摘要
该文提出了一种基于运动对象外观模型的分层聚类算法,用于从对应同一轨迹的运动对象序列中提取关键对象。采用边方向直方图和主颜色直方图描述运动对象的特征,兼顾形状和颜色特征的全局与局部变化。实验表明,使用该方法提取的关键对象数量较少,且能较好地反映运动目标的形态变化。
This paper proposes a hierarchy cluster method based on the appearance model feature of moving object for the extraction of key objects from the moving objects belong to the same trajectory. In order to take into account the color and shape,global and local changes of the moving objects,the paper uses the dominant color histogram and edge direction histogram as the object＇s appearance model feature. Experiment results show that the key objects extracted by proposing method have a little numbers and can well represent the morphological changes of the moving objects.
出处
《杭州电子科技大学学报（自然科学版）》
2014年第2期73-76,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
基金
浙江省科技计划公益技术研究资助项目(2010C33050)
关键词
运动对象
聚类
主颜色直方图
边方向直方图
key objects
cluster
dominant color histogram
edge direction histogram
