Enhancing ISAR Image Efficiently via Convolutional Reweighted l₁ Minimization

Area of Science:

• Radar imaging
• Signal processing
• Computational imaging

Background:

• Sparse aperture data in Inverse Synthetic Aperture Radar (ISAR) imaging leads to artifacts like grating and side lobes.
• Standard sparse signal recovery (SSR) for ISAR images often results in isolated scatterers, hindering target structure interpretation.

Purpose of the Study:

• To propose a novel approach for enhancing ISAR images reconstructed from sparse aperture data.
• To address the limitations of existing methods in recognizing target structures due to isolated scatterers.

Main Methods:

• Development of a convolutional reweighted l1 minimization model to capture the structural sparsity of ISAR images.
• Reconstruction of ISAR images by iteratively solving reweighted l1 problems, incorporating neighborhood information.
• Utilizing the alternating direction of multipliers (ADMM) and linearized approximation for computational efficiency.

Main Results:

• The proposed algorithm effectively enhances ISAR images obtained from sparse aperture data.
• The method demonstrates robustness against noise in the imaging process.
• Experimental results confirm the algorithm's high computational efficiency.

Conclusions:

• The convolutional reweighted l1 minimization model successfully models structural sparsity, leading to improved ISAR image quality.
• The proposed method offers a significant advancement in ISAR imaging for sparse aperture scenarios.
• The technique provides a practical and efficient solution for enhanced ISAR image reconstruction.