Suspicious behavior recognition using deep learning

Focusing on keypoint 2D scaling

Authors

  • Yeonji Park Kwangwoon University
  • Yoojin Jeong Kwangwoon University
  • Chaebong Sohn Kwangwoon University

DOI:

https://doi.org/10.37944/jams.v4i1.78

Keywords:

defense and security technology, surveillance camera, suspicious person, behavior recognition, OpenPose

Abstract

The purpose of this study is to reinforce the defense and security system by recognizing the behaviors of suspicious person both inside and outside the military using deep learning. Surveillance cameras help detect criminals and people who are acting unusual. However, it is inefficient in that the administrator must monitor all the images transmitted from the camera. It incurs a large cost and is vulnerable to human error. Therefore, in this study, we propose a method to find a person who should be watched carefully only with surveillance camera images. For this purpose, the video data of doubtful behaviors were collected. In addition, after applying a algorithm that generalizes different heights and motions for each person in the input images, we trained through a model combining CNN, bidirectional LSTM, and DNN. As a result, the accuracy of the behavior recognition of suspicious behaviors was improved. Therefore, if deep learning is applied to existing surveillance cameras, it is expected that it will be possible to find the dubious person efficiently.

Author Biographies

Yeonji Park, Kwangwoon University

(Co-Author) Kwangwoon University, Department of Electronics and Communications Engineering, Master’s Course Student

e-mail: [email protected]

Yoojin Jeong, Kwangwoon University

(First Author) Kwangwoon University, Department of Electronics and Communications Engineering, Master Candidate

e-mail: [email protected]

Chaebong Sohn, Kwangwoon University

(Corresponding Author) Kwangwoon University, Department of Electronics and Communications Engineering, Professor

e-mail: [email protected]

Suspicious Behavior Recognition

Downloads

Published

2021-04-30

How to Cite

Park, Y., Jeong, Y., & Sohn, C. (2021). Suspicious behavior recognition using deep learning: Focusing on keypoint 2D scaling. Journal of Advances in Military Studies, 4(1), 43-59. https://doi.org/10.37944/jams.v4i1.78