Human ID of Freestyle Walking Based on Smartphone and Dual-Tree Complex Wavelet Transform

Mobile devices are becoming increasingly sophisticated and the latest generation of smartphone now incorporates many diverse and powerful sensors. These sensors include GPS sensors, vision sensors, audio sensors, vector sensors, direction sensors, rotary vector sensors, acceleration sensors and so on. The availability of these sensors in mass-marketed communication devices creates exciting new opportunities for data mining and data mining applications. In this paper we describe and evaluate a system that to use smartphone based on accelerometers and rotary vector sensors to perform gait recognition, a task which involves that identifying the human ID is performing. To implement our system, we collect labeled accelerometers and rotary vector meters data from twelve users as they perform freestyle walking, which is the most frequent activity in our daily life. We use Dual-tree Complex Wavelet Transform (DT-CWT) as feature extraction tool. Then we use general machine learning algorithm tool WEKA for classification. we use the resulting training data to induce a predictive model for gait recognition. This work is significant because the activity recognition model permits us to passively gain useful knowledge about the habits of millions of users-just by making them to carry mobile phones in their pockets. Our work has a wide range of applications, including that protecting user's private information in mobile devices is extremely important in these days.