Page 132 - ITU KALEIDOSCOPE, ATLANTA 2019

2019 ITU Kaleidoscope Academic Conference

video surveillance. The system is designed to work within a [8] A. Núñez-Marcos, G. Azkune, and I. Arganda-
confined location such as hospitals, indoor rooms and public Carreras, "Vision-based fall detection with
places. The system has been tested on two different datasets convolutional neural networks", Wireless
of MOT and UR Fall and evaluated the performance of both Communications and Mobile computing, Volume
models. The MFPT model’s precision and accuracy denote 2017, Article ID 9474806, 2017.
the fact that multiple feature-based models help in achieving
higher efficiency. The proposed system achieved 94.67% [9] H. S. Parekh, D. G. Thakore, and U. K. Jaliya, "A
precision in tracking and 98.01% accuracy in elderly fall survey on object detection and tracking methods,
detection. The usage of LSTM model in both the models has " International Journal of Innovative Research in
aided in representing time-series data effectively. The Computer and Communication Engineering vol 2, no.
proposed system for elderly healthcare in homes and 2 pp. 2970-2979, 2014.
hospitals can be standardized in ITU-T Study Group 16,
which is the parent group of Focus Group on Artificial [10] A. Aggarwal, S. Biswas, S. Singh, S. Sural, and A. K.
Intelligence for Health (FG-AI4H). The proposed work can Majumdar, "Object tracking using background
be extended to detect different activities apart from fall subtraction and motion estimation in MPEG videos,"
detection, and recognize and report in the cases of anomalies. Asian Conference on Computer Vision, Springer, pp.
The fall detection module consisting of a HoG feature-based 121-130, 2006.
LSTM training network is the standardization item.
[11] S. Aslani,, and H. M. Nasab, "Optical flow based
REFERENCES moving object detection and tracking for traffic
surveillance," International Journal of Electrical,
[1] World Population Aging Report, United Nations Electronics, Communication, Energy Science and
2017, Engineering” vol. 7, no. 9 pp. 789-793, 2013.
https://www.un.org/en/development/desa/population
/publications/pdf/ageing/WPA2017_Highlights.pdf [12] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner,
"Gradient-based learning applied to document
[2] WHO Guidelines on Integrated Care for Older recognition," Proceedings of the IEEE, vol. 86, no.
People: 11, pp. 2278-2324, 1998.
https://www.who.int/ageing/publications/guidelines-
icope/en/ [13] R. Girshick, J. Donahue, T. Darrell, and J. Malik,
"Rich feature hierarchies for accurate object detection
[3] S. Van der Spek, J. Van Schaick, P. De Bois, and R. and semantic segmentation," IEEE conference on
De Haan, "Sensing human activity: GPS Computer Vision and Pattern Recognition, pp. 580-
tracking," Sensors vol. 9 no. 4, pp. 3033-3055, 2009. 587, 2014.

[4] S. K. Opoku, "An indoor tracking system based on [14] D. Held, S. Thrun, and S. Savarese, "Learning to track
bluetooth technology," arXiv preprint arXiv at 100 fps with deep regression networks," European
1209.3053, 2012. Conference on Computer Vision, pp. 749-765., 2016.
https://arxiv.org/ftp/arxiv/papers/1209/1209.3053.pd
f [15] X. Liu, Y. Zhou, J. Zhao, R. Yao, B. Liu and Y.
Zheng, "Siamese Convolutional Neural Networks for
[5] M. Shankar, J. B. Burchett, Q. Hao, B. D. Guenther, Remote Sensing Scene Classification," IEEE
and D. J. Brady, "Human-tracking systems using Geoscience and Remote Sensing Letters. vol. 16, no.
pyroelectric infrared detectors," Optical 8, pp. 1200-1204, Aug. 2019.
Engineering, vol 45, no. 10, 2006.
[16] ITU focus group on “Artificial Intelligence for
[6] F. Wu, H. Zhao, Y. Zhao, and H. Zhong, Health” https://www.itu.int/en/ITU-
"Development of a wearable-sensor-based fall T/focusgroups/ai4h/Pages/default.aspx
detection system," International Journal of
Telemedicine and Applications, Vol 2015 Article ID [17] N. Dalal, and B. Triggs, "Histograms of oriented
576364, 2015 gradients for human detection,” IEEE Conference on
Computer Vision & Pattern Recognition (CVPR'05),
[7] N. Zerrouki, and A. Houacine, "Combined curvelets vol. 1, pp. 886-893. IEEE Computer Society, 2005.
and hidden Markov models for human fall
detection", Multimedia Tools and Applications vol [18] S. Hochreiter, and J. Schmidhuber, "Long short-term
77 no. 5 pp. 6405-6424, 2018. memory," Neural Computation vol. 9, no. 8 pp. 1735-
1780, 1997.

– 112 –