A novel multisensoric system recording and analyzing human biometric features for biometric and biomedical applications

摘要: The recording and analyzing human motor control movements are fundamental parts of both behavioral biometrics biomedical research studies. dynamics functions fingers, hand wrist can be studied while handwriting, drawing or gestures. major difficulties the acquisition high quality data characterization acquired accurately efficiently. A multisensoric Biometric Smart Pen BiSP device has been developed which ability to measure handwriting gesturing during writing on paper pad free in space. Hence, biometric features recorded terms pen refill pressures, finger grip pressures holding pen, inclination, acceleration tilts three dimensions. sensing is unique provides excellent fine skills a writer. by have not only analyzed recognize person handwritten object applications but also characterize dysfunctions writer due Parkinson’s disease for diagnostics medical applications. problems processing automatic feature selection classification especially short sequence input (e.g., single characters). For above applications, conducted that involves improved procedures diverse techniques acquisition, signal pattern recognition based signals (time series) obtained from novel digital pen. Many including enhanced pre-processing dimension reduction methods as well extraction implemented. this, dynamic time warping (DTW) together with its variants support vector machine (SVM) established evaluated analysis. In biometrics, performances authentication using new designed system obtained. application, personal neuro-motoric order medication. brief, evaluate newly (sensor software methods) several experiments performed determine feasibility register analyze multiple