摘要
为了实现人-机-环的自然交互,解决按键以及图形界面等交互方法存在的用户学习成本高、操作不便等问题,提出了一种采用主动声传感系统的局域固有频率识别的实体交互方法。该方法首先运用主动声传感技术构造多频率恒频声源信号,以激励交互实体系统按固有频率产生振动响应;然后识别并提取局域固有频率特征,对不同时长和幅值进行参数组合,选择稳定性最佳的组合模式定义成不同交互手势;最后选择KNN算法进行特征分类和识别。设计了以木质台面控制雾化玻璃状态的交互实验,实验结果表明:所提局域固有频率识别方法能有效地过滤噪声,对交互过程中边界条件的改变响应迅速且结果准确;相较于传统操控方法,所提方法的最大优势在于运用了自然人机交互方式,能够有效降低用户的学习成本,使“物物皆界面,处处可交互”成为可能。
To realize the multi-channel natural interaction and solve users’problems of high learning cost and inconvenient operation in the key interaction and graphical interface interaction methods,this paper proposes a tangible interaction method based on recognizing the local natural frequency of active acoustic sensing system.Firstly,the active acoustic sensing technology is used to construct multi-constant-frequency sound signals,and to excite interactive entity system generating vibration resonance according to the natural frequency.Then,the local natural frequency characteristics are identified and extracted,and the parameters with different durations and amplitudes are combined.The combination of parameters with the highest stability is selected and defined as different interactive gestures.Finally,the KNN classification algorithm is selected for feature learning and recognition.An application experiment using wooden countertops to control fog glass is designed.Experimental results show that the system can effectively filter noise and respond the change of boundary conditions quickly with accurate results in the interaction process.Compared with traditional interaction methods,the method has the advantage of using a natural and intuitive way to effectively reduce users’learning cost,making the concept of“things are all interface,everywhere can be interactive”possible.
作者
李晓玲
张丽霞
李佳伟
高远
朱慧进
郑子明
LI Xiaoling;ZHANG Lixia;LI Jiawei;GAO Yuan;ZHU Huijin;ZHENG Ziming(School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an 710049, China;International Joint Research Center for Digital Medical Devices and Instruments, Xi’an Jiaotong University, Xi’an 710049, China)
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2020年第6期10-16,35,共8页
Journal of Xi'an Jiaotong University
基金
陕西省重点研发计划资助项目(2018GY-142)。
关键词
自然人机交互
局域固有频率
实体交互
主动声传感
手势
natural human-computer interaction
local natural frequency
tangible interaction
active acoustic sensing
gesture
