PROJECT TITLE :
Home Automation Oriented Gesture Classification From Inertial Measurements
In this paper, a machine learning (ML) approach is presented that exploits accelerometers information to house gesture recognition (GR) issues. The proposed methodology aims at providing high accuracy classification for home automation systems, which are typically user freelance, device freelance, and device orientation freelance, an heterogeneous state of affairs that has not been fully investigated in previous GR literature. The approach illustrated in this paper is composed of 3 main steps: event identification; feature extraction; and ML-primarily based classification. The elements of the novelty of the proposed approach are 1) a preprocessing phase based mostly on principal component analysis to increase the performance in real-world state of affairs conditions and 2) the event of parsimonious novel classification techniques primarily based on sparse Bayesian learning. This methodology is tested on 2 datasets of four gesture classes (horizontal, vertical, circles, and eight-shaped movements) and on a more dataset with eight categories. In order to authentically describe a true-world home automation surroundings, the gesture movements are collected from a lot of than 30 people who freely perform any gesture. It leads to a dictionary of twelve and 20 different movements, respectively, within the case of the four-category and also the eight-class databases.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here