基于Leap Motion传感器的弹琴触键手势自动控制系统设计  

Design of an automatic control system for piano touch gesture based on Leap Motion sensor

作　　者：李婷婷 王靖 骆亚丽 刘红梅 LI Tingting;WANG Jing;LUO Yali;LIU Hongmei(Aksuvocational and Technical College,Aksu Xinjiang 843000,China)

机构地区：[1]阿克苏职业技术学院,新疆阿克苏843000

出　　处：《自动化与仪器仪表》2025年第2期223-227,共5页Automation & Instrumentation

摘　　要：人工智能技术的发展为传统乐器学习带来了新的助力。为了提升弹琴触键手势的识别率,帮助钢琴弹奏者控制自己的错误手势,研究构建了弹琴触键手势自动控制系统,并设计了该系统的功能模块。研究采用了Leap Motion传感器来采集数据,并设计了特征提取模块。研究设计了基于支持向量机和改进K最近邻算法的手势识别模型。结果显示,手势识别模型在训练集和测试集上准确率的最大值分别为98.88%和98.59%,平均识别时间为15.85 ms。研究设计系统的最大和最小吞吐量分别为82 400 bit/s和80 300 bit/s,响应时间最大值和最小值分别为33.51 ms和27.40 ms。研究设计系统和模型的性能良好,能够为弹琴触键手势的识别提供技术上的支持,促进钢琴弹奏者控制自身的错误手势。The development of artificial intelligence technology has brought new assistance to the learning of traditional musical instruments.In order to improve the recognition rate of piano touch gestures and help piano players control their erroneous gestures,an automatic control system for piano touch gestures was studied and constructed,and the functional modules of the system were de-signed.The study used Leap Motion sensors to collect data and designed a feature extraction module.A gesture recognition model based on support vector machine and improved K-nearest neighbor algorithm was studied and designed.The results showed that the maximum accuracy of the gesture recognition model on the training and testing sets were 98.88%and 98.59%,respectively,with an average recognition time of 15.85 ms.The maximum and minimum throughput of the research and design system are 82400 bit/s and 80300 bit/s,respectively,and the maximum and minimum response times are 33.51 ms and 27.40 ms,respectively.The research and design system and model have good performance and can provide technical support for the recognition of piano touch gestures,promoting piano players to control their own erroneous gestures.

关 键 词：Leap Motion 钢琴 手势 识别 模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]