EUVIMEDEUVIMED entdecken
Crossref

Accurate felt‐tip pen brands classification based on a convolutional neural network using data augmentation

Xiaobin Wang, Lei Yang, Ruili Chen, Wei Guo, Xun Han, Aolin Zhang

Journal of Forensic Sciences · 2025-01 · 70 (1) · 170-178

DOI: 10.1111/1556-4029.15658 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Forensic Sciences, online erschienen am 13.11.2024. Verfasst von Xiaobin Wang und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Investigation, People's Public Security University of China, Beijing, China. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 26 Literaturangaben. Gefördert von: The Opening Project of Intelligent Policing Key Laboratory of Sichuan Province, No.ZNJW2-23KFMS007.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ink analysis played an important role in document examination, but the limited dataset made it difficult for many algorithms to distinguish inks accurately. This article aimed to evaluate the feasibility of two data augmentation (DA) methods, Gaussian noise data augmentation (GNDA) and extended multiplicative signal augmentation (EMSA), for the classification of felt‐tip pen ink brands. Four brands of felt‐tip pens were analyzed using FT‐IR spectroscopy. Five classification models were used, convolutional neural network (CNN), K‐nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and partial least squares discriminant analysis (PLS‐DA). The results showed that the datasets generated by GNDA and EMSA are similar to the original datasets and have some diversity. The EMSA method had optimal classification results when combined with CNN, with classification accuracy (ACC), precision (PRE), recall (REC) and F1 score reaching 99.86%, 99.87%, 99.86%, 99.86%, and 99.86%, compared with GNDA‐CNN method (ACC = 80.90%, PRE = 87.34%, REC = 81.62%, F1 score = 79.23%). This study shows that when raw spectral data is small, DA methods can be combined with neural network models to identify ink brands effectively.

Schlagwörter

Stichwörter der Autor:innen

classificationconvolutional neural networkdata augmentationfelt‐tip penink

Autor:innen und Institutionen

  • School of Investigation, People's Public Security University of China, Beijing, China.
  • Intelligent Policing Key Laboratory of Sichuan Province, Luzhou, China.
  • Department of Transportation Management, Sichuan Police College, Luzhou, China.

Förderung

  • The Opening Project of Intelligent Policing Key Laboratory of Sichuan Province, No.ZNJW2-23KFMS007

Zitieren

APA

Wang, X., Yang, L., Chen, R., Guo, W., Han, X., & Zhang, A. (2025). Accurate felt‐tip pen brands classification based on a convolutional neural network using data augmentation. Journal of Forensic Sciences, 70(1), 170-178. https://doi.org/10.1111/1556-4029.15658

Vancouver

Wang X, Yang L, Chen R, Guo W, Han X, Zhang A. Accurate felt‐tip pen brands classification based on a convolutional neural network using data augmentation. Journal of Forensic Sciences. 2025;70(1):170-178. doi:10.1111/1556-4029.15658

BibTeX
@article{Wang2025,
  author = {Wang, Xiaobin and Yang, Lei and Chen, Ruili and Guo, Wei and Han, Xun and Zhang, Aolin},
  title = {Accurate felt‐tip pen brands classification based on a convolutional neural network using data augmentation},
  journal = {Journal of Forensic Sciences},
  year = {2025},
  volume = {70},
  number = {1},
  pages = {170--178},
  doi = {10.1111/1556-4029.15658}
}
Kostenlos starten →