近年來拜醫療科技的進步，台灣社會面臨人口結構高齡化的問題，而 隨這年輕子女的外移，獨居老人照護問題也比以往更加需要被關注，如何 有效且即時的對獨居老人的活動量進行量測是目前很重要的一個議題。本 論文以深度類神經網路為基礎開發了一套居家生活動作辨識系統，提出兩 套特徵截取方式，分別為以數值方式以及圖像方式將彩色攝影機所截取的 骨架資料萃取特徵，再以類神經網路抓分析動作，並將動作留存下來，以 備使用者隨時查看，系統內更有一種情境的跌倒偵測，防止意外的發生， 而在獨居老人的使用情境更可提供給長期照顧機構做為獨居老人活動力的 參考。
本論文設計的十種居家生活動作，包括一種基本的跌倒動作，用來識 別跌倒的情況，並以此蒐集資料集用以訓練及測試類神經網路，並在角度 推廣性以及不同人間之推廣性都有九成以上的推廣能力，在實際測試系統 時有 92.93%的準確率，能證明本系統在居家動作辨識上有很好的可靠性供 使用者參考。

In recent years, because of the improvement of medical technology, Taiwan is facing the severe problem of population aging. Since young people move out for work or marriage, the health care of independent-living Elderly is more important than ever. How to measure the activities of daily living for the elderly in an effective way is the crucial issue nowadays. In this paper, we developed a DNN-based System for the Recognition of the Activities of Daily Living. The system estimates skeleton data from the color image, which is recorded from webcam or surveillance system, and using the neural network like CNN, BPN or DNN to classify these features proposed by this paper. After recognized motions, we log the data in order to give the user a daily report.
In this paper, we design ten different activities of daily living including one Scene of falling movement, and testing these data with angular tolerance and person independent experiments. In these experiments, we obtained a great result of over 90% recognition rate. Even in the real-life test, this system precision rate can also achieve 92.93%. With these experiments, we can prove that the system is good enough to provide a robust report to the user for consulting.

[1] 106 年第 10 週內政統計通報. [Online]. Available: http://www.moi. gov.tw/stat/news_content.aspx?sn=11735. [Accessed: 4-Jul-2017].
[2] Kinect Interactive Games. [Online]. Avaliable: http://x-tech.am/kinect- interactive-games-game-development-company/. [Accessed: 4-Jul-
2017].
[3] B. D. Lukas, and T. Kanade, “An iterative image registration technique
with an application to stereo vision,” 7th International Joint Conference on Artificial Intelligence, pp. 674-679, 1981.
[4] I. Laptev, and T. Lindeberg, “Local Descriptors for Spatio-Temporal Recognition,” Spatial Coherence for Visual Motion Analysis, Vol. 3667,
pp. 91-103, 2004.
[5] A. F. Bobick, and J. W. Davis, “The recognition of human movement
using temporal templates,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 23 No. 3 pp. 257-267, 2001.
[6] S. Danafar, and N. Gheissari, “Action recognition for surveillance applications using optic flow and SVM,” Proceedings of the Asian
Conference on Computer Vision, Vol. 4844, pp. 457-466, 2007.
[7] I. Laptev, and T Lindeberg, “Space–time interest points,” Proceedings of the International Conference on Computer Vision, vol. 1 pp. 432-
439, 2003.
68
[8] C. Harris, and M.J. Stephens. “A combined corner and edge detector,” Alvey Vision Conference, pp. 147–152, 1988.
[9] D. G. Lowe, “Distinctive image features from scale-invariant keypoints”, International Journal of Computer Vision, Vol. 60 No. 2 pp.
91-110, 2004.
[10] H. Bay, A. Ess, T. Tuytelaars, and L. J. V. Gool, “SURF: Speeded up
robust features,” Computer Vision and Image Understanding, Vol. 110 No. 3 pp. 346-359, 2008.
[11] Wikipedia Kinect. [Online]. Available: https://en.wikipedia.org/wiki/
Kinect. [Accessed: 22-May-2017].
