Articles | Volume 22, issue 2
Research article
25 Jan 2022
Research article |  | 25 Jan 2022

New particle formation event detection with Mask R-CNN

Peifeng Su, Jorma Joutsensaari, Lubna Dada, Martha Arbayani Zaidan, Tuomo Nieminen, Xinyang Li, Yusheng Wu, Stefano Decesari, Sasu Tarkoma, Tuukka Petäjä, Markku Kulmala, and Petri Pellikka

Data sets

Smart-SMEAR: on-line data exploration and visualization tool for SMEAR stations H. Junninen, A. Lauri, P. Keronen, P. Aalto, V. Hiltunen, P. Hari, and M. Kulmala

Model code and software

maskNPF P. Su, J. Joutsensaari, L. Dada, M. A. Zaidan, T. Nieminen, X. Li, Y. Wu, S. Decesari, S. Tarkoma, T. Petäjä, M. Kulmala, and P. Pellikka

Short summary
We regarded the banana shapes in the surface plots as a special kind of object (similar to cats) and applied an instance segmentation technique to automatically identify the new particle formation (NPF) events (especially the strongest ones), in addition to their growth rates, start times, and end times. The automatic method generalized well on datasets collected in different sites, which is useful for long-term data series analysis and obtaining statistical properties of NPF events.
Final-revised paper