| 基于惯性传感器MPU6050的手势识别方法 |
| |
| 引用本文： | 张平,刘祚时.基于惯性传感器MPU6050的手势识别方法[J].传感器与微系统,2018(1):46-49,53. |
| |
| 作者姓名： | 张平 刘祚时 |
| |
| 作者单位： | 江西理工大学机电工程学院,江西赣州,341000 |
| |
| 基金项目： | 国家自然科学基金资助项目，江西省科技计划资助项目，江西省研究生创新专项资金资助项目 |
| |
| 摘 要： | 针对基于加速度规律的手势识别方法未充分利用陀螺仪的数据进行手势分类和识别的问题,提出了一种基于MPU6050惯性传感器的特征提取手势识别方法,通过提取加速度和姿态角信号在手势上的特征量,利用决策树对手势进行预分类,结合加速度和姿态角的变化规律完成了手势的具体识别.依据预定义手势选择10位试验对象进行测试,获得了96.4%的平均识别率,识别时间小于0.005 s.方法对基于自带数字运动处理器的惯性传感器的手势识别具有一定的参考价值.
|
| 关 键 词： | MPU6050 手势识别 特征提取 数字运动处理器 MPU6050 gesture recognition feature extraction digital motion processor(DMP) |
Gesture recognition method based on inertial sensor MPU6050 |
| |
| ZHANG Ping,LIU Zuo-shi.Gesture recognition method based on inertial sensor MPU6050[J].Transducer and Microsystem Technology,2018(1):46-49,53. |
| |
| Authors: | ZHANG Ping LIU Zuo-shi |
| |
| Abstract: | Aiming at problem that gesture recognition method based on acceleration does not make full use of data of gyro to classify and recognize gesture,a method for gesture recognition based on MPU6050 sensor is proposed. Through extracting acceleration signal and attitude angle signal,reflect on gesture characteristic quantity.Gestures are presorted by decision tree. Gestures are recognized according to acceleration and attitude angle changing regulation.The gesture verification test is carried out on ten experimenters,average recognition rate is 96.4% and recognition time is less than 0.005 s. The method has certain reference value for gesture recognition based on inertial sensor with digital motion processor(DMP). |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
|
