Personalized Unusual Event Detection Algorithm at Smart Home via Daily Activity and Vision Pattern
AUTHORS
Ergashev Odiljon,Software department, Korea National University of Transportation
Hwijune Park,Software department, Korea National University of Transportation
Junho Ahn,Junho Ahn, Address: Software, Korea National University of Transportation. Daehak-ro, Daesowon-myeon, Chungju, Korea
ABSTRACT
There currently are used smartphone applications and video camera systems for emergency detections in home environments. The mobile applications can identify the user behaviors to classify accident situations in indoor areas while carrying the phones on the body or the vision systems detect the video-based accident motions in a limited range of a camera installed at home. To compensate for these limitations, we propose a fusion algorithm to detect personalized unusual events like serious injuries and even near death at home via daily activity and vision pattern. We designed and implemented the fusion classification algorithm based on user activity detection with the smartphone accelerometer, and the behavior identification with a video camera installed at home. We evaluated both the activity and vision pattern algorithms and simulated the fusion algorithm with high-accuracy performance in scenarios.
KEYWORDS
Unusual event, Activity, Vision, Emergency, Fusion, Home
REFERENCES
[1] Alexander Trowbridge, Living alone? You're not the only one, CBS news, Aug 29, (2013),
[2] https://www.cbsnews.com/news/living-alone-youre-not-the-only-one/
[3] Monisha Mohan, Arun P.S, ACCELEROMETER–BASED HUMAN FALL DETECTION AND RESPONSE USING SMARTPHONES, International Journal of Computer Engineering In Research Trends,5, (2017).
[4] Zishan Zahidul Islam, Syed Mahir Tazwar, Md. Zahidul Islam, Seiichi Serikawa and Md. Atiqur Rahman Ahad, Automatic Fall Detection System of Unsupervised Elderly People Using Smartphone, IIAE International Conference on Intelligent Systems and Image Processing, 7, (2017)
[5] Glen Debard, Marc Mertens, Toon Goedeme, Tinne Tuytelaars and Bart Vanrumst Three Ways to Improve the Performance of Real-Life Camera-Based Fall Detection Systems Journal of Sensors (2017)
[6] Miao Yu, Liyun Gong, Stefanos Kollias Computer vision based fall detection by a convolutional neural network ACM (2017) 11
[7] Koldo de Miguel, Alberto Brunete, Miguel Hernando and Ernesto Gambao, Home Camera-Based Fall Detection System for the Elderly, Multidisciplinary Digital Publishing Institute (MDPI),Sensors,21, (2017)
[8] Fouzi Harroua , Nabil Zerroukib , Ying Suna , Amrane Houacineb,Vision-based fall detection system for improving safety of elderly people,IEEE Instrumentation and Measurement Society, 21, (2017)
[9] Philip Geismann. Georg Schneider, A Two-staged Approach to Vision-based Pedestrian Recognition Using Haar and HOG Features,IEEE,6 (2008)
[10] Junho Ahn, Richard Han, myBlackBox: Blackbox Mobile Cloud Systems for Personalized Unusual Event Detection, Sensors, (2016), Volume 16, Issue 5, 753, 20 pages.(CrossRef)(Google Scholar)
[11] Junho Ahn, James Williamson, Mike Gartrell, Richard Han, Qin Lv, and Shivakant Mishra. (2015).
[12] Supporting Healthy Grocery Shopping via Mobile Augmented Reality. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), vol. 12, 1s, Article 16 (October 2015), 24 pages.
[13] Huang J, Rathod V, Sun C, Zhu M, Korattikara A, Fathi A, Fischer I, Wojna Z, Song Y, Guadarrama S, Murphy K., Speed/accuracy trade-offs for modern convolutional object detectors. CVPR (2017), https://github.com/tensorflow/models/tree/master/research/object_detection