摘要
针对某小型地面无人作战平台控制手势识别率低的问题,该文提出了一种控制手势识别方法。首先,对控制手势的肌电图(EMG)信号进行预处理,提取平均绝对值、平均绝对值斜率、波长和方差4种特征构成特征集,输入支持向量机(SVM)中进行分类;然后,针对其中的相近手势引入了角速度信号和EMG信号特征融合的方法进行识别,并进行了手势控制地面无人作战平台的实验验证。结果表明,该文方法的识别率较传统Hudgins特征集识别方法提高了3.29%,在引入角速度信号后相近手势识别率提高了13%,手势整体识别率提高了5.56%。
Aiming at the problem of low control gesture recognition rate of a small ground unmanned combat platform,a control gesture recognition method is proposed here.Firstly,the electromyography(EMG)signals of the control gesture are preprocessed,and the four features of average absolute value,average absolute value slope,wavelength and variance are extracted to form a feature set,which is input into support vector machine(SVM)for classification.Then,the angular velocity signal and EMG signal feature fusion is introduced for similar gestures’identification,and the experimental verification of gesture-controlled ground unmanned platform is carried out.The results show that the recognition rate of the proposed method is 3.29%higher than that of the traditional Hudgins feature set recognition method.After the angular velocity signal is introduced,the recognition rate of similar gestures is increased by 13%,the overall recognition rate of gestures is increased by 5.56%.
作者
梅武松
陈科仲
李忠新
Mei Wusong;Chen Kezhong;Li Zhongxin(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China;Military Representative Bureau of Army Equipment Department in Chongqing,Chongqing 400060,China)
出处
《南京理工大学学报》
CAS
CSCD
北大核心
2022年第3期262-269,共8页
Journal of Nanjing University of Science and Technology
基金
中央高校基本科研业务费专项资金(30918012203)。
关键词
地面无人作战平台
手势识别
肌电图信号
支持向量机
角速度
特征融合
ground unmanned combat platform
gesture recognition
electromyography signal
support vector machine
angular velocity
feature fusion
