Gesture recognition is the core technology of most gesture interaction systems. The proposed gesture recognition system adopts STM32 F103 single-chip microcomputer minimum system as the main control chip

combines with FDCC2214 capacitive sensor

collects the capacitance value received by FDC2214 through IIC method

implements weighted average algorithm in STM32 to realize training and recognition of gesture signal

and the specific model test results is displayed on OLED LCD. Through training mode

the system can store the memory of the hand gestures of the trainees

and then realize the recognition of the training people’s guessing games（ stone

scissors

cloth） and stroke games（ 1

2

3

4

5） in the judgment mode. After testing

the gesture recognition system works well

it has high recognition rate

and can meet the real-time requirements.