Sitting Posture Recognition Using a Spiking Neural Network

To increase the quality of citizens' lives, we designed a personalized smart chair system to recognize sitting behaviors. The can receive surface pressure data from sensor and provide feedback for guiding user towards proper postures. We used liquid state machine logistic regression classifier construct spiking neural network classifying 15 allow this read our into neurons, an algorithm encode map-like cosine-rank sparsity data. experimental results consisting postures 19 participants show that prediction precision SNN is 88.52%.