摘要
针对城市快速路交通状态划分问题,提出一种改进的模糊C-均值(FCM)算法.为了解决FCM算法对初始聚类中心敏感、聚类前必须对聚类数和模糊加权指数给出恰当赋值等问题.首先采用减法聚类得到最大聚类数及相应的初始聚类中心,然后基于模糊决策的方法优选参数m,最终将聚类有效性函数融入FCM聚类,动态确定交通状态的分类.对上述方法通过Matlab6.5编程得出结果,对比分析表明提出的方法能够提高城市快速路交通流状态的分类效果.
For the recognition of urban expressway traffic flow situation, this paper proposes an improved algorithm using improved fuzzy C-means clustering method. FCM is sensitive to the initial cluster centers and needs to determine fuzzy weighting exponent and the number of cluster type. In order to solve those problems, we use subtractive clustering to get the maximum of cluster number and cluster centers, and use fuzzy decision theory to optimize the parameter m. FCM with the cluster validity function is applied to perform the cluster, programming by matlab 6.5 to confirm the proposed methodology. A comparative analysis shows positive and encouraging results.
出处
《西南民族大学学报（自然科学版）》
CAS
2012年第6期905-909,共5页
Journal of Southwest Minzu University(Natural Science Edition)
基金
四川省科技支撑计划项目(2011FZ0050)
关键词
FCM算法
减法聚类
模糊决策
fuzzy C-means algorithm
subtractive clustering
fuzzy decision
partition fuzzy degree
