摘要
立体足迹重压面的分布反映了人体的行为特征和心理特征,立体足迹的分割和描述是足迹生物特征识别的基础。本文提出了一种基于高斯曲率和平均曲率的足迹深度图象交互式分割方法,可以直观、有效地分割出我们感兴趣的重压面区域。并在此基础上提出了基于足迹全局和区域特征的描述方法,特征均能反映人体固有的生理特征,具有较强的鲁棒性。
The distribution of footprint heavy pressure surface refleets the behavior feature and the physiological feature of human body. So. footprint heavy presstlre surface pick-up and description is the foundation of footprint biological feature identification. We put forward a kind of footprint heavy pressure surfaces pick-up method based on Gauss curvature and average curvature. This method can piek-up effectively footprint heavy pressure surfaces that we arc interested. And on this foundation, we still put forward the description method based on footprint range image overall and regional features which can reflect the built-in hiological feature of human body. The result of experiment shows that this description method is simple and practicality.
出处
《微计算机信息》
北大核心
2005年第09X期103-104,125,共3页
Control & Automation
基金
国家自然科学基金资助项目:60272004
关键词
重压面
曲率
特征描述
heavy pressure surface
curvature
feature description.
