Researchers have developed RCL-Mamba, a novel dual-domain State Space Model designed to improve image restoration for Rotational Scanning Computed Laminography (RCL). This method addresses rotational blur in projection data and sparse artifacts in image reconstructions, common issues in rapid non-destructive testing. By employing a cascaded joint processing strategy and a Mamba-CNN dual-branch module, RCL-Mamba effectively corrects blur and suppresses artifacts while preserving fine details. Evaluations show it significantly outperforms existing methods and can reduce scanning views by up to 8-fold without quality degradation, enhancing inspection efficiency for components like printed circuit boards. AI
IMPACT This research offers a more efficient and accurate method for 3D imaging in industrial inspection, potentially speeding up quality control processes.
RANK_REASON Academic paper detailing a new model and its application. [lever_c_demoted from research: ic=1 ai=1.0]
- Mamba-CNN
- nondestructive testing
- printed circuit board
- RCL-Mamba
- Rotational Scanning Computed Laminography
- State Space Model
- Xuyang Duan
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →