基于柔性力敏传感器的左右脚动态识别方法 |
| |
引用本文: | 汪飞跃,姚志明,许胜强,魏凯,杨先军. 基于柔性力敏传感器的左右脚动态识别方法[J]. 传感技术学报, 2015, 28(7): 964-971. DOI: 10.3969/j.issn.1004-1699.2015.07.004 |
| |
作者姓名: | 汪飞跃 姚志明 许胜强 魏凯 杨先军 |
| |
作者单位: | 1. 安徽大学电子信息工程学院,合肥230009; 中国科学院合肥智能机械研究所安徽省仿生感知与先进机器人技术重点实验室,合肥230031;2. 中国科学院合肥智能机械研究所安徽省仿生感知与先进机器人技术重点实验室,合肥230031; 江苏省体育科学研究所,南京210033; 北京体育大学,北京100084;3. 中国科学院合肥智能机械研究所安徽省仿生感知与先进机器人技术重点实验室,合肥,230031 |
| |
基金项目: | 安徽省科技攻关计划课题项目(1301042215,1501021042);国家科技支撑计划项目(2013BAH14F01) |
| |
摘 要: | 在利用柔性力敏传感器获取动态足底压力分布数据时,能够准确快速自动区分左右脚的数据将极大提升数据的可视性和分析的便利性。为此,提出了一种基于足底压力和脚印外观形状的左右脚动态识别方法。首先,基于足底动力学原理,利用连通域的图像分割算法对足底压力数据进行聚类分析,得到每一步压力脚印的时间和坐标范围;在此基础上进一步分离出完整的单步压力数据;最后利用单步压力数据刻画脚印轮廓,并根据轮廓的外观特征进行左右脚识别。本文提出的方法可应用于步态分析、临床辅助诊断、步态识别等领域。通过108个实测数据样本的测试表明:本文方法的识别率高达94.5%,并具有较好的鲁棒性。
|
关 键 词: | 柔性力敏传感器 连通区域算法 轮廓特征 聚类分析 脚印识别 |
Dynamic Footprint Recognition Method Based on Flexible Force-sensitive Sensor |
| |
Abstract: | Distinguishing the left and right footprint data accurately,rapidly and automatically by using the flexible force-sensitive sensor for acquiring the dynamic plantar pressure distribution data can greatly improve the visibility of data and the convenience of data analysis. Therefore,a novel method for footprints recognition based on the plan?tar pressure and the appearance shape features of the footprint is presented. The concrete implementation steps are as follows. Firstly,on account of the principle of the plantar kinetics,the connected component algorithm is applied to data clustering analysis,the coordinate and time range of the target footprint are obtained at the same time. Then we gain the complete plantar pressure data of each footprint and extract the appearance shape features of them. Fi?nally,footprint recognition is accomplished on the basis of the plantar pressure and the appearance shape features. The proposed method can be applied to many fields,such as gait analysis,clinical diagnosis and gait recognition. Experimental results of 108 samples in normal walking pattern show that the proposed method not only has a high recognition rate (94.5%),but also has strong robustness. |
| |
Keywords: | flexible force-sensitive sensor connected component algorithm shape features data clustering analy-sis footprint recognition |
本文献已被 万方数据 等数据库收录! |
| 点击此处可从《传感技术学报》浏览原始摘要信息 |
|
点击此处可从《传感技术学报》下载全文 |