Phase-Interface Instance Segmentation as a Visual Sensor for Laboratory Process Monitoring
This addresses the challenge of reliable monitoring in laboratory automation for chemical processes, but it is incremental as it builds on existing YOLO models.
The paper tackles the problem of visual monitoring in chemical experiments by formulating it as phase-interface instance segmentation, introducing the CTG 2.0 dataset and proposing LGA-RCM-YOLO, which achieves 84.4% AP@0.5 and 58.43% AP@0.5-0.95, improving over a baseline by 6.42 and 8.75 AP points.
Reliable visual monitoring of chemical experiments remains challenging in transparent glassware, where weak phase boundaries and optical artifacts degrade conventional segmentation. We formulate laboratory phenomena as the time evolution of phase interfaces and introduce the Chemical Transparent Glasses dataset 2.0 (CTG 2.0), a vessel-aware benchmark with 3,668 images, 23 glassware categories, and five multiphase interface types for phase-interface instance segmentation. Building on YOLO11m-seg, we propose LGA-RCM-YOLO, which combines Local-Global Attention (LGA) for robust semantic representation and a Rectangular Self-Calibration Module (RCM) for boundary refinement of thin, elongated interfaces. On CTG 2.0, the proposed model achieves 84.4% AP@0.5 and 58.43% AP@0.5-0.95, improving over the YOLO11m baseline by 6.42 and 8.75 AP points, respectively, while maintaining near real-time inference (13.67 FPS, RTX 3060). An auxiliary color-attribute head further labels liquid instances as colored or colorless with 98.71% precision and 98.32% recall. Finally, we demonstrate continuous process monitoring in separatory-funnel phase separation and crystallization, showing that phase-interface instance segmentation can serve as a practical visual sensor for laboratory automation.