APSIPA Transactions on Signal and Information Processing > Vol 13 > Issue 1

Continual Learning Based Personalized Abnormal Behavior Recognition Alarm System

Hyewon Song, Yonsei University, South Korea, Jiwoo Kang, Sookmyung Women's University, South Korea, jwkang@sookmyung.ac.kr , Taewan Kim, Dongduk Women's University, South Korea, kimtwan21@dongduk.ac.kr
 
Suggested Citation
Hyewon Song, Jiwoo Kang and Taewan Kim (2024), "Continual Learning Based Personalized Abnormal Behavior Recognition Alarm System", APSIPA Transactions on Signal and Information Processing: Vol. 13: No. 1, e23. http://dx.doi.org/10.1561/116.20240033

Publication Date: 25 Sep 2024
© 2024 H. Song, J. Kang and T. Kim
 
Subjects
Detection and estimation,  Pattern recognition and learning,  Activity and gesture recognition,  Motion estimation and registration,  Video analysis and event recognition,  Image and video processing,  Classification and prediction,  Deep learning,  Artificial intelligence methods in security and privacy
 
Keywords
Abnormal behavior recognitioncontinual learningalarm systemDeepStreammental illness
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 105 times

In this article:
Introduction 
Related Works 
Abnormal Behaviors 
Abnormal Behavior Recognition Alarm System 
Database 
Experiments 
Conclusion 
References 

Abstract

The number of people with mental illness is increasing because of stress or environmental influences. They require close monitoring because of their unpredictable behaviors; however, this is challenging given the lack of adequate numbers of medical staff. To overcome this problem, we propose a personalized abnormal behavior recognition alarm system for closed wards. The proposed system utilizes real-time video analysis to detect and track the locations of the patients, enabling recognition of their abnormal behaviors. In addition, new definitions are provided for specific abnormal behaviors that commonly occur in closed wards, with the adaptation of continual learning in the system. This architecture allows the creation of an abnormal behavior dataset while enhancing the recognition accuracy. The average abnormal behavior recognition accuracy with this system is over 92%. According to test results in real hospitals, about 84% of the medical staff were satisfied with the proposed system. Through the proposed alarm system, the staff could implement immediate actions without careful monitoring. By reducing the probability of occurrence of dangerous incidents, the system not only benefits the health of the patients but also enhances the working environment of the medical staff.

DOI:10.1561/116.20240033