论文部分内容阅读
This paper introduces a novel approach for human motion recognition via motion feature vectors collected by A Micro Inertial Measurement Unit (μIMU).First,μIMU that is 56x23x15mm3 in size is built.The unit consists of three dimensional MEMS accelerometers,gyroscopes,a Bluetooth module and a Micro Controller Unit (MCU).It records human motion information,and,through a serial port to computer,the data can be saved for later on process.Second,a human motion database is setup by recording the motion data from the μIMU,which is worn by an experimentalist on left hip.The motions include fall,walk,stand,run and step upstairs,each motion has 100 samples.Third,Support Vector Machine (SVM) training process is used for human motion multi-classification.FFT is used for feature generation and optimal parameter searching process is done for the best SVM kernel function.Experimental results show that for the given 5 different motions,the total correct recognition rate is 92%,of which the fall motion can be classified from others with a 100% recognition rate.