A gesture recognition device based on FDC2214

Abstract

The present invention belongs to the field of human-computer interaction technology, and specifically relates to a gesture recognition device based on FDC2214. The gesture recognition device includes an MSP430G2553 main control unit, an FDC2214 capacitive sensor detection unit and a human-computer interaction unit; the main control unit adopts an MSP430G2553 chip and includes a power supply circuit, a clock crystal oscillator circuit and a reset circuit to ensure the normal operation of the processor. The power supply circuit is provided with a ground filter capacitor at the input and output ends of the voltage stabilizing chip to provide a stable 3.3V voltage for the entire device. The clock circuit includes a low-speed crystal oscillator circuit, which is used when the main control unit is dormant. The present invention overcomes the shortcomings of the prior art solutions, such as high cost, complex algorithms or reduced accuracy and stability due to the influence of the external environment. After a short training of the user, the present invention not only has a high recognition accuracy, but also has high recognition speed and stability.

Description

Technical Field

[0001] The invention belongs to the technical field of human-computer interaction, and in particular relates to a gesture recognition device based on FDC2214. Background Art

[0002] With the rise of electronic information technology and smart devices, people have higher and higher requirements for the convenience and real-time performance of human-computer interaction. At present, emerging technologies such as gestures, face recognition, and fingerprint recognition have been widely used in the field of human-computer interaction. As an intuitive and natural way of communication in people's daily lives, gestures play an important role in the field of human-computer interaction.

[0003] After years of development, gesture recognition technology has become increasingly mature, and many products are trying to add gesture recognition to smart wearable devices, VR, smart homes, and other fields. At present, the mainstream solutions for gesture recognition are mainly based on computer vision, infrared sensors, inertial sensors, and optical technology. Based on computer vision recognition technology, its recognition process is generally divided into four parts: gesture image capture, gesture segmentation, gesture feature extraction, and gesture detection; its recognition sensitivity and accuracy are determined by the camera and processor. However, due to the high cost of cameras and the difficulty in developing recognition algorithms, its technology has not been widely popularized. The solution for gesture recognition based on infrared sensors uses infrared tubes as input and output signals, and uses a single-chip microcomputer to control and detect and process signals. It is a low-cost gesture detection solution. The recognition distance of the infrared tube solution depends on the performance of the infrared tube, and the infrared tube is greatly affected by light intensity, which will seriously affect performance and stability. This solution has not been fully adopted. The gesture recognition solution based on inertial sensors mainly uses accelerometers and gyroscopes to obtain the hand's motion status information to determine the gesture. The recognition results are relatively accurate and stable, but the recognition requires the user to wear a specific device on the hand for recognition, so the scope of application is greatly limited. Summary of the invention

[0004] In order to solve the problems of the prior art, the present invention invents a gesture recognition device based on a FDC2214 capacitive digital sensor.

[0005] The technical solution of the present invention:

[0006] A gesture recognition device based on FDC2214 includes an MSP430G2553 main control unit, an FDC2214 capacitive sensor detection unit and a human-computer interaction unit;

[0007] The MSP430G2553 main control unit adopts the MSP430G2553 chip as a core processor, and includes a power supply circuit, a clock crystal oscillator circuit and a reset circuit to ensure the normal operation of the processor.

[0008] The power supply circuit uses the LM1117-3.3 voltage regulator chip with an input voltage range of 5-12V. Ground filter capacitors are set at the input and output ends of the voltage regulator chip to provide a stable 3.3V voltage for the entire device. The clock circuit includes a 32.768kHz low-speed crystal oscillator circuit, which is used when the main control unit is dormant.

[0009] The reset circuit uses a manual RC reset circuit, which is used to reset the main control unit when an abnormality occurs.

[0010] The FDC2214 capacitance sensor detection unit comprises an FDC2214 chip and a capacitance to be measured.

