Developing an Automatic Evaluation of Exertion Using a Smart Phone

Human activity recognition (HAR) is an important task in the field of wearable technology and the Internet of Things (IoT). This abstract presents a study on using Edge Impulse and an accelerometer sensor to recognize human activities using an auto-encoder (EON) tuner. The accelerometer sensor data is collected from a wearable device worn on the wrist. The data is then preprocessed and used to train an EON tuner model for recognizing different activities such as sitting, running, and walking. The results of the study show that the EON tuner model can achieve high accuracy in recognizing different human activities. The use of Edge Impulse and the accelerometer sensor in combination with the EON tuner provides a simple and efficient method for recognizing human activities in real-world scenarios. This research highlights the potential of using Edge Impulse and EON tuner for developing low-cost and low-power human activity recognition systems for IoT applications.