Real-time monitoring system for potentially dangerous activities detection
Aleksandra Postawka · 2017
Cognitive impairments are an unavoidable community problem. People suffering from such diseases need all day long attention with varying care difficulty depending on the type of disorder. What makes care harder in the case of autism is the frequent occurrence of self aggressive behaviors. The monitoring system is supposed to detect such situations and differentiate them from similar normal activities. In this paper the Averaged Hidden Markov Models are used for potentially dangerous activities detection in the real-time monitoring system. The acceleration measure has been used in order to discriminate dangerous and normal situations. Additionally, algorithms for real-time activity recognition have been presented. The experiments have been conducted for a set of data containing hitting and touching sequences obtained from the depth sensor.