Activity Detection
- Citation Author(s):
-
Vahideh Shirmohammadli (Simon Fraser University)
- Submitted by:
- Vahideh Shirmohammadli
- Last updated:
- DOI:
- 10.21227/qjnr-re50
- Research Article Link:
294 views
- Categories:
Abstract
Wearable sensors can efficiently detect human body movements with the ability to be extended to an Internet-of-Things (IoT) platform. In this platform, integration of wearable sensors, smartphones, and activity recognition takes place on a web-based application. A SensorTile kit from STMicroelectronics is an IoT module that packs powerful processing capabilities and is mostly used for activity detection. It consists of a 3-axis accelerometer, 3-axis gyroscope, 3-axis Magnetometer and Pressure, Temperature, and Humidity sensors. We have conducted the Data acquisition using this specific module once the person was running, walking, standing, sitting, and so on.
Instructions:
This is a guide to open and read the data and Matlab codes for activity detection based on accelerometers.
The Matlab Code makes use of built-in algorithms in Matlab.
Requirement: 2020b The MathWorks, Inc.
1- Unzip both 'data' and 'code' files into one folder.
2- Run 'MainCode' and read the instructions.
3- 'MainCode' calls three Matlab functions, which was created to aviod any complexity in the main code: Acceleration, FeatureExtraction, and normalization.
4- To build a specific Model, enable the determined lines in the 'Training' section.
5- For more visualization or algorithms you may add your code lines appropriately.
Enjoy!
