基于Leap motion3D体感传感器的钢琴弹奏手部动作跟踪自适应判别系统研究

Research on adaptive identification system of piano hand motion tracking based on Leap motion 3D motion sensor

钢琴演奏的手部动作跟踪,对于演奏者的技巧提高有重要意义。研究针对钢琴弹奏手部动作进行跟踪自适应判别系统研究。实验结果表明,对于不同的动作手势,包括顺指、穿指、跨指和扩指,均采用了350组训练数据以及150组测试数据。顺指手势的识别正确率达到了96%,穿指和跨指手势的识别正确率均为93%,而扩指手势的识别正确率为98%,以上结果验证了系统在动态手势识别方面的高准确性。研究成果有助于帮助改进演奏者的演奏技巧,提高演奏水平。

The tracking of hand movements in piano performance is of great significance for the improvement of performers’skills.Research on a tracking adaptive discrimination system for hand movements in piano playing.The experimental results showed that 350 sets of training data and 150 sets of test data were used for different interactive gestures,including forward,through,cross,and extended fingers.The recognition accuracy of the forward pointing gesture reached 96%,while the recognition accuracy of the through pointing and cross pointing gestures was 93%,while the recognition accuracy of the extended pointing gesture was 98%.The above results verify the high accuracy of the system in dynamic gesture recognition.The research results can help improve performers’performance skills and enhance their performance level.