A New Algorithm on Mice Action Analysis

摘要：

Mice action data is important part of the experimental data in neurology, physiological etc. subjects. A algorithm was proposed on action analysis of mice in this paper. Difference of mice in interframe was extracted while the position relationship between mice contours and change information was taken into account. Then K-Means clustering algorithm was used to cluster the information while the cluster center was used as the key frame. In the recognition phase, the similarity measure function was defined according to practical significance of the center of K-Means algorithm, the most similar sample class with it was the result. Experimental results showed that the action analysis algorithm is very effective.

关键词： action analysis mice similarity change information K-Means algorithm

作者: Liurong Hong

作者单位: School of Computer & Technology, Huaibei Normal University, AnHui,235000,China

会议类型: 国际会议

会议名称: 2011 International Conference on Mechatronics and Applied Mechanics(2011年机电一体化与应用力学国际会议 ICMAM 2011)

会议地点: 香港

会议语种:英文

页码: 494-498

在线出版日期: 2011-12-27（万方平台首次上网日期，不代表论文的发表时间）