A high-rank synthetic aperture radar echo focusing method based on one-dimensional kernelization and structured matrix completion
By employing a one-dimensional kernelization and structured matrix completion method, the problems of image blurring and artifacts in SAR echo undersampling imaging were solved, achieving high-fidelity target reconstruction and focusing, and improving imaging quality.
CN122345846APending Publication Date: 2026-07-07UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-26
- Publication Date
- 2026-07-07
AI Technical Summary
Technical Problem
Existing SAR undersampling imaging methods suffer from severe image blurring and artifacts in dense target scenes, failing to effectively improve imaging quality.
Method used
A method based on one-dimensional kernelized and structured matrix completion (1-D KSMC) is adopted. Undersampled SAR data is projected to a high-dimensional space through one-dimensional Hankel mapping and RBF kernel function. Matrix completion is performed by utilizing low-rank characteristics. The objective function is solved alternately by coordinate descent method to recover complete SAR echo data.
Benefits of technology
It effectively suppressed the blurring and artifacts caused by undersampling, achieved high-fidelity target reconstruction and focusing, and improved imaging quality.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122345846A_ABST
Abstract
The application discloses a high-rank synthetic aperture radar echo focusing method based on one-dimensional kernelization and structured matrix completion, and is applied to the technical field of radar signal processing and imaging, and aims at the problems of blurred reconstructed images and serious artifacts of existing SAR echo undersampling imaging methods in dense target scenes due to high-rank characteristics of echoes; the application innovatively combines one-dimensional Hankel mapping and kernelization matrix completion, projects undersampling SAR data into a higher dimension by using RBF kernel, and makes high-rank echo data which cannot be processed by a traditional matrix completion method have low-rank recoverability in a transformation space.
Need to check novelty before this filing date? Find Prior Art