Page 223 - Read Online

Research Article | Open Access

Journal of Materials Informatics

Yuan et al. J. Mater. Inf. 2026, 6, 17 DOI:10.20517/jmi.2025.78

Computer vision for efficient object detection and
segmentation in molecular image analysis

Shaoxuan Yuan , Zhiwen Zhu , Jiayi Lu , Liangliang Cai 1,* , Qiang Sun 1
1
1,2
1
Keywords:
YOLOv9, object detection,
instance segmentation,
incremental learning,
surface chemistry

Citation: Yuan, S.; Zhu, Z.;
Lu, J.; Cai, L.; Sun, Q.
Computer vision for efficient
object detection and
segmentation in molecular
image analysis. J. Mater. Inf.
2026, 6, 17.
https://dx.doi.org/10.20517
/jmi.2025.78

Received: 11 Sep 2025
First Decision: 23 Oct
2025 Abstract
Revised: 27 Nov 2025
Accepted: 1 Dec 2025 Image recognition, classification, and analysis of large sets of high-resolution molecular
Published: 31 Mar 2026 images are time-consuming and labor-intensive, even for human experts, due to the lack of
standardized approaches. In recent years, machine learning has emerged as a powerful tool
Academic Editors:
William Yi Wang, Hao Li for automating image data analysis in materials science. In this work, we developed a
Copy Editor: computer vision program for efficient object detection and instance segmentation, offering
Pei-Yun Wang a fast alternative to manual molecular image analysis. By integrating You Only Look Once
Production Editor: version 9 (YOLOv9) with an incremental learning strategy and hyperparameter
Pei-Yun Wang
optimization, the system enables accurate detection, classification, and segmentation of
molecular species across diverse scanning tunneling microscopy datasets. Our results
demonstrate robust performance and minimal forgetting rates across multiple molecular
categories, enabling scalable and updatable surface image analysis workflows. We
anticipate that computer vision methods will see increasing applications in image data
analysis within the field of on-surface chemistry.

INTRODUCTION
In recent years, the rise of artificial intelligence (AI) technologies and advances in

1 Materials Genome Institute, Shanghai University, Shanghai 200444, China.
2 Faculty of Science, Katholieke Universiteit Leuven, Leuven 3001, Belgium.

* Correspondence to: Prof. Liangliang Cai, Materials Genome Institute, Shanghai University, Shanghai 200444, China. E-mail:
cailiangliangmgi@shu.edu.cn

www.oaepublish.com Submit a Manuscript: https://ucenter.oaepublish.com