Project: Automatic Realtime System for Traffic Security and Violation Detection in Roads
Every day millions of vehicles are being driven on the roads worldwide, making control and monitoring is becoming more and more important. Currently, this need has led to the creation of traffic control centres in every city or roadway, allowing better management of traffic, and acting against accidents and traffic violations that occur in roads. _x000D__x000D_The appearance of these control centres and their demand of improvements, have created a market focused on the increment of their capacities, such as operability, usability and information management. Therefore, currently, there are a set of applications that cover small niches of this market, but that are not yet matured enough._x000D__x000D_The CO objective of the project is the creation and evaluation of a new tool for the automatic traffic management, based on the already installed network of visual sensors, with the purpose of improving the current capacities. For this aim, the project presents a set of tools, structured in functionality layers, which are described briefly below:_x000D__x000D_Firstly, the present applications do not provide an automatic configuration of the system for the traffic control centres. The installation or change of location of any visual sensor of the control network of roads involves a manual configuration, being this more complex when more complete is the information that must provide the application. In order to solve this defect of the present solutions, SAFEROADS provides a semantic understanding of the critical components of the circulation routes (lanes, verges, etc.) by means of the creation of an abstract definition of the road that allows an automatic or semiautomatic configuration. _x000D_Additionally, this automatic configuration reduces the human error factor as much as possible. On the other hand, it ensures the right functionality of the system before changes of the visual sensors, by means of a reconfiguration in real time. These functionalities are gathered in the most inner layer called SoUL (Semantic Understanding Layer) of SAFEROADS's application._x000D__x000D_In addition, current traffic control centres use computer vision techniques to be able to detect the riskiest situations that could take place in roads in order to improve the security. These improvements in traffic management procedures provide a large number of indirect benefits like reduction of CO2 emissions by traffic jams or reduction of accidents and deaths. Moreover, this kind of solutions provides data about traffic flows that is very useful for behaviour analysis and corrective measure decision._x000D_But the two CO problems of these applications are: first, the small number of incidents that can detect and the poor information and statistics that they calculate, and second, extreme dependency of the scenario, because it is necessary different products for different scenarios (roads, bridges, tunnels, etc.) despite they are conceptually the same._x000D_For this reason, SAFEROADS defines an intermediate layer, called CONSuL (CONtrol and Statistics Layer), in order to design one of the most complete tool for the control and data management of traffic that is independent of application scenario and can detect and calculate more relevant incidents and statistics than current solutions, for example, relationship between incidents and weather conditions, use of lane by type of vehicle, etc._x000D__x000D_Finally, in the field of traffic management there are not many solutions to detect violations, but the prevention of these violations is very important to avoid high-risk situations. The current solutions are designed to detect just a violation per solution, so it is necessary to integrate as many solutions as violations. Furthermore, many of these solutions need a very specific system configuration and it is very complex and expensive to install._x000D_SafeRoad's solution adds a punitive function, by means of the upper layer, called DEVIL (DEtection of traffic VIolation Layer), that allows an automatic detection of traffic violations. This functionality is possible by means of the video analysis, in which is captured and stored all the needed evidences to initiate the punitive actions applicable to the infraction and according to traffic laws. _x000D__x000D_A solid consortium of highly relevant and complementing Ps has been formed to achieve the expected objectives. Treelogic is a R&D-performing SME and, as the project coordinator, has wide experience in managing national and European research projects. In addition to this, Treelogic actively participates in the development of this solution using computer vision techniques involving their experts on digital processing._x000D__x000D_Citilog SA is another R&D-performing SME with significant experience in the field of the design of traffic management applications, including current successful applications to management the traffic in roads. Additionaly, Citilog has wide experience in European project as IVISCAT, a previous Eurostars project (Call 8).
|
Acronym
|
SAFEROADS
(Reference Number: 8217)
|
|
Duration
|
01/10/2013 - 31/12/2015
|
|
Project Topic
|
The goal is to develop and test a new integrated system for traffic management. This application is focused on the control of real time traffic, collection of statistics about traffic flow, and development of a security system in order to detect high-risk situations and traffic violations.
|
|
Network
|
Eurostars
|
|
Call
|
Eurostars Cut-Off 10
|
Project partner
