臺灣博碩士論文加值系統

本論文重點在於建構一種以電腦視覺為基礎的老人跌倒偵測演算法，利用 IP Camera 將拍攝到的
畫面資訊做多重的演算法處理，包括三向相減法(Three-Phase Difference)、光流法(Optical Flow)、
以及高斯混合模型(Gaussian Mixture Model)。此外，該演算法經過優化後，在單一電腦上能夠支援多
隻IP Camera。除了演算法本身之外，經由UI 設計包裝，輔以中央監控系統CMS(Central Monitoring
System)，能夠針對每個老年人的環境與健康情形做控管的動作，當發生任何異常狀況時，隨即通知
醫護人員，並根據使用者需求，發出不同種類的警報訊息。本論文提出的多個演算法，成功的整合
中央監控系統以及使用者介面，是一套獨立且完整的系統。

This thesis aims to design an elderly fall-detection algorithm based on intelligent video
surveillance. We use images captured by IP cameras for computations using the Retinex,
Three-Phase Dierence, Optical Flow, and Gaussian Mixture Model. In addition, we opti-
mize the algorithm to support multi-IP cameras on a single computer. Using the designed
interface and the central monitoring system, we can monitor elderly persons' living environ-
ments and health conditions. This system is designed to notify medical sta immediately
when accidents occur, and distinct types of alarm are used according to the situation. The
algorithm presents here integrates the central monitoring system and the user interface,
which is an independent complete system.

Contents
1 Introduction 5
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3 Thesis Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Literature Survey 11
2.1 Elderly Fall Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1.1 Single-Camera System . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.1.2 Multi-Camera System . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Behavior Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 System Architecture 18
3.1 Elderly Fall Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2 Flow Chart of Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4 Algorithm 22
4.1 Retinex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.1 Single-scale Retinex . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.2 Multi-scale Retinex . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.1.3 Multi-scale Retinex color restoration . . . . . . . . . . . . . . . . . . 24
4.2 Three-Phase Dierence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.3 Optical Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.4 Gaussian Mixture Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5 Experimental Results 34
5.1 False Alarm Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
6 Conclusion and Future Work 45

