A hybrid HMM/ANN model for activity recognition in the home using binary sensors

Abstract

Activities of daily living are good indicators of the health status of elderly, and activity recognition in a smart environment is a well-known problem that has been previously addressed by several studies. This paper presents a hybrid model based on ANN (Artificial Neural Network) and HMM (Hidden Markov Modeling) techniques in order to tackle the task of activity recognition in a home setting. The output scores of the ANN, after processing, are used as observation probabilities in the model. We evaluate our approach comparing it with classical probabilistic models using three datasets obtained from real data streams. Finally, we show how our approach achieves significative better recognition performance, at a confidence interval of 95%, in several features spaces, proving the hybrid approach to be better suited for the addressed domain. © 2012 Springer-Verlag.