以深度資訊為主的手部形狀與運動軌跡方法及其手勢比對應用

The Application of Gestures Recognition Based on Depth Information Using Shape Classification and Trajectory Matching

張厥煒(Chueh-Wei Chung) ； 邱莉雅(Li-Ya Chiu)

《臺北科技大學學報》 46卷1期 (2014/06) Pp. 15-26

摘要

手為人體中最靈巧的肢體部位，根據個人習慣和生活背景的不同，每個人都有其獨特的手勢和動作方式，這些手勢動作會根據時間和空間的不同形成唯一的獨特性。本論文基於此想法，將多組手勢動作透過排列組合和後續的解鎖辨識，應用在門禁系統上。而主要方法是利用手部座標和手部區域的深度影像資料，經由大量的手型特徵訓練和手勢座標運動軌跡的記錄，建立以多個手勢連續動作為主的手勢動作單元，如握拳、揮手等等，最後透過動作單元的順序排列，組合成一組密碼序列，而此組密碼的比對手口辨識將為門禁系統的解鎖條件。

關鍵字

深度攝影機 ； 手勢辨識 ； 門禁系統

並列摘要

According to different personal habits and lifestyles, everyone has unique hand gestures and movements which form the uniqueness on the basis of time and space differences. Based on this concept, the thesis uses hand coordinates and depth data of hand area, and by further employing a lot of hand shape feature training and records of hand gesture coordinates' trajectory, to create hand gesture units based on multiple continuous movements, such as making a fist and waving; subsequently, by arranging the order of the movement units, it forms a set of password sequence that allows users to successfully unlock by doing the hand gestures according to the password sequence.

並列關鍵字

Depth sensor ； hand gesture recognition ； Access Control system