Phase-Interface Instance Segmentation as a Visual Sensor for Laboratory Process Monitoring

arXiv cs.CV / 3/12/2026

📰 NewsTools & Practical UsageModels & Research

共有:

Key Points

It presents phase-interface instance segmentation as a visual sensor for monitoring chemical experiments in transparent glassware.
It introduces the CTG 2.0 dataset with 3,668 images, 23 glassware categories, and five multiphase interface types for benchmarking.
It proposes LGA-RCM-YOLO, combining Local-Global Attention and a Rectangular Self-Calibration Module to refine boundaries, achieving 84.4% AP@0.5 and 58.43% AP@0.5-0.95 and outperforming the YOLO11m baseline by 6.42 and 8.75 AP points.
It demonstrates near real-time inference at 13.67 FPS on an RTX 3060 and an auxiliary color-attribute head that achieves 98.71% precision and 98.32% recall, enabling continuous process monitoring in separatory-funnel phase separation and crystallization as a practical visual sensor for lab automation.

Abstract

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.