Urban Road Short-term Traffic Flow Forecast Based On Video Analysis

Urban road traffic plays an significant effect in the development of national economy,but the traffic congestion problem has gradually become a serious "urban disease",so that transportation couldn't become the "fast lane" of the economic development,but become the "stumbling block" of the economic development.The problem of traffic congestion not only has negative impact on economic development,but also has serious impact on society,environment,residents' life and psychology.Solving traffic congestion problem is the primary goal to promote the current social development.Traffic control and induction are important methods to solve traffic congestion in the method of solving traffic jam.However,short-term traffic flow forecast is necessary to achieve traffic management,not only serve as the basic way for managers to predict the running state of road network,but also plays an important role in the management of road network.Traffic parameter extraction can provide data support for short-term traffic flow forecasting,with the rapid development of image processing and related technology,the use of video detection technology to extract traffic parameters gradually reflect its superiority,avoid some problems of the traditional traffic parameter detector.Currently,short-term traffic flow prediction is mainly problematic,the problem are mainly derived from two aspects: first,the traffic parameters extraction is not accurate;second,short-term traffic flow forecasting based on single data.In this paper,we analyze the video data to extract traffic parameters,and integrate the real-time data and historical data to forecast the short-term traffic flow.Firstly,this paper introduces the significance of the research,analyzes the advantages and disadvantages of the existing traffic parameter detector,and selects the prediction model which is used in this paper.Secondly,using video detection method and image processing technology to extract traffic parameters.Thirdly,short term traffic flow forecasting based on real time data and historical data.Finally,the effectiveness of the video based traffic parameter extraction method and the short-term traffic flow forecasting method based on information fusion are verified by simulation.