4 Conclusions
In this paper, we study fall detection algorithm based on machine learning. We extract
the combined acceleration, the angular velocity of XOY plane, and the pitch angle as
the characteristics of the falling process, in which pitch angle is introduced to distinguish the two sets of falls. In the preprocessing of data, we exploit the “full-field
positioning” to obtain the pitch angle. Then, we proposed “sliding window method” to
update the data. To solve the difficulty in locating the occurrence of falling, we proposed the “maximum satisfaction rate” method, which affects the marking of falling
process in turn. Then, the XGBoost is used for threshold accurately training. Finally,
the falling detection model we designed has received high accuracy with perfect balance between the sensitivity and specificity.
Acknowledgements. This work is supported by the 2016 Computer and Science Institute,
Beijing University of Posts and Telecommunications.
This work is also supported by National Key R&D Program of China under grant number
SQ2018YFC200148-03.
References
1. “Falls”. World Health Organization. 16 January 2018. https://www.who.int/news-room/factsheets/detail/falls
2. The World Bank (2017) Population ages 65 and above (% of total). https://data.worldbank.
org/indicator/SP.POP.65UP.TO.ZS
3. Shieh W-Y, Huang J-C (2009) Speedup the multi-camera video-surveillance system for elder
falling detection. In: 2009 International conference on embedded software and systems.
IEEE
4. Chen J et al (2006) Wearable sensors for reliable fall detection. In: 2005 IEEE Engineering
in medicine and biology 27th annual conference. IEEE
5. Petelenz TJ, Peterson SC, Jacobsen SC (2002) Elderly fall monitoring method and device. U.
S. Patent No. 6,433,690. 13 Aug 2002
6. Yuan X et al (2015) Fall detection analysis with wearable MEMS-based sensors. In: 2015
16th International conference on electronic packaging technology (ICEPT). IEEE
7. Gumus M, Kiran MS (2017) Crude oil price forecasting using XGBoost. In: 2017
International conference on computer science and engineering (UBMK). IEEE
8. Chen Z et al (2018) XGBoost classifier for DDoS attack detection and analysis in SDNbased cloud. In: 2018 IEEE international conference on big data and smart computing
(BigComp). IEEE
9. Diebel J (2006) Representing attitude: euler angles, unit quaternions, and rotation vectors.
Matrix 58(15-16):1–35
10. Shoemake K (1985) Animating rotation with quaternion curves. In: ACM SIGGRAPH
computer graphics, vol 19, no 3. ACM
11. Noury N et al (2007) Fall detection-principles and methods. In: 2007 29th Annual
international conference of the IEEE engineering in medicine and biology society. IEEE
56
M. Xiao et al.
In this paper, we study fall detection algorithm based on machine learning. We extract
the combined acceleration, the angular velocity of XOY plane, and the pitch angle as
the characteristics of the falling process, in which pitch angle is introduced to distinguish the two sets of falls. In the preprocessing of data, we exploit the “full-field
positioning” to obtain the pitch angle. Then, we proposed “sliding window method” to
update the data. To solve the difficulty in locating the occurrence of falling, we proposed the “maximum satisfaction rate” method, which affects the marking of falling
process in turn. Then, the XGBoost is used for threshold accurately training. Finally,
the falling detection model we designed has received high accuracy with perfect balance between the sensitivity and specificity.
Acknowledgements. This work is supported by the 2016 Computer and Science Institute,
Beijing University of Posts and Telecommunications.
This work is also supported by National Key R&D Program of China under grant number
SQ2018YFC200148-03.
References
1. “Falls”. World Health Organization. 16 January 2018. https://www.who.int/news-room/factsheets/detail/falls
2. The World Bank (2017) Population ages 65 and above (% of total). https://data.worldbank.
org/indicator/SP.POP.65UP.TO.ZS
3. Shieh W-Y, Huang J-C (2009) Speedup the multi-camera video-surveillance system for elder
falling detection. In: 2009 International conference on embedded software and systems.
IEEE
4. Chen J et al (2006) Wearable sensors for reliable fall detection. In: 2005 IEEE Engineering
in medicine and biology 27th annual conference. IEEE
5. Petelenz TJ, Peterson SC, Jacobsen SC (2002) Elderly fall monitoring method and device. U.
S. Patent No. 6,433,690. 13 Aug 2002
6. Yuan X et al (2015) Fall detection analysis with wearable MEMS-based sensors. In: 2015
16th International conference on electronic packaging technology (ICEPT). IEEE
7. Gumus M, Kiran MS (2017) Crude oil price forecasting using XGBoost. In: 2017
International conference on computer science and engineering (UBMK). IEEE
8. Chen Z et al (2018) XGBoost classifier for DDoS attack detection and analysis in SDNbased cloud. In: 2018 IEEE international conference on big data and smart computing
(BigComp). IEEE
9. Diebel J (2006) Representing attitude: euler angles, unit quaternions, and rotation vectors.
Matrix 58(15-16):1–35
10. Shoemake K (1985) Animating rotation with quaternion curves. In: ACM SIGGRAPH
computer graphics, vol 19, no 3. ACM
11. Noury N et al (2007) Fall detection-principles and methods. In: 2007 29th Annual
international conference of the IEEE engineering in medicine and biology society. IEEE
56
M. Xiao et al.