Towards Web 3.0
Activity Recognition Process: From Sensor Data to the Final Activity
Latest trends in Ambient Intelligence are focused towards user monitoring systems. These systems are constantly monitoring users in order to detect changes in users’ behavior, to indicate decreased performance, to analyze users’ daily-living dynamics and many more. To successfully fulfill each of these objectives, it is important to know the current user’s activity. In order to recognize activity from sensors data, several steps are first applied. Sensor measurements have to be synchronized, then filtered, furthermore segmented and finally activity can be extracted from the data. In order to recognize final activity, process of removing spurious transitions is also applied over the sequence of predicted activities.
Keywords ambient intelligence, sensors, activity recognition, ICT
The problem of population aging rapidly is considered as a pressing issue over the last few years. Consequently, the number of elderly people who need more intensive healthcare and need to be admitted into a retirement home is also increasing. However, some of these elderly could continue with their normal living habits, if there was someone/something that would constantly monitor their health and take appropriate actions when their health would be at risk. This pressing issue was and currently is addressed by many research groups in the field of ambient intelligence. Furthermore, common goal of these researches is to create novel methods for health monitoring and use them in intelligent healthcare applications. One of the most important components in these applications is activity recognition. It is the basis of several other important components, e.g....
You are not authenticated to view the full text of this chapter or article.
This site requires a subscription or purchase to access the full text of books or journals.
Do you have any questions? Contact us.Or login to access all content.