基于RFID技术的动态手势指令识别方法仿真  

Simulation of Dynamic Gesture Instruction Recognition Method Based on RFID Technology

作　　者：许燕[1] 王维兰[1] XU Yan;WANG Wei-lan(School of Mathematics and Computer Science Northwest Minzu University,Gansu Lanzhou 730030,China)

机构地区：[1]西北民族大学数学与计算机科学学院,甘肃兰州730030

出　　处：《计算机仿真》2025年第2期248-251,487,共5页Computer Simulation

基　　金：中央高校基本科研项目(31920220131);武汉学院2020年度科技创新团队项目(KYT202001)。

摘　　要：与静态手势指令不同,动态手势指令通过手的运动轨迹和速度来表达方向、速度、力度等内容,信息更加丰富,因此识别难度更大。为了有效提升动态手势指令识别效果,提出一种基于RFID技术的动态手势指令识别方法。通过RFID技术采集动态手势指令,采用小波变换对动态手势指令图像展开分解处理,获取初步去噪动态手势指令图像。同时引入多尺度双边滤波算法对动态手势指令图像再次处理。将经过去噪处理后的动态手势指令图像转换为两种变换特征,将其作为特征识别的依据。针对变换特征使用多中心模糊C均值聚类算法提取动态手势指令特征,将全部特征的编码信息存储到知识库中,通过改进编辑距离计算待测特征动作序列和知识库中动态手势指令编码信息的相似度,根据相似度匹配完成动态手势指令识别。仿真测试结果表明,采用所提方法可以有效提升动态手势指令识别结果的准确性,且识别效果较好。At present,dynamic gesture commands express direction,speed,force,and other content through the movement trajectory and speed of the hand,thus providing richer information.To effectively improve the recognition of dynamic gesture commands,this article put forward a method of recognizing dynamic gesture commands based on RFID technology.Firstly,RFID technology was used to collect dynamic gesture commands.Moreover,wavelet transform was used to decompose the dynamic gesture command image,and thus to obtain a preliminary denoised image.Simultaneously,a multi-scale bilateral filtering algorithm was introduced to reprocess dynamic gesture command images.Then,these images were converted into two types of transformation features as the basis for feature recognition.Aiming at the transformation features,a multi-center fuzzy C-means clustering algorithm was used to extract the dynamic gesture command features,thus storing the encoding information of all features in the knowledge base.Furthermore,the improved edit-distance was used to calculate the similarity between the action sequence of the feature to be detected and the encoding information of dynamic gesture command in the knowledge base.Finally,the recognition for dynamic gesture command was completed by similarity matching.Simulation results prove that the proposed method can effectively improve the accuracy of dynamic gesture command recognition.

关 键 词：手势识别 小波变换 多中心模糊均值聚类算法 相似度 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]