日常生活健康對於人類來說是非常重要的,由於科技的進步,智慧型手機、平板、智慧手錶等皆搭載微機電感測元件,透過感測元件來記錄身體健康狀況是最快最具經濟效益的方法,而本研究利用智慧型手機與手錶之感測元件來做為定位與人體活動動作辨別。過去大多數的系統因為穿戴式儀器較為龐大不易安置在身上,也不易穿載外出,所以很少實際使用於日常生活當中,為了解決此問題,本研究使用操作簡易且攜帶方便的智慧型手錶作為行為辨別之檢測工具。 在即時移動定位方面,本研究利用手機藍芽裝置獲得iBeacon之RSSI,求出手機與iBeacon之間的距離,透過三角定位與接近定位演算法求出目的座標,並將座標與RSSI儲存於地圖資料庫(Map-Box)中。當使用者欲即時定位,系統會從Map-Box中透過貝氏定理選取出最佳的座標,以達到快速且準確的定位。 在行為辨識方面,本研究利用智慧手機透過無線感測網路(Wireless Sensor Network, WSN)收集智慧手錶中加速度計、轉子加速度計和陀螺儀資料數據,並制定人類行為識別和分類方法。透過本研究所提出之方法步驟分類人體動作行為狀態以增加行為辨識準確率。 本研究透過行為分類和Map-Box即時動態定位技術達到適時掌握使用者位置,並且透過位置篩選動作提高行為分類的準確率。當我們能獲得在室內空間中使用者的位置與行為資訊,此時便可確認使用者的健康與安全。
The health of the human is very import in daily life. Due to the advance of the technology, the smart phone, the tablet and the smart watch have Micro Electrical-Mechanical Systems (MEMS). The healthy status is recorded by sensor that it is the fastest and the most efficient method. In this research, we locate our position and identify the human activity action. In the past, the systems are not easy to put on the body and bring it to outside because the wearable devices are too large. The systems are seldom used in daily life. In order to solve this problem, this search uses the smart watch to identify the behavior. In terms of real-time mobile location, first, we can obtain the iBeacon’s RSSI through the blue-tooth wireless systems, and derive the distance between the smart phone and iBeacon using the RSSI. Finally, we also find the destination coordinate of the by smart phone using triangulation and proximity position methods. Furthermore, we save these coordinates and RSSI information into the Map-Box database. Our proposed system can choose the best coordinate through Bayes’ Theorem from Map-Box database. Therefore the users can obtain accurate position information rapidly. Besides, in terms of behavior identification, the sensor data is collected from the smart watch. We proposed the human behavior identification and the classify method. We can classify the human behavior status and increase the accuracy of behavior identification through our proposed method. Finally, our proposed method can obtain users’ location information accurately and increase the human behavior identification accurately. Furthermore, we can ensure the user is healthy and safe for a home environment.