摘要： 传统输入设备无法满足人们的随意性输入需求。为此，提出一种基于陀螺仪传感器的三维手势识别方案。硬件架构由陀螺传感器信息采集模块、单片机信息处理模块以及射频无线传输模块组成。利用多功能滤波器进行数据预处理，设计一种基于角度的特征提取算法，提取三维手势特征。实验结果表明，该方案的平均识别率达到99.3%，能较好地实现3D空间的鼠标输入功能和键盘输入功能。

关键词: 3D输入, 手势识别, 陀螺传感器, 人机交互, 支持向量机, 多类分类器

Abstract: Traditional input devices are increasingly unable to meet people’s needs of random input. Aiming at the defect, a 3D gesture recognition scheme based on gyroscope sensor is proposed. The hardware architecture of scheme includes a high accuracy data acquisition hardware system based on gyroscope and a MCU based on atmega16 and a wireless transmission module based on nRf24API. The gesture recognition algorithm includes data preprocessing module based on multisection filter. A feature extraction algorithm is based on angular movement to extract the 3D gesture feature. Experimental results show that the average recognition rate of this scheme is 99.3% which can well realize 3D space mouse input function and the keyboard function.

Key words: 3D input, gesture recognition, gyro sensor, human-computer interaction, Support Vector Machine(SVM), multi-class classifier

中图分类号: 

• TP393