Advancing from Automated to Autonomous Beamline by Leveraging Computer Vision

2025/06/01 by Baolu Li, Li, Baolu, Hongkai Yu +11
Computer Science · Earth and Planetary Sciences · Engineering · #3D Surveying and Cultural Heritage #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image and Object Detection Techniques #Satellite Image Processing and Photogrammetry

paper · pdf · doi:10.48550/arxiv.2506.00836

openalex publication_date 2025/06/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The synchrotron light source, a cutting-edge large-scale user facility, requires autonomous synchrotron beamline operations, a crucial technique that should enable experiments to be conducted automatically, reliably, and safely with minimum human intervention. However, current state-of-the-art synchrotron beamlines still heavily rely on human safety oversight. To bridge the gap between automated and autonomous operation, a computer vision-based system is proposed, integrating deep learning and multiview cameras for real-time collision detection. The system utilizes equipment segmentation, tracking, and geometric analysis to assess potential collisions with transfer learning that enhances robustness. In addition, an interactive annotation module has been developed to improve the adaptability to new object classes. Experiments on a real beamline dataset demonstrate high accuracy, real-time performance, and strong potential for autonomous synchrotron beamline operations.

Citations

• Materials Discovery in Combinatorial and High-throughput Synthesis and Processing: A New Frontier for SPM
• Closing the loop: Autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments
• Micrograph segmentations for DDEVD
• Segment Anything
• Associating Objects with Transformers for Video Object Segmentation
• Adversarial Style Mining for One-Shot Unsupervised Domain Adaptation
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal\n Networks
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Related

• Virtual Scientific Companion for Synchrotron Beamlines: A Prototype
• Machine Learning For Beamline Steering
• First Steps Toward Incorporating Image Based Diagnostics Into Particle Accelerator Control Systems Using Convolutional Neural Networks
• Action-Attentive Deep Reinforcement Learning for Autonomous Alignment of Beamlines
• Beam Detection Based on Machine Learning Algorithms