论文部分内容阅读
利用主成分分析方法,研究新型的基于主成分分析的交通事件自动检测算法。主成分分析方法是一种常用的在线性数据中进行特征提取的方法。本文首先对输入数据进行主成分分析,然后在主成分空间中选取占特征值变化范围90%的特征值作为主成分,计算其主成分核。最后对每个实例的主成分核进行卡方分配的适应性检验,以判断该实例是否为事件实例。对I-880高速公路事件数据的测试结果表明,该算法获得了较好的检测效果。
Using principal component analysis method, this paper studies a new automatic traffic incident detection algorithm based on principal component analysis. Principal component analysis is a commonly used method for feature extraction in linear data. In this paper, the principal component analysis is carried out on the input data firstly, and then the eigenvalues accounting for 90% of the variation range of the eigenvalues are selected as the principal components in the principal component space to calculate the principal component kernel. Finally, the principal component kernel of each instance is tested for the chi-square assignment to determine whether the instance is an event instance. The test results of the I-880 freeway incident data show that the algorithm has a good detection effect.