[1] National policy foundation of Taiwan. http://www.npf.org.tw.
[2] T. Ji, T. Mukherjee, W. Mun-Fei, W. Mengdi, and T. Yap-Peng. Fall incidents detection
for intelligent video surveillance. In Internation Conf. on Information, Communications,
and Signal Processing, 2005.
[3] Community Development Quarterly. http://sowf.moi.gov.tw/19/cdj.htm.
[4] C. Rougier, J. Meunier, A. St-Arnaud, and J. Rousseau. Robust video surveillance for
fall detection based on human shape deformation. In IEEE Transactions on Circuits
and Systems for Video Technology, 2011.
[5] C. Rougier and J. Meunier. Demo : Fall detection using 3d head trajectory extracted
from a single camera video sequence. In Journal of Telemedicine and Telecare, 2005.
[6] H. Foroughi, A. Naseri, A. Saberi, and H.S. Yazdi. Approach for human fall detection
using integrated time motion image and neural network. In Proceedings of Internation
Conference on Signal Processing, 2008.
[7] The elderly population in United State Bureau of the Census.
http://www.census.gov/population/www/pop-prole/elderpop.html.
[8] C. Zhang, Y. Tian, and E. Capezuti. Privacy preserving automatic fall detection for
elderly using rgbd cameras. In Computers Helping People with Specail Needs Lecture
Notes in Computer Science, volume 7382, pages 625-633, 2012.
[9] C. Zhang. Rgb-d camera-based activity analysis. In Singal and Information Processing
Association Annual Summit and Conference APSIPA ASC, pages 1-6, 2012.
[10] M. Shaou-Gang, S. Pei-Hsu, and H. Chia-Yuan. A customized human fall detection
system using omni-camera images and personal information. In IEEE Proceedings of the
1st Distributed Diagnosis and Home Healthcare (D2H2) Conference Arlington, Virginia,
USA, pages 39-42, 2006.
[11] R. Cucchiara, A. Prati, and R. Vezzani. A multi-camera vision system for fall detection
and alarm generation. In Expert Systems Journal, volume 24, pages 334-345, 2007.
[12] H. Nait-Charif and S. McKenna. Activity summarization and fall detection in a sup-
portive home environment. In Proc. 17th ICPR, volume 4, pages 323-326, 2004.
[13] K. Brewer, C. Ciolek, and M. F. Delaune. Falls in community dwelling older adults:
Introduction to the problem. In APTA Continuing Education Series, pages 38-46, 2007.
[14] M. Kangas, A. Konttila, P. Londgren, I. Winblad, and T. Jasmsa. Comparion of low-
complexity fall detection algorithms for body attached accelerometers. In Gait Posture,
volume 28, pages 285{291, 2008.
[15] M. Nyan, F. E. Tay, and E. Murugasu. A wearable system for pre-impact fall detection.
In Biomech, volume 41, pages 3475-3481, 2008.
[16] iLife. Fall Detection Sensor [Online]. http://www.falldetection.com/ilifefds.asp.
[17] Directalert. Wireless Emergency Response System [Online].
http://www.directalert.ca/emergency/help-button.php.
[18] EDAO. http://www.edao.com/domicile/overview/.
[19] HONcode. http://www.hon.ch/honcode.
[20] Elder Falling Detection Project V FATE. http://www.project-fate.eu.
[21] WiMax Home care robot. http://www.vnax.net.tw.
[22] Pressure Sensitive Pad for Elder. http://www.cmuh.cmu.edu.tw/web/18910/excellence.
[23] B. Toreyin, Y. Dedeoglu, and A. Cetin. Hmm based falling person detection using both
audio and video. In Proc. IEEE Int. Workshop Hum.-Comput. Interaction, pages 1-4,
2005.
[24] D. Anderson, J. Keller, M. Skubic, X. Chen, and Z. He. Recognizing falls from silhou-
ettes. In Proc. Int. Conference IEEE EMBS, pages 6388-6391, 2006.
[25] T. Lee and A. Mihailidis. An intelligent emergency response system: preliminary de-
velopment and testing of automated fall detection. In J. Telemed. Telecare, volume 11,
pages 194-198, 2005.
[26] Homa Foroughi, Baharak Shakeri Aski, and Hamidreza Pourreza. Intelligent video
surveillance for monitoring fall detection of elderly in home environments. In Proceedings
of 11th International Conference on Computer and Information Technology, 2008.
[27] A. Sixsmith and N. BJohnson. A smart sensor to detect the falls of the elderly. In IEEE
Prevasive Computation, volume 3, pages 42-47, 2004.
[28] C. Rougier, J. Meunier, A. St-Arnaud, and J. Rousseau. Monocular 3-d head tracking
to detect falls of elderly people. In Proc. Int. Conference IEEE Eng. Med. Biol. Soc.,
pages 6384-6387, 2006.
[29] C. Rougier, J. Meunier, A. St-Arnaud, and J. Rousseau. Fall detection from hu-
man shape and motion history using video surveillance. In Proc. 21st Int. Conference
AINAW, volume 2, pages 875-880, 2007.
[30] N. Thome, S. Miguet, and S. Ambellouis. A real-time, multiview fall detection system: A
lhmm-based approach. In IEEE Transaction Circuits System Video Technol., volume 18,
pages 1522-1532, 2008.
[31] D. Anderson, R. H. Luke, Skubic M. Keller, J. M. and, M. Rantz, and M. Aud. Linguistic
summarization of video for fall detection using voxel person and fuzzy logic. In Computer
Vision Image Understanding, volume 113, pages 80-89, 2009.
[32] E. Auvinet, L. Reveret, A. St-Arnaud, J. Rousseau, and J. Meunier. Fall detection
using multiple cameras. In Proc. 30th Annu. Int. Conference IEEE Eng. Med. Biol.
Soc., pages 2554-2557, 2008.
[33] L. Hazelho, J. Han, and P. H. N. de With. Video-based fall detection in the home using
principal component analysis. In Proc. Advance Concepts Intelligent Vision System,
volume 1, pages 298-309, 2008.
[34] D. Glen, K. Peter, D. Mieke, V. Ellen, V.D.N. Jonas, D. Eddy, M. Koen, G. Toon,
T. Tinne, and V. Bart. Camera based fall detection using multiple features validated
with real life video. In Workshop Proceedings of the 7th International Conference on
Intelligent Environments, volume 10, 2011.
[35] H. Nait-Charif and S.J. McKenna. Acitivity summarisation and fall detection in a
supportive home environment. In Proc. of the 17th International Conf. on Pattern
Recognition, 2004.
[36] M. Popescu, Y. Li, M. Skublic, and M. Rantz. An acoustic fall detector system that uses
sound height information to reduce the false alarm rate. In Engineering in Medicine and
Biology Society, 30th Annual International Conference of the IEEE, pages 4628-4631,
2008.
[37] A. Williams, D. Ganesan, and A. Hanson. Aging in place: Fall detection and localization
in a distributed smart camera network. In Proc. of the 15th international conf. on
Multimedia, pages 892-901, 2007.
[38] G. Diraco, A. Leone, and P. Siciliano. An active vision system for fall detection and
posture recognition in elderly healthcare. In IEEE In Design, Automation and Test in
Europe Conf. and Exhibition, pages 1536-1541, 2010.
[39] G.L. Foresti, L. Marcenaro, and C.S. Regazzoni. Automatic detection and indexing of
video-event shots for surveillance applications. In IEEE Transactions on Multimedia,
pages 459-471, 2002.
[40] C.J. Cohen, F. Morelli, and K.A. Scott. A surveillance system for the recognition of
intent within individuals and crowds. In IEEE Conference In Technologies for Homeland
Security, pages 559-565, 2008.
[41] S. Mitra and T. Acharya. Gesture recognition: A survey. In IEEE Transaction on
Systems, Man, and Cybernetics - part C: Applications and Reviews, pages 311{324,
2007.
[42] H. Farid. Blind inverse gamma correction. In IEEE Transaction on Image Processing,
volume 10, pages 1428-1433, 2001.
[43] P. E. Trahanias. Color image enhancement through 3-d histogram equalization. In 11th
IAPR International Conference on Image, Speech and Signal Analysis, Proceedings,
pages 545-547, 1992.
[44] L. Ching-Liang, T. Luo-Wei, W. Yuan-Kai, and F. Kuo-Chin. Robust face recognition
under illumination and facial expression variations. In Proceedings of the Ninth Inter-
national Conf. on Machine Learning and Cybernetics, Qingdao., pages 11-14, 2010.
[45] Z. Rahman, D. Jobson, and G. Woodell. Retinex processing for automatic image en-
hancement. In The Human Vision and Electronic Imaging VII Conf., volume 4662,
pages 390-401, 2002.
[46] S. S. Agaian, K. Panetta, and A. M. Grigoryan. A new measure of image enhancement.
In presented at the IASTED International Conf. Signal Processing Communication,
Marbella, Spain, 2000.
[47] L. Xu, J. Jia, and Y. Matsushita. Motion detail preserving optical
ow estimation.
In IEEE Transactions on pattern analysis and machine intelligence, volume 34, pages
1744-1757, 2012.
[48] Z. Zivkovic. Improved adaptive gaussian mixture model for background subtraction. In
Proc. ICPR, 2004.