基于Adaboost算法的高速公路事件检测

本文介绍Adaboost方法的基本原理及算法；阐述了高速公路事件检测原理并进行了参数选择，确定了神经网络的结构，提出利用Adaboost方法进行高速公路事件检测并给出了该方法事件检测的算法步骤，最后进行了仿真实验。实验结果表明，该算法可以大大提高弱分类算法的性能，具有较高的检测率和较低的误报率，适于高速公路事件检测。

Freeway Incident Detection Based on the Adaboost Algorithm

This paper describes the principle and algorithm of the Adaboost method. By introducing the principle of free- way incident detection and parameter choice, the topology of neural networks is employed in this paper, and an improved freeway incident-detection algorithm and the steps of its processing are presented based on the Adaboost algorithm. In addi- tion, a simulation experiment is conducted to test the feasibility and validity of this algorithm. The result of the experiment shows that this algorithm can highly enhance the performance of weak classification algorithms with a higher detection rate and a lower false alarm rate, which assesses the effectiveness of its application to freeway incident detection.