Based on Intelligent Space Local Behavior Analysis for Service Object
Haitao Li, Hanqing Zhang · 2014
Abstract—According to the position behavior characteristics of service object, this paper firstly determines the environmental mark point by combining the position point clustering with map topological node. By using the state of family environmental mark point to transfer fitting continuous position trajectory, the overall position and behavior pattern of of service object can be analyzed; the possible target position and residence time, and whether it’s in an abnormal state can be determined. At the same time, because of the high computational complexity and poor real-time performance existing in the traditional modeling and identification process, it further introduces the environmental mark point state residence time distribution, combined with the retention time of service object in different markers region, to optimization the model. Finally, this paper detects the service object position, and the experiments show that the model can depict the position movement well, and can help analyze and judge according to the anomaly detection results. Keywords- Position Behavior; Environmental mark point; Anomaly Detection; Trajectory: Service object