Robot-based automation in chemistry accelerates material development but lacks safety in unsupervised environments. This paper introduces DenseSSD, a deep learning-based object detector, to improve safety by precisely locating transparent chemical vessels. DenseSSD achieves over 95% mean average precision (mAP), significantly outperforming existing detectors. Its robustness across varying conditions (solution colors, camera angles, backgrounds) makes it suitable for diverse lab settings, enhancing safety in automated material synthesis.
Publisher
npj | computational materials
Published On
Feb 22, 2024
Authors
Leslie Ching Ow Tiong, Hyuk Jun Yoo, Nayeon Kim, Chansoo Kim, Kwan-Young Lee, Sang Soo Han, Donghun Kim
Tags
robot-based automation
object detection
safety
deep learning
chemical vessels
material synthesis
Related Publications
Explore these studies to deepen your understanding of the subject.