Human behavior ontologies: Surveillance task

The video sequences analysis is very important topic at the time of using this strategy for surveillance at security places. The problem consist in understand any success in real life saved in a video camera. This process for person recognition, objects, vehicles, danger places, alarms, etc, is the principal goal of this work. The propose is to use ontologies to classify events in recorded videos; for example, people in one place, vehicles in the security zone, etc; generating rules to know what does this event do and trigger this rules at the right time. For this solution we used the relationship between Protégé and Jade generating the agents that use the natural language for the alarms and actions described in the events analysis. The first test was realized in the office with excellent results.