In this project, I have implemented a Python Flask based server which recieves requests from the Node MCU containing sensor values.
I have trained a ML classifier containing multi-layered neural networks on the data collected while the person is walking, running and stationary.
- Features: Inertial signals obtained from the accelerometer and gyroscope of Node MCU
- Classes: Walking, Running, Stationary
The classifier predicts the class using the sensor values recieved from the Node MCU. The server correspondingly updates the time spent in the corresponding activity. The server then returns a JSON containing key-value pairs of the classes and their corresponding time-spent value.