[0011] The IN0A terminal of the input channel 0 of the FDC2214 chip is connected to one end of the inductor L1 and one end of the capacitor C9. The IN0B terminal of the input channel 0 is connected to the other end of the inductor L1 and the capacitor C9. The inductor L1 and the capacitor C9 form an LC oscillation circuit. One end of the measured capacitor CSensor1 is connected to one end of the inductor L1, and the other end of the measured capacitor CSensor1 is connected to GND. The IN1A terminal of the input channel 1 is connected to one end of the inductor L2 and one end of the capacitor C11. The IN1B terminal of the input channel 1 is connected to the other end of the inductor L2 and the capacitor C11. The inductor L2 and the capacitor C11 form an LC oscillation circuit. One end of the measured capacitor CSensor2 is connected to one end of the inductor L2, and the other end of the measured capacitor CSensor2 is connected to GND. The IN3A terminal of input channel 3 is connected to one end of inductor L3 and one end of capacitor C12, the IN3B terminal of input channel 3 is connected to the other end of inductor L3 and capacitor C12, the inductor L3 and capacitor C12 form an LC oscillation circuit, one end of the measured capacitor CSensor3 is connected to one end of inductor L3, and the other end of the measured capacitor CSensor3 is connected to GND. The IN4A terminal of input channel 4 is connected to one end of inductor L4 and one end of capacitor C9, the IN4B terminal of input channel 4 is connected to the other end of inductor L4 and capacitor C13, the inductor L4 and capacitor C13 form an LC oscillation circuit, one end of the measured capacitor CSensor4 is connected to one end of inductor L4, and the other end of the measured capacitor CSensor4 is connected to GND. The LC circuit composed of the inductor and the capacitor will generate an oscillation frequency f S .

[0012] Furthermore, the capacitor to be measured is four copper sheets with different areas, and the four copper sheets with different areas are arranged in the shape of five fingers spread out.

[0013] The VDD pin of the FDC2214 chip is connected to the output of the power supply circuit, and the ADDR, GND, and PAD pins are connected to GND. The CLKIN of the FDC2214 chip is connected to the output of an external 40Mhz high-precision crystal oscillator through a series 0 ohm resistor. The 40Mhz high-precision crystal oscillator provides a reference frequency f for the FDC2214 chip. REF , to measure the oscillation frequency f generated by the LC oscillation circuit of each channel S The FDC2214 chip will measure the oscillation frequency f S Converted into digital quantity, at the same time, the IIC pins SCL and SDA of the FDC2214 chip are connected to the two pins P1.5 and P2.0 of the MSP430G2553 chip, and the MSP430G2553 reads the conversion result of the FDC2214 chip through the IO simulation IIC protocol.

[0014] The human-computer interaction unit includes an OLED display screen and a key circuit. The present invention adopts an OLED display screen with low power consumption and small size. The OLED_SCL and OLED_SDA pins of the OLED display screen are respectively connected to the pins P2.1 and P2.2 of the MSP430G2553 chip. The IO simulates the IIC protocol to communicate to realize multi-functional display. The pin 1 of the OLED display screen is connected to the output end of the power supply circuit, and the pin 2 is connected to GND. One end of the key KEY1 of the key circuit is connected to the pin P2.5 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY2 is connected to the pin P2.4 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY3 is connected to the pin P2.3 of the MSP430G2553 chip, and the other end is connected to GND.

[0015] The system program means that after the gesture recognition device is turned on, the main control chip MSP430G2553 and the sensor FDC2214 are initialized first. After the initialization is completed, the main page is displayed on the OLED display screen, prompting the user to use the button to select the recognition mode or enter the training mode for training. If the user presses button 1, it enters mode 1 to recognize 5 gestures, namely gestures 1, 2, 3, 4, and 5. If the user presses button 2, it enters mode 2 to recognize 3 gestures, namely rock, scissors, and cloth. If the user presses button 3, it enters the training mode, which is set to adapt to people with different hand shapes. After entering the training mode, the OLED display screen prompts the user to use the button to select the mode again. If the user presses button 1, it enters the training mode of mode 1; if the user presses button 2, it enters the training mode of mode 2. After entering the training mode, the trainee is required to perform the five gestures of mode 1 or the three gestures of mode 2 in sequence. After the trainee performs the gesture, the main control unit will learn and enter the trainee's hand shape data for storage. After completing the training, the gesture recognition program will be directly entered. After the user makes a gesture, the main control unit will compare and judge the capacitance value of the measured capacitor and recognize the gesture.

