RCL-Mamba: A Dual-domain State Space Model for Measurement-oriented Image Restoration in Rotational Sparse-View Scanning Computed Laminography

2026-06-30

AI Summary

This work addresses the cross-domain degradation in rotational sparse-view computed laminography (RCL), where rotation-induced blur in the projection domain coexists with sparse-view artifacts in the image domain. We propose RCL-Mamba, the first dual-domain collaborative framework that integrates state space models into RCL reconstruction. Our approach employs a cascaded strategy: it first corrects rotational blur in the projection domain and subsequently suppresses sparse artifacts in the image domain. A novel Mamba-CNN dual-branch module adaptively balances large-scale blur correction with fine local detail recovery. Evaluated with only 64 projection views—down from the conventional 512—RCL-Mamba achieves significantly superior reconstruction quality compared to existing methods, enhances inspection throughput by approximately eightfold, and accurately preserves critical structural features such as via holes, pad boundaries, and fine trace contours, ensuring high measurement fidelity in high-throughput industrial scenarios.

Abstract

Rotational Scanning Computed Laminography (RCL) is widely utilized for the Non-Destructive Testing(NDT) of large planar components. However, to facilitate rapid inspection, continuous sparse-view scanning is often employed, where the angular integration effect during exposure induces rotational blur in the projection domain. Furthermore, the data incompleteness inherent in sparse sampling manifests as sparse artifacts in the reconstructed image domain. To address these cross-domain degradations, this paper proposes RCL-Mamba, a measurement-oriented dual-domain State Space Model (SSM)-based image restoration network. The framework adopts a cascaded joint processing strategy: it first corrects the rotational blur in the projection domain and subsequently suppresses the sparse artifacts in the image domain. Additionally, we design a Mamba-CNN dual-branch module to adaptively balance large-scale blur correction with local detail recovery. Evaluations on both simulated datasets and real-world Printed Circuit Board (PCB) scans demonstrate that RCL-Mamba outperforms existing baselines in blur removal, artifact suppression, and structural preservation. Line-profile-based structural measurement further verifies that the proposed method better preserves via/pad boundaries and slender trace profiles. Crucially, by reducing the required scanning views from 512 to 64, our method enhances inspection efficiency by approximately 8-fold without compromising reconstruction quality, offering a robust measurement-oriented restoration solution for high-throughput RCL inspection with improved structural measurement fidelity.

Problem

• Rotational Scanning Computed Laminography
• sparse-view scanning
• rotational blur
• sparse artifacts
• image restoration

Innovation

• State Space Model
• Dual-domain Restoration
• Rotational Scanning Computed Laminography
• Measurement-oriented Imaging
• Sparse-view Reconstruction