چكيده به لاتين
Detection of anomalies between lines is an important issue in the field of monitoring. Abnormalities can be an event that does not match the expected pattern; the abnormal discovery problem is divided into two areas: the detection of an anomalies of a line and anomaly detection between a trajectory set.
Although the detection of anomalies between the trajectories is an important issue and is commonly used in important areas such as monitoring and control, little work has been done in this area, and most of the algorithms proposed for this area have weaknesses and Limitations are suffering. In general, most of the work done is trying to find traffic by focusing on the empirical relationship between many traffic factors such as speed and flow.
In this research, different methods of traffic density detection were first investigated and then a high level comparison and general classification on these methods were presented. In the next step, and because raw data is not suitable for processing, a preprocess is performed on raw data that has the steps of uniformizing the time units of the points and matching operations on the map. Subsequently, a congestion-based approach to traffic detection has been proposed, with several benefits, including dynamic zoning of traffic areas, segregation of zones, classification and storage of lines that are present in traffic areas at a time interval. and switching the flow of traffic towards the designated areas for traffic and eliminating other lines to reduce computational time. At the end of the research, various experiments have been carried out using a standard data set on the desired algorithm and the results have been evaluated