[0016] The beneficial effects of the present invention are as follows: the present invention proposes a new gesture recognition solution, the adopted FDC2214 capacitive sensor is not affected by the environment such as light intensity, and has high accuracy and stability. The shortcomings of the existing technical solutions, such as high cost, complex algorithm, or reduced accuracy and stability due to the influence of the external environment, are overcome. After a short training of the user, the present invention not only has a high recognition accuracy, but also has high recognition speed and stability. At the same time, the present invention is a low-cost and convenient gesture recognition technical solution, which is convenient for wide application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a system structure diagram of a gesture recognition device;

[0018] Figure 2 This is the circuit schematic diagram of the gesture recognition device.

[0019] Figure 3 It is a flowchart of the gesture recognition device system program;

[0020] Figure 4 This is the layout diagram of the capacitor to be measured.

[0021] Among them: 1 FDC2214 capacitance sensor detection unit; 2 power supply circuit; 3 MSP430G2553 chip; 4 crystal oscillator circuit; 5 MSP430G2553 main control unit; 6 reset circuit; 7 human-computer interaction unit. DETAILED DESCRIPTION

[0022] The present invention is further described in conjunction with the technical solution and the accompanying drawings. The FDC2214-based gesture recognition device of the embodiment includes an MSP430G2553 main control unit 5 , an FDC2214 capacitive sensor detection unit 1 and a human-computer interaction unit 7 .

[0023] The output terminal 3V3 of the power supply circuit 2 is connected to the power pin VDD of the MSP430G2553 chip 3, the GND of the power supply circuit 2 is connected to the pin GND of the MSP430G2553 chip 3, and the output terminal 3V3 of the power supply circuit 2 is simultaneously connected to the power pins of the FDC2214 capacitive sensor detection unit 1 and the human-computer interaction unit, providing a stable voltage for each unit. The crystal oscillator circuit 4 and the reset circuit 6 are connected to the corresponding pins of the MSP430G2553 chip. The FDC2214 capacitive sensor detection unit 1 is connected to the MSP430G2553 main control unit 5, and communicates through the IIC protocol. The human-computer interaction unit 7 is connected to the MSP430G2553, and communicates through the IIC protocol. Figure 2 As shown, the power pins VDD and GND of the FDC2214 chip are connected to the power supply unit, the IIC pins SCL and SDA of the FDC2214 chip are connected to the two pins P1.5 and P2.0 of the MSP430G2553 chip, and the MSP430G2553 reads the conversion result of the FDC2214 chip through the IO simulation IIC protocol, and sets a pull-up resistor with a resistance of 4.7k ohms on the SCL pin and the SDA pin. The measured capacitor, that is, Figure 2 The gray part in the circuit is connected to one end of the LC oscillator circuit to form an LC oscillator circuit with variable capacitance, thereby generating an oscillation frequency f with variable oscillation frequency. S The LC circuit is connected to the four channels of the chip respectively. The oscillation frequency f generated by the LC circuit S After being converted into digital quantity by FDC2214 chip. Figure 3 As shown, the present invention starts working after initializing MSP430G2553 and FDC2214, and selects the working mode by pressing a button, performs gesture detection and recognition in the corresponding working mode and displays the recognition result. Figure 4 As shown, the capacitors 1, 2, 3, and 4 to be measured are four copper sheets of different areas, and the capacitors to be measured are arranged in the shape of five fingers in the device.

[0024] The specific implementation steps of the embodiment are as follows:

[0025] 1) MSP430G2553 master control unit

[0026] The MSP430G2553 main control unit 5 uses the MSP430G2553 chip 3 as the core processor. Around the chip are a power supply circuit 2, a clock crystal circuit 4, and a reset circuit 6 to ensure the normal operation of the processor. The power supply circuit uses the LM1117-3.3 voltage regulator chip with a large input voltage range. At the same time, ground filter capacitors C1, C2, and C3 are set at the input terminal VBAT and the output terminal 3V3 of the voltage regulator chip, so as to output a stable 3.3V voltage for the entire gesture recognition device. The clock circuit includes a 32.768kHz low-speed crystal oscillator circuit for use when the main control unit 5 is dormant. The two ends of the crystal oscillator are respectively connected to the pins P2.6 and P2.7 of the MSP430G2553 chip. In order to ensure that the system can return to normal when it is abnormal, a classic manual button RC reset circuit 6 is set. When the button is pressed, a low-level pulse will be generated at the RESET pin of the MSP430G2553 chip 3 to force the main control unit to reset.

