以區間第二型模糊模型判斷使用者動作之方法

根據內政部的調查報告中指出臺灣地區六十五歲以上人口佔總人口數11.74%，其中實際獨居或僅與配偶同居比例為27.9%，獨居老人的生活照顧是一項被關注的焦點，跌倒偵測更是一個廣泛的研究課題，跌倒不僅危害老年人健康，而且會留下副作用，如內心的陰影，讓他們在日常生活中害怕會再次跌倒。跌倒相關的研究大多以可攜式感測器、影像判別及環境感測器方式，以確定使用者是否有跌倒意外事故發生，但以影像判別及環境感測器方式，考慮諸多的環境不確定因素和成本考量，相對於可攜式感測器的方式不僅可有效獲得身體局部動作資訊，更可大幅降低外在環境因素導致的誤判率。因此，本研究利用智慧型手機設計一套適用於居家環境之跌倒偵測系統，透過智慧型手機內建的重力加速度感測器擷取使用者分析可能會導致意外跌倒的日常活動。本論文提出以區間第二型模糊模型來自動檢測老年人是否發生跌倒，並以多層級判別系統進一步確定老年人活動型態。多層級判別系統利用訊號向量強度和訊號強度面積的方法來區分跌倒、慢跑等激烈動作與其他各項緩和的動作，使得跌倒偵測工作可以進一步簡化，實驗結果顯示所提系統能100%正確識別跌倒，至於正常行走和慢跑活動的準確率分別為92.59%與83.52%。

關鍵字

區間第二型模糊模型；跌倒偵測；智慧型手機；重力加速度感測器

並列摘要

According to the Ministry of Interior, Taiwan survey reports 11.74% of the Taiwan’s population age over 65 years. The ratio who lives alone or with their companion goes up to 27.9%. Keeping them save from falling or dropping is one of the major concerns but to accomplish that they need full time assistance which is sometimes hard to maintain. This problem can be resolved by developing such a system which can detect every time they collapse or have a fall; that system is called fall detection. It is nowadays one of the most researched areas and this study also focuses on the same context. Portable sensors, image recognition and environmental sensors are mostly used to determine such type of incidents but it is analyzed that image recognition and environmental sensors are not favorable as they have uncertainty factors and cost considerations. Portable sensors are considered most feasible and effective because they not only maintains information about body movements but also reduces the impact of false external environmental factors. This study approaches from analyzing elderly daily activities that may cause fall accidents. Interval type-2 fuzzy models are proposed to automatically detect whether the elderly fall occurred, and multilayer detection system is devised to further identify elderly activity types. Signal Vector Magnitude (SVM) and Signal Magnitude Area (SMA) methods are used to discriminate fall activities from daily routine tasks so that detection can be further improved and made efficient. Experimental results show that the proposed system can be 100% correct in case of fall identification, as for the accuracy of regular activities such as walking and jogging it declines a bit to 92.59% and 83.52%, respectively.

並列關鍵字

interval type-2 fuzzy model ； fall detection ； smart phones ； accelerometer

參考文獻

2.財團法人門諾社會福利慈善事業基金會，「用愛關懷，溫暖到宅~~102 年度東部獨居老人照顧計」，http://www.mf.org.tw/item06/10201.pdf。

3.A. M. Khan, Y. K. Lee, S. Y. Lee and T. S. Kim, “A triaxial accelerometer-based physical-activity recognition via augmented-signal features and a hierarchical recognizer,” IEEE Trans. on Information Technology in Biomedicine, vol. 14, no. 5, pp.1166-1172, Sep. 2010.

4.F. Bianchi, S. J. Redmond, M. R. Narayanan, S. Cerutti and N. H. Lovell, “Barometric pressure and triaxial accelerometry-based falls event detection,” IEEE Trans. on Neural Systems and Rehabilitation Engineering, vol. 18, no. 6, pp.619-627, Dec. 2010.

5.D. Curone, A. Tognetti, E. L. Secco, G. Anania, N. Carbonaro, D. D. Rossi and G. Magenes, “Heart rate and accelerometer data fusion for activity assessment of rescuers during emergency interventions,” IEEE Trans. on Information Technology in Biomedicine, vol. 14, no. 3, pp.702-710, May 2010.

6.M. Sekine, T. Tamura, M. Akay, T. Fujimoto, T. Togawa and Y. Fukui, “Discrimination of walking patterns using wavelet-based fractal analysis,” IEEE Trans. on Neural System and Rehabilitation Engineering, vol. 10, no. 3, pp.188-196, Sep. 2002.

國際替代計量

以區間第二型模糊模型判斷使用者動作之方法

全文下載

主題瀏覽