Sitting Posture Recognition Using a Spiking Neural Network

Abstract

To increase the quality of citizens' lives, we designed a personalized smart\nchair system to recognize sitting behaviors. The system can receive surface\npressure data from the designed sensor and provide feedback for guiding the\nuser towards proper sitting postures. We used a liquid state machine and a\nlogistic regression classifier to construct a spiking neural network for\nclassifying 15 sitting postures. To allow this system to read our pressure data\ninto the spiking neurons, we designed an algorithm to encode map-like data into\ncosine-rank sparsity data. The experimental results consisting of 15 sitting\npostures from 19 participants show that the prediction precision of our SNN is\n88.52%.\n