摘要
交通事故检测是ITS的重要组成部分,文章就现有的模式识别法、事故理论、统计预测和人工智能四类不同的交通异常检测算法进行分析,重点分析各种方法中存在的优点和不足之处,以便为建立具有较高检测率、较低误报率和检测时间的检测算法提供科学的参考。
Accident detection is an important section on ITS. In this paper, the four category accident detection algorithms that include pattern-based algorithms, catastrophe theory, statistical methods and artificial intelligence are analyzed. Its advantage and disadvantage are major analyzed in order to provide a reference when constructing a detection algorithm that has the properties of higher detection rate, lower false alarm rate and time detection.
出处
《交通科技与经济》
2005年第3期50-52,共3页
Technology & Economy in Areas of Communications
基金
四川省交通科研基金(2003.3-7-10)
