarization SAR image classification based on complex and real numbers, which preprocesses the full polarization SAR sample data, extracts multiple polarization features by using the reflection symmetry decomposition method and performs normalization processing, and divides them into real polarization features and complex polarization features, and constructs corresponding training data, and inputs the training data into an image classification composite network for training, wherein the image classification composite network includes a real branch network and a complex branch network that process real polarization features and complex polarization features respectively, and a fusion classification network that fuses the output features of the real branch network and the complex branch network and classifies according to the fused features, and inputs the real polarization features and the complex polarization features corresponding to the full polarization data into the trained image classification composite network to realize image classification. The present method can effectively improve the classification accuracy of polarization images.
Need to check novelty before this filing date? Find Prior Art