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

TL;DR

This paper introduces a new dataset and a specialized neural network model for accurately segmenting phase interfaces in transparent glassware, enabling improved visual monitoring of chemical experiments.

Contribution

It presents the CTG 2.0 dataset and a novel LGA-RCM-YOLO model that enhances phase-interface segmentation accuracy and robustness in laboratory settings.

Findings

Achieves 84.4% AP@0.5 on CTG 2.0 dataset.

Improves segmentation AP by over 8 points compared to baseline.

Demonstrates real-time inference and practical application in process monitoring.

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.