[0027] 2) FDC2214 capacitive sensor detection unit

[0028] like Figure 2 As shown, a 0.1uF and a 1uF ground capacitor filter are connected in parallel between the VDD and GND pins of the FDC2214 chip. At the same time, VDD is connected to the 3.3V output of the power supply unit, and GND is connected to the ground of the power supply unit. The IIC pins SCL and SDA of the FDC2214 chip are connected to the pins SCL and SDA of the IIC peripheral of the MSP430G2553 chip respectively. An 18uH inductor and a 33pF capacitor are connected to the input of the four measurement channels of the FDC2214 chip to form an LC oscillation circuit. The measured capacitors 1, 2, 3, 4 ( Figure 2 The middle gray part) is connected to the LC circuit, which will generate an oscillation frequency f that changes with the capacitance value being measured. S The FDC2214 chip uses a high-precision 40MHz reference frequency from an external source to measure the oscillation frequency f S , and converted into digital quantity for storage, MSP430G2553 reads the measurement results through SCL and SDA pins using IIC protocol.

[0029] 3) Human-computer interaction unit

[0030] The human-computer interaction unit 7 includes an OLED display screen and a key circuit. This example uses a low-power, small-volume OLED display screen to connect to the main control chip 3, and communicates through IO simulation IIC protocol to achieve multi-functional display. One end of the key KEY1 of the key circuit is connected to the pin P2.5 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY2 is connected to the pin P2.4 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY3 is connected to the pin P2.3 of the MSP430G2553 chip, and the other end is connected to GND. The main control chip 3 determines whether the key is pressed by detecting the level of the pin to which the key is connected, and executes the program corresponding to the key.

[0031] 4) System program flow

[0032] After the gesture recognition device is turned on, the main control chip and sensor FDC2214 are initialized first. After the initialization is completed, the main page is displayed on the OLED display screen, prompting the user to use the button to select to start recognition or enter the training mode for training. If the user presses button 1, it enters mode 1 and recognizes five gestures, namely gestures 1, 2, 3, 4, and 5. If the user presses button 2, it enters mode 2 and recognizes three gestures, namely rock, scissors, and cloth. If the user presses button 3, it enters the training mode, which is set to adapt to people with different hand shapes. After entering the training mode, the OLED display screen prompts the user to use the button to select the mode again. If the user presses button 1, it enters the training mode of mode 1; if the user presses button 2, it enters the training mode of mode 2. After entering the training mode, the trainee is required to make the five gestures of mode 1 or the three gestures of mode 2 in sequence. After the trainee makes the gesture, the main control unit will learn and enter the trainee's hand shape data for storage. After the training is completed, the gesture recognition program will be directly entered. After the user makes a gesture, the main control unit will compare and judge the capacitance value of the measured capacitor and recognize the gesture.

