A smartphone-based real-time simple activity recognition

This paper explores a physical activity monitoring application in smartphones. The combination of smartphone application and sensors could perform as a biomedical sensor to monitor physical activities. Thai 3 Axis (TH3AX) was developed as a real-time physical activities monitoring application such as standing, walking, and running. The smartphone was attached to right hip of 10 young, healthy adult subjects to collect accelerometer data to determine threshold range for each activity. After finding thresholds, the results were used as range for a predictive model of activity recognition. The second experiment was scheduled with 6 young healthy subjects performed three physical activities (standing, walking, and running) to evaluate TH3AX application. Then, a diagnostic test was computed to test TH3AX's sensitivity, specificity, and accuracy. Results showed that TH3AX sensitivity, specificity, and accuracy are 0.981, 0.988, and 0.986, respectively.

[1]  Rebecca Vieyra,et al.  Analyzing Forces on Amusement Park Rides with Mobile Devices , 2014 .

[2]  Shyamal Patel,et al.  A review of wearable sensors and systems with application in rehabilitation , 2012, Journal of NeuroEngineering and Rehabilitation.

[3]  Daniel P. Siewiorek,et al.  Activity recognition and monitoring using multiple sensors on different body positions , 2006, International Workshop on Wearable and Implantable Body Sensor Networks (BSN'06).

[4]  Diane J. Cook,et al.  Simple and Complex Activity Recognition through Smart Phones , 2012, 2012 Eighth International Conference on Intelligent Environments.

[5]  A. Akobeng.  Understanding diagnostic tests 1: sensitivity, specificity and predictive values , 2007, Acta paediatrica.

[6]  Peter H Veltink,et al.  Accelerometer and rate gyroscope measurement of kinematics: an inexpensive alternative to optical motion analysis systems. , 2002, Journal of biomechanics.

[7]  Cheol-Hoon Lee,et al.  Development of a Single 3-Axis Accelerometer Sensor Based Wearable Gesture Recognition Band , 2007, UIC.

[8]  Y.-K. Lee,et al.  Human Activity Recognition via an Accelerometer-Enabled-Smartphone Using Kernel Discriminant Analysis , 2010, 2010 5th International Conference on Future Information Technology.

[9]  Michael L. Littman,et al.  Activity Recognition from Accelerometer Data , 2005, AAAI.

[10]  Davide Anguita,et al.  Human Activity Recognition on Smartphones Using a Multiclass Hardware-Friendly Support Vector Machine , 2012, IWAAL.

[11]  M. Brock,et al.  The use of “overall accuracy” to evaluate the validity of screening or diagnostic tests , 2004, Journal of General Internal Medicine.

[12]  Hanghang Tong,et al.  Activity recognition with smartphone sensors , 2014 .

[13]  Angelo M. Sabatini,et al.  Machine Learning Methods for Classifying Human Physical Activity from On-Body Accelerometers , 2010, Sensors.