本文提出一套非接觸式心脈量測技術，藉由多組視訊感測裝置拍攝皮膚以分析出心脈率。本技術首先擷取膚色區域之色彩特徵訊號，經由多裝置訊號之資料同步後進行獨立分量分析法，最終挑選具代表性之獨立分量以估測出心脈率。本方法以血氧濃度計作為對照組，於可控環境下之平均誤差值達2%以下，並具下列優點：（1）僅需一般視訊感測裝置，成本低且取得容易；（2）非接觸式量測生理訊號，減低受測者不適感；（3）抽取多組視訊感測裝置之穩定訊號，具低誤差率及高穩定性。

We propose a contact-free heart rate (HR) measurement method by analyzing sequential images of multiple video sensor devices (VSDs). In this method, skin-like pixels are firstly detected from multiple VSDs for extracting the color features. These color features in a period of time are further synchronized and analyzed by independent component analysis (ICA). A representative component is finally selected to measure the HR, which achieves under 2% deviation on average compared with a pulse oximeter in the controllable environment. The advantages of the proposed method include: (1) it uses low cost and high accessibility VSD; (2) it eases users' uncomfortableness by utilizing contact-free measurement; (3) it achieves the low error rate and the high stability by integrating multiple VSDs.