[12] J. Shottom, A. Fitzgibbon, M. Cook, T. Sharp, M. Finocchio, R.Moore, A. Kipman and A. Blake “Real-Time Human Pose Recognition in Parts from Single Depth Images,” IEEE Computer Vision and Pattern
Recognition, pp. 1297-1304, 2011.
[13] H. T. Kam, “Random Decision Forests,” International Conference on
Document Analysis and Recognition, Vol. 1, pp. 278–282, 1995.
[14] D. Comaniciu, and P. Meer, “Mean Shift: A Robust Approach Toward Feature Space Analysis,” Pattern Analysis and Machine Intelligence,
Vol. 24, pp. 603-619, 2002.
69
[15] Houston Ballet II dancers Liana Carpio & Chunwai Chan perform Ben Stevenson's Sylvia. Photo: Amitava Sarkar, [Online]. Available: https://houstonballet.wordpress.com/2011/08/23/18th-annual-theater- district-open-house/. [Accessed: 22-May-2017].
[16] Z. Cao, T. Simon, S. E. Wei, and Y. Sheikh, “Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields,” Conference on Computer
Vision and Pattern Recognition, 2017
[17] 維基百科:感知器. [Online]. Available: https://zh.wikipedia.org/wiki/
感知器. [Accessed: 22-May-2017].
[18] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” Nature, Vol. 323, pp. 533-
536, 1986.
[19] 蘇木春、張孝德，機械學習:類神經網路、模糊系統以及基因演算
法則，全華圖書股份有限公司，民國一百零一年。
[20] V. Nair, and G. Hinton, “Rectified linear units improve restricted Boltzmann machines,” 27th International Conference on Machine
Learning, pp. 807, 2010.
[21] S. Hochreiter , Y. Bengio , P. Frasconi, and J. Schmidhuber, “Gradient
flow in recurrent nets: the difficulty of learning long-term dependencies,” International Conference on Machine Learning, pp.
807-814, 2001.
70
[22] Using large-scale brain simulations for machine learning and A.I. [Online]. Available: https://googleblog.blogspot.tw/2012/06/using- large-scale-brain-simulations-for.html. [Accessed: 22-May-2017].
[23] Get off the deep learning bandwagon and get some perspective. [Online]. Available: http://www.pyimagesearch.com/2014/06/09/get- deep-learning-bandwagon-get-perspective/. [Accessed: 22-May-2017].
[24] M. D. Zeiler, and R. Fergus, “Visualizing and Understanding Convolutional Networks,” 13th European Conference on Computer
Vision, Vol. 8689, pp. 818-833, 2014.
[25] Deep learning for complete beginners: convolutional neural networks
with keras. [Online]. Available: https://cambridgespark.com/content/ tutorials/convolutional-neural-networks-with-keras/index.html. [Accessed: 22-May-2017].
[26] Wiki- Convolutional neural network. [Online]. Available: https://en. wikipedia.org/wiki/Convolutional_neural_network. [Accessed: 22-
May-2017].
[27] TensorFlow. [Online]. Avaliable: https://www.tensorflow.org.
[Accessed: 4-Jul-2017].
[28] An in-depth look at Google’s first Tensor Processing Unit (TPU). [Online]. Avaliable: https://cloud.google.com/blog/big-data/2017/05/
an-in-depth-look-at-googles-first-tensor-processing-unit-tpu.[Accessed: 4-Jul-2017].
71
[29] Caffe. [Online]. Avaliable: http://caffe.berkeleyvision.org. [Accessed: 4-Jul-17].
[30] Wiki atan2. [Online]. Available: https://en.wikipedia.org/wiki/Atan2. [Accessed: 4-Jul-2017].
[31] Panasonic Shop. [Online]. Available: http://shop.panasonic.com/
support-only/DMC-LX3K.html-q=lx3&start=1.[Accessed: 4-Jul-2017].