EUVIMEDEUVIMED entdecken
Crossref

A deep learning-based recognition for dangerous objects imaged in X-ray security inspection device

Qiuyue Wei, Shenlan Ma, Shaojie Tang, Baolei Li, Jiandong Shen, Yuanfei Xu, Jiulun Fan

Journal of X-Ray Science and Technology: Clinical Applications of Diagnosis and Therapeutics · 27.01.2023 · 31 (1) · 13-26

DOI: 10.3233/xst-221210 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of X-Ray Science and Technology: Clinical Applications of Diagnosis and Therapeutics, online erschienen am 20.10.2022. Verfasst von Qiuyue Wei und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Automation, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi, China. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 22 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Several limitations in algorithms and datasets in the field of X-ray security inspection result in the low accuracy of X-ray image inspection. In the literature, there have been rare studies proposed and datasets prepared for the topic of dangerous objects segmentation. In this work, we contribute a purely manual segmentation for labeling the existing X-ray security inspection dataset namely, SIXRay, with the pixel-level semantic information of dangerous objects. We also propose a composition method for X-ray security inspection images to effectively augment the positive samples. This composition method can quickly obtain the positive sample images using affine transformation and HSV features of X-ray images. Furthermore, to improve the recognition accuracy, especially for adjacent and overlapping dangerous objects, we propose to combine the target detection algorithm (i.e., the softer-non maximum suppression, Softer-NMS) with Mask RCNN, which is named as the Softer-Mask RCNN. Compared with the original model (i.e., Mask RCNN), the Softer-Mask RCNN improves by 3.4% in accuracy (mAP), and 6.2% with adding synthetic data. The study result indicates that our proposed method in this work can effectively improve the recognition performance of dangerous objects depicting in the X-ray security inspection images.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningX-RaysRadiographyAlgorithms

Stichwörter der Autor:innen

Deep learningdata augmentationmask RCNNobject recognitionsecurity inspection

Autor:innen und Institutionen

  • School of Automation, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi, China.
  • Xi'an Key Laboratory of Advanced Control and Intelligent Process, Xi'an, Shaanxi, China.
  • Automatic Sorting Technology Research Center, Xi'an University of Posts and Telecommunications, State Post Bureau of the People's Republic of China, Xi'an, Shaanxi, China.
  • Beijing Hangxing Machinery Manufacturing Co., Ltd, Dongcheng, Beijing, China.
  • School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi, China.

Zitieren

APA

Wei, Q., Ma, S., Tang, S., Li, B., Shen, J., Xu, Y., & Fan, J. (2023). A deep learning-based recognition for dangerous objects imaged in X-ray security inspection device. Journal of X-Ray Science and Technology: Clinical Applications of Diagnosis and Therapeutics, 31(1), 13-26. https://doi.org/10.3233/xst-221210

Vancouver

Wei Q, Ma S, Tang S, Li B, Shen J, Xu Y, et al. A deep learning-based recognition for dangerous objects imaged in X-ray security inspection device. Journal of X-Ray Science and Technology: Clinical Applications of Diagnosis and Therapeutics. 2023;31(1):13-26. doi:10.3233/xst-221210

BibTeX
@article{Wei2023,
  author = {Wei, Qiuyue and Ma, Shenlan and Tang, Shaojie and Li, Baolei and Shen, Jiandong and Xu, Yuanfei and Fan, Jiulun},
  title = {A deep learning-based recognition for dangerous objects imaged in X-ray security inspection device},
  journal = {Journal of X-Ray Science and Technology: Clinical Applications of Diagnosis and Therapeutics},
  year = {2023},
  volume = {31},
  number = {1},
  pages = {13--26},
  doi = {10.3233/xst-221210}
}
Kostenlos starten →