Treadmill control system for facial expression application

Abstract

The invention relates to the technical field of fitness equipment, in particular to a treadmill control system for facial expression application, which comprises a CPU control module, a mobile phone terminal, an image analysis processing module, a motor driving system module, an EMC module, a power supply, a wireless module and a Bluetooth module, and is characterized in that the mobile phone terminal comprises a mobile phone APP and a mobile phone camera; a signal input port of the CPU control module is connected with a signal output port of the image analyzing and processing module, and a mobile signal port of the CPU control module is connected and paired with a mobile phone end through a Bluetooth module or a wireless module and transmits data information. Compared with the prior art, by integrating two control modes of facial expression recognition and manual operation, a user can freely switch according to scenes, the system supports real-time state synchronization, resource coordination is optimized through a unified main control module, electromagnetic interference is reduced through the design of an EMC module and a double-grounding power supply, and the system is simple in structure and convenient to use. And the stable operation of the motor driving system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of fitness equipment, and more specifically, to a treadmill control system applied with facial expressions. Background Art

[0002] The control systems of traditional treadmills are mostly limited to a single manual operation mode, lacking intelligence and interactivity. The user experience is relatively monotonous, the control method is single, unable to meet the personalized and interesting needs of users. The user authentication means are backward, with insufficient security and convenience. The hardware coordination is poor, and the data interaction ability is limited. There is no real-time sharing of motion data, multi-device linkage, and dynamic feedback mechanism, and at the same time, there is a lack of intelligent decision-making ability, which restricts the intelligent upgrade of fitness equipment and the improvement of user experience. Summary of the Invention

[0003] In order to overcome the deficiencies of the prior art, the present invention provides a treadmill control system applied with facial expressions. Through the innovative integration of multi-modal interaction, intelligent algorithms, and modular hardware, it solves the problems of single control method, insufficient security, and poor scalability of traditional treadmills, and has significant advantages in terms of user experience, security, and technical compatibility.

[0004] To achieve the above object, a treadmill control system applied with facial expressions is designed, including a CPU control module, a mobile phone terminal, an image analysis and processing module, a motor drive system module, an EMC module, a power supply, a wireless module, and a Bluetooth module. The mobile phone terminal includes a mobile phone APP and a mobile phone camera. The signal input port of the CPU control module is connected to the signal output port of the image analysis and processing module. The control signal output port of the CPU control module is connected to the control signal input port of the motor drive system module. The power input port of the motor drive system module is connected to the power filter output port of the EMC module. The power filter input port of the EMC module is connected to the output port of the power supply. The mobile signal port of the CPU control module and the mobile phone terminal are connected and paired through the Bluetooth module or the wireless module to transmit data information.

[0005] It includes the following steps: Step 1, the system is initialized to detect whether there is an access request at the external port. If not, it enters the sleep state after waiting for a period of time. If there is, it performs signal access and detects whether the fitness program on the device side is the main interface program. If so, it calls the instruction, and the system enters different mode selections of "Virtual Reality Running Mode", "Target Mode", and "Quick Start Mode". After obtaining the manually selected mode and detecting again that it is not the main interface, it continues with Step 2; Step 2, it is judged whether it is a running mode program. If it is a running mode program, it calls the instruction, and the system enters the exercise target and exercise program interface, and calls the instructions for manually setting the time, distance, and calories to be consumed. If it is not a running mode program, it continues with Step 3; Step 3, it is judged whether there is access to wireless local area network or Bluetooth transmission data. If there is an access request, it performs signal access, obtains the user mode selection of the mobile phone APP, and judges whether the expression recognition mode is selected. If it is the expression recognition mode, it calls the image acquisition of the mobile phone camera. The image analysis and processing module detects and extracts features from the face image collected by the mobile phone camera to judge the user state. The CPU control module judges whether the treadmill continues to operate or stops according to the feedback of the image analysis and processing module, and controls the motor drive system module to control the actions of the treadmill; if it is not the expression recognition mode, it starts the manual mode, obtains the "Start", "Pause", "Increase Speed", and "Decrease Speed" signals of the manual operation in the mobile phone APP, and controls the speed and slope of the treadmill. If there is no access request, it continues with Step 4; Step 4, it is judged whether it is a sports data program. If it is a sports data program, it calls the display function instruction to display the historical sports parameter data and can share it. If it is not a sports data program, it continues with Step 5; Step 5, it is judged whether it is a system setting. If it is a system setting, it calls the instruction, and the system enters the setting mode interface to obtain the set brightness and language switching parameters. If it is not a system setting, it returns to Step 1.

[0006] The CPU control module is set inside the treadmill control platform. The treadmill control platform is installed at the front end of the treadmill, and there is a handle paddle on the treadmill control platform.

[0007] The working voltage of the power supply is 220V, and the frequency is 50Hz or 60Hz.

[0008] The model of the wireless module is MTK7981B, and the model of the Bluetooth module is CC2640R2F.

[0009] The two signal output ends of the motor drive system module are respectively connected to the drive end of the AC motor and the drive end of the lifting motor.

[0010] The described image analysis and processing module uses LBP texture feature extraction to compare the gray values of neighboring pixels with each pixel in the image. If the gray value of a neighboring pixel is greater than or equal to that of the central pixel, it is marked as 1; otherwise, it is marked as 0. The formula is , and the comparison results are arranged in a clockwise order to form a binary number, which is then converted from an 8-bit binary number to a decimal number as the LBP value of the pixel. The formula is .

[0011] Compared with the prior art, the present invention integrates two control modes of facial expression recognition and manual operation. Users can freely switch according to the scenario. The system supports real-time status synchronization, optimizes resource coordination through a unified main control module, sets an EMC module and a dual-grounding power supply design to reduce electromagnetic interference, and ensures the stable operation of the motor drive system. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a module connection diagram.

