目的：感測科技能精準且即時地呈現動作資訊，穿戴感測裝置更是受到研究與商業應用的關注，逐漸取代了影像與動力學的方法；感測技術應用在游泳領域中，不僅能優化技術提升運動表現，更可提供即時回饋資訊給教練，有助於提升訓練效果；本研究提出一個以配戴慣性感測器即可準確地計算划手次數及動作分類的演算法，期以能應用於游泳教學與訓練中。方法：以12名大專游泳選手為研究對象，分別配戴慣性感測器及G 牌游泳運動手錶進行四式游泳的實驗，演算感測裝置所記錄的資訊，以相對誤差百分比評定兩種感測裝置在游泳划手次數及動作分類之準確率。結果：本研究所提之感測演算法在游泳四式的划手次數準確率達93.64%，在動作判定的準確率為95.8%；而使用游泳運動手錶在划手次數及動作判定的準確率分別為90.24%及85.4%，由結果顯示本研究所提出之方法優於游泳運動手錶之準確率。結論：本研究所設計的方法實作於Android 平台上，所提之感測波形演算法能有效計算划手次數及動作分類，未來可應用於其他週期性的運動項目上。

Purpose: Sensing technology can accurately and instantly recording motion information, Body worn inertial somsers have received much attention recently from both research and commercial communities as an alternative to video-based or kinetics approaches. In the swimming analysis field, this technology may improved stroke mechanics, race performance and real-time feedback to the coach, For help to enhance training effect. This research proposes a scheme, which can effectively and stroke counts by inertial sensors. And we hope it can be applied to swim teaching and training. Methods: 12 university-level sportsmen participate in our experiments. They carry inertial sensors or Garmin Swim watches to swim using four kinds of stroke styles. We used sensor data to compute, and use the percentage relative error to compare accuracy in stroke count and identify stroke styles. Results: The proposed scheme can achieve 93.64% and 95.8% accuracies on counting stroke counts and identifying stroke styles, respectively. On the other hand, the accuracies of Garmin Swim are 90.24% and 85.4%, respectively. From the results, our scheme can outperform the Garmin Swim watch. Conclusion: We have implemented the design scheme on Android platform. The results indicate that the designed scheme can effectively counts strokes and identifies stroke styles. In the future, we are planning to apply the design scheme on those sport types, which have regular posture movements.