以Kinect為基礎的網球揮拍動作分析與應用

Kinect-Based Tennis Swing Motion Analysis And Its Application

林駿維(Chun-Wei Lin)

指導教授 ： 謝景棠

淡江大學/工學院/電機工程學系碩士在職專班/碩士(2014年)

https://doi.org/10.6846/TKU.2014.00128

摘要

在本論文中利用微軟推出的Kinect，提出一套以Kinect為基礎的網球揮拍動作分
析與應用。使用者可免於一般運動擷取技術中需要將裝許多感測器在身。我們透過
OpenNI擷取的骨架座標，建立一套經由三維直角座標的座標系，並使用身體部分的
關節點對身體軀幹做投影，視為揮拍動作辨識之特徵值。
針對網球揮拍動作之軌跡，並以此特徵值透過動態時間校正(Dynamic Time
Warping)，可以算出時間序列的最佳化，以軌跡路徑做相似度的比較分辨各種揮拍動
作，並透過角度判斷式來評估揮拍動作。系統會針對其動作的揮拍高度、身體重心的
高低及重心轉移快慢進行評估，以達到辨識網球揮拍及評估揮拍動作之功能。

關鍵字

網球揮拍動作分析與應用 ； 動態時間校正

並列摘要

Use Microsoft's Kinect in this paper, we release a set of detection and identification of Kinect with tennis racquet and swing movement of the evaluation study. Users may be exempted from the general motion capture technology requires a lot of sensors will be wearing on body. via OpenNI retrieved skeleton coordinates, establishment of a three-dimensional Cartesian coordinate system, use the joints of the body and parts of the body torso to make projections, considered the swing motion identification eigenvalues.
For tennis swing movement of the track, thus eigenvalues by the Dynamic Time Warping, it can calculate the optimal sequence of time, in order to trace the path similarity comparisons do distinguish between the various swing action, and through angles judgment to evaluate the swing-style action. The system will focus on the swing-heights for their actions, the body center of gravity height and weight transfer speed is evaluated to achieve recognition tennis swing as well as assessment functions of swing action.

並列關鍵字

Motion Analysis And Its Application ； Dynamic Time Warping

參考文獻

[8]張量鈞，基於立體視覺之手掌位移辨識系統，淡江大學電機工程學系碩士論文，民國九十九年。

[3] 楊勝雄，3D 人體動作之分析與檢索，淡江大學資訊工程學系碩士班碩士論文，民國九十五年。

[5].Takeshi Yabe and Katsumi Tanaka. “Similarity Retrieval of Human Motion as Multi Stream Time Series Data” , Database Applications in Non-Traditional Environments, 1999

[6].Shih-Pin Chao, Chih-Yi Chiu, Jui-Hsiang Chao, Shi-Nine Yang and Tsang-Ken Lin. “Motion Retrieval and Its Application to Motion Synthesis”,International Conference on Distributed Computing Systems Workshops 2004

[14] Elena Kostadinova, Veselka Boeva, Liliana Boneva., “An Integrative DTW-based Imputation Method for Gene Expression Time Series Data”