| abstract |
The invention discloses an underwater acoustic 3D touch screen system based on leaked Lamb waves and convolutional neural networks, which can effectively predict the position information and pressure information of touch screen points when the screen surface is covered by liquid. The main structure of the device is Including the glass plate (3), the encapsulation water barrier, the signal generator (1), the piezoelectric ceramic sheet, the post-signal amplifier (6) and the jetson TX1 development version (7). The device transmits pulse waves through one piezoelectric ceramic sheet and the other two piezoelectric ceramic sheets receive acoustic wave signals, and the original data containing the contact information can be obtained. After training the convolutional neural network, the received raw data can be directly input into the trained neural network to obtain the position information and pressure information of the touch point, which can be applied in the field of underwater touch writing. |