Chemistry-Physics Lab Materials Dataset

Laboratory experiments are fundamental to science education, yet resource constraints often limit students’ access to hands-on learning experiences. While object detection technology offers promising solutions for automated material identification and assistance, existing datasets like CABD (21 classes) and Chemical Experiment Image Dataset (7 classes) are limited in scope. We present two comprehensive datasets for laboratory material detection: a Chemistry dataset comprising 1,191 images across 60 classes and a Physics dataset containing 1,749 images across 76 classes. Both datasets feature high-quality images captured in authentic laboratory settings. We validate these datasets using three state-of-the-art object detection models: YOLOv5nu, YOLOv8, and YOLOv11. The models achieve exceptional performance on both datasets, with YOLOv8 demonstrating the highest accuracy, reaching mean Average Precision (mAP@0.5) of 0.99 or 1.0 for most chemistry classes and comparable results for physics materials. Our key contributions include: (1) the first comprehensive open-source dataset of physics laboratory materials; (2) a significantly expanded chemistry dataset that addresses the limitations of existing collections; and (3) extensive validation using multiple YOLO architectures to ensure dataset reliability. These datasets enable the development of robust computer vision applications for laboratory education, potentially improving resource accessibility and student learning outcomes in STEM fields. The datasets[1] and trained models are made publicly available to support further research and development in educational technology.[1] https://github.com/danison2/Chemistry-Physics-Lab-Materials-Dataset