Method for traffic safety state detection of urban road based on cloud architecture

Aiming at the problems of poor real-time performance and accuracy, high complexity and energy consumption in current traffic safety state detection, a method for traffic safety state detection of urban road based on cloud architecture is proposed. According to the overall structure of traffic safety state detection of urban road under cloud architecture, data mining processing of urban road traffic is realized through missing data repair and error data filtering. Based on the results of data processing, the road traffic system is divided into three parts: participants, objects and traffic organization and management. The representative indicators of the three parts are selected as indicators of traffic safety status detection. Rough set is introduced to determine the weight of detection indicator by constructing decision table, calculating attribute dependence degree, calculating the importance degree of an attribute and normalizing the calculation result of indicator importance degree. Combining the comprehensive score of the detection indicators and the comprehensive weight of the indicators, the detection model is constructed to realize the traffic safety state detection. The experimental results show that the method has high detection efficiency and accuracy, low complexity and energy consumption, as well as strong robustness.