心電圖信號不僅是一個非常有用的臨床診斷工具,而且也是一種很好的用於身份認證的生物特徵識別工具。它具有普遍性,唯一性和穩定性。最近,有人開始研究用心電圖信號作為身份識別的手段,並取得了一些成果。有鑑於大部分的文獻僅使用醫用心電圖儀進行研究,雜訊控制和取樣狀態都有嚴苛的規範,和實際使用於門禁管制的情況有天壤之別,本研究的主要目標就是要利用市售的簡易心電圖儀(目前選取『達楷生醫科技股份有限公司』研發販售的InstantCheck 隨身
型心電圖記錄儀)模擬實際門禁管制的情況,進行心電圖訊號的擷取、特徵判定和身份鑑定,以提
供作為日後實際門禁管制使用時的參考。
A feasibility study on the potential of the electrocardiogram (ECG) for secure access system is presented.
A test set of 20 ECG recordings from healthy subjects was obtained emulating a realistic scenario for ECG biometrics by using short measurements of 30s length in combination with a handheld ECG Recorder. The long-term stability of the individual ECG was investigated during time periods up to 30 days. Verification and identification was done by utilizing the artificial neural network with some simple signal pre-processing steps. The experimental results showed that the identification rate of 20 subjects reached 100% with the Radial Basis Function Neural Network. From the result we can see that using the electrocardiogram in the secure access system is promising.