Claims

1. A gesture recognition device based on FDC2214, characterized in that: It includes MSP430G2553 main control unit, FDC2214 capacitive sensor detection unit and human-computer interaction unit; The MSP430G2553 main control unit includes an MSP430G2553 chip, a power supply circuit, a clock crystal oscillator circuit and a reset circuit; The power supply circuit uses the LM1117-3.3 voltage regulator chip. The input and output ends of the LM1117-3.3 voltage regulator chip are respectively provided with ground filter capacitors to provide a stable 3.3V voltage for the entire device; the clock crystal oscillator circuit includes a 32.768kHz low-speed crystal oscillator circuit, and the two ends of the crystal oscillator are respectively connected to the pins P2.6 and P2.7 of the MSP430G2553 chip; the reset circuit uses a manual RC reset circuit, and the RC reset circuit is connected to the pin TEST of the MSP430G2553 chip; The FDC2214 capacitance sensor detection unit includes an FDC2214 chip and a capacitor to be measured, and the capacitor to be measured is four copper sheets of different areas; The IN0A terminal of the input channel 0 of the FDC2214 chip is connected to one end of the inductor L1 and one end of the capacitor C9. The IN0B terminal of the input channel 0 is connected to the other end of the inductor L1 and the capacitor C9. The inductor L1 and the capacitor C9 form an LC oscillation circuit. One end of the measured capacitor CSensor1 is connected to one end of the inductor L1, and the other end of the measured capacitor CSensor1 is connected to GND. The IN1A terminal of the input channel 1 is connected to one end of the inductor L2 and one end of the capacitor C11. The IN1B terminal of the input channel 1 is connected to the other end of the inductor L2 and the capacitor C11. The inductor L2 and the capacitor C11 form an LC oscillation circuit. One end of the measured capacitor CSensor2 is connected to one end of the inductor L2, and the other end of the measured capacitor CSensor2 is connected to GND. The IN1A terminal of the input channel 1 is connected to one end of the inductor L2 and one end of the capacitor C11. The IN1B terminal of the input channel 1 is connected to the other end of the inductor L2 and the capacitor C11. The inductor L2 and the capacitor C11 form an LC oscillation circuit. One end of the measured capacitor CSensor2 is connected to one end of the inductor L2, and the other end of the measured capacitor CSensor2 is connected to GND. The IN3A terminal of input channel 4 is connected to one end of inductor L3 and one end of capacitor C12, the IN3B terminal of input channel 3 is connected to the other end of inductor L3 and capacitor C12, the inductor L3 and capacitor C12 form an LC oscillation circuit, one end of measured capacitor CSensor3 is connected to one end of inductor L3, and the other end of measured capacitor CSensor3 is connected to GND; the IN4A terminal of input channel 4 is connected to one end of inductor L4 and one end of capacitor C9, the IN4B terminal of input channel 4 is connected to the other end of inductor L4 and capacitor C13, the inductor L4 and capacitor C13 form an LC oscillation circuit, one end of measured capacitor CSensor4 is connected to one end of inductor L4, and the other end of measured capacitor CSensor4 is connected to GND; the LC circuit composed of inductor and capacitor will generate an oscillation frequency f S ; The VDD pin of the FDC2214 chip is connected to the output end of the power supply circuit, and the ADDR, GND, and PAD pins are connected to GND. The CLKIN of the FDC2214 chip is connected to the output end of an external 40Mhz high-precision crystal oscillator through a series 0 ohm resistor. The 40Mhz high-precision crystal oscillator provides a reference frequency f for the FDC2214 chip. REF ; The FDC2214 chip will measure the oscillation frequency f S Convert to digital quantity, and connect the IIC pins SCL and SDA of FDC2214 chip to the two pins P1.5 and P2.0 of MSP430G2553 chip. MSP430G2553 reads the conversion result of FDC2214 chip through IO simulation IIC protocol. The human-computer interaction unit includes an OLED display screen and a key circuit; the OLED_SCL and OLED_SDA pins of the OLED display screen are respectively connected to the pins P2.1 and P2.2 of the MSP430G2553 chip, and the multi-function display is realized by communicating through the IO simulation IIC protocol, the pin 1 of the OLED display screen is connected to the output end of the power supply circuit, and the pin 2 is connected to GND; one end of the key KEY1 of the key circuit is connected to the pin P2.5 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY2 is connected to the pin P2.4 of the MSP430G2553 chip, and the other end is connected to GND; one end of the key KEY3 is connected to the pin P2.3 of the MSP430G2553 chip, and the other end is connected to GND; After the gesture recognition device is turned on, the main control chip and sensor FDC2214 are initialized first. After the initialization is completed, the main page is displayed on the OLED display screen, prompting the user to use the button to select to start recognition or enter the training mode for training; if the user presses button 1, it enters mode 1 to recognize 5 gestures, namely gestures 1, 2, 3, 4, and 5; if the user presses button 2, it enters mode 2 to recognize 3 gestures, namely rock, scissors, and cloth; if the user presses button 3, it enters training mode. The training mode is set to adapt to people with different hand shapes. After entering the training mode, the OLED display screen prompts the user to use the button to select the mode again. If the user presses button 1, it enters the training mode of mode 1; if the user presses button 2, it enters the training mode of mode 2; after entering the training mode, the trainee is required to make five gestures of mode 1 or three gestures of mode 2 in sequence. After the trainee makes the gesture, the main control unit will learn and enter the trainee's hand shape data for storage; after completing the training, it will directly enter the gesture recognition program. After the user makes the gesture, the main control unit will compare and judge the capacitance value of the measured capacitor and recognize the gesture.

2. A gesture recognition device based on FDC2214 as claimed in claim 1, characterized in that: Four copper pieces of different areas are arranged in the shape of five fingers spread out.

Images