[0013] Figure 2 is a program flow chart of the main control system module of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] The following further describes the present invention with reference to the drawings.

[0015] As Figure 1 shown, a treadmill control system for facial expression applications includes a CPU control module, a mobile phone terminal, an image analysis and processing module, a motor drive system module, an EMC module, a power supply, a wireless module, and a Bluetooth module. The mobile phone terminal 4 includes a mobile phone APP and a mobile phone camera. The mobile phone camera 4.2 is responsible for capturing the facial expressions of users in real time.

[0016] The signal input port of the CPU control module 2 is connected to the signal output port of the image analysis and processing module 3. The control signal output port of the CPU control module 2 is connected to the control signal input port of the motor drive system module 5. The power input port of the motor drive system module 5 is connected to the power filter output port of the EMC module 6. The power filter input port of the EMC module 6 is connected to the output port of the power supply 7.

[0017] The mobile signal port of the CPU control module 2 is connected and paired with the mobile phone terminal 4 through the Bluetooth module 14 or the wireless module 13 to transmit data information, for establishing a Bluetooth or wireless connection with a smart mobile device. The model of the wireless module 13 is MTK7981B, and the model of the Bluetooth module 14 is CC2640R2F.

[0018] The power supply terminal of the CPU control module 2 is connected to the power output port of the 12V power supply module 10. The 12V power supply module 10 supplies power to the CPU control module 2. The CPU control module 2 is provided with a wireless charging interface 11 and a USB interface 12, and the wireless charging interface 11 and the USB interface 12 are used to provide mobile phone charging.

[0019] The CPU control module 2 is used to receive the operation instructions of the manual control module and the control instructions of the face expression recognition module, and control the running state of the treadmill according to the instructions. The image analysis and processing module 3 is responsible for face detection and positioning, expression classification, and instruction generation, and is used to analyze the face expressions captured by the camera and convert the analysis results into control instructions. The information collected through the mobile phone camera is transmitted to the CPU control module 2 through the Bluetooth module 14 or the wireless module 13.

[0020] The CPU control module 2 is arranged inside the treadmill control platform. The treadmill control platform is installed at the front end of the treadmill. A handle paddle can be provided on the treadmill control platform, which can be used to receive external operation instructions.

[0021] The working voltage of the power supply 7 is 220V, and the frequency is 50Hz or 60Hz.

[0022] The two signal output ends of the motor drive system module 5 are respectively connected to the drive end of the AC motor 8 and the drive end of the lift motor 9.

[0023] Such as Figure 2As shown in the figure, the present invention includes the following steps: Step 1, the system is initialized to detect whether there is an access request at the external port. If not, it enters the sleep state after waiting for a period of time. If there is, signal access is performed, and it is detected whether the fitness program on the device side is the main interface program. If so, an instruction is called, and the system enters different mode selections of "real - scene running mode", "target mode", and "quick start mode". After obtaining the manually selected mode and detecting again that it is not the main interface, proceed to Step 2; Step 2, determine whether it is a running - mode program. If it is a running - mode program, an instruction is called, and the system enters the exercise target and exercise program interface, and calls the instructions for manually setting the time, distance, and calories to be consumed. If it is not a running - mode program, proceed to Step 3; Step 3, determine whether there is access to wireless local - area network or Bluetooth - transmitted data. If there is an access request, signal access is performed, and the user - mode selection of the mobile phone APP is obtained. Determine whether the expression recognition mode is selected. If it is the expression recognition mode, call the mobile - phone camera image acquisition. The image analysis and processing module detects and extracts features from the face image collected by the mobile - phone camera to judge the user's state. The CPU control module judges whether the treadmill continues to operate or stops according to the feedback of the image analysis and processing module, and controls the motor drive system module to control the movement of the treadmill; if it is not the expression recognition mode, start the manual mode, obtain the "start", "pause", "increase speed", and "decrease speed" signals of the manual operation in the mobile phone APP, and control the speed and slope of the treadmill. If there is no access request, proceed to Step 4; Step 4, determine whether it is a sports - data program. If it is a sports - data program, call the display - function instruction to display the historical sports - parameter data, and it can be shared. If it is not a sports - data program, proceed to Step 5; Step 5, determine whether it is a system setting. If it is a system setting, call an instruction, and the system enters the setting - mode interface to obtain the set brightness and language - switching parameters. If it is not a system setting, return to Step 1.

[0024] When the user selects the "manual" mode, the operation of the user is directly received as a command through the buttons in the mobile - phone APP on the mobile - phone side or the handle paddles of the treadmill control platform. The CPU control module 2 controls the running state of the treadmill according to the received instruction.

[0025] When the user selects the treadmill expression recognition mode, the mobile phone camera is used to capture the user's facial expression in real time, and the information is transmitted into the image analysis and processing module 3 through the Bluetooth module 14 or the wireless module 13. The image analysis and processing module 3 performs grayscale processing on the image collected by the camera through preprocessing. The face detection and feature localization module determines the face position using the face detection algorithm in the OpenCV library based on the results of the preprocessing. By synthesizing the pixel values of each data row into a 48*48 expression map, a training model is generated, which is divided into a training set and a validation set. Each picture is classified into which type of expression and stored in a csv file. The training set and test set are marked, and the image dataset is loaded by rewriting. Subsequently, according to the data format accepted by the api of nn.Conv2d in the convolutional neural network, network model B is used to build a network model, and RRelu is used as the activation function to build a convolutional neural network model. After processing the information, the instruction generation module converts the recognized expression into a corresponding control command and sends the control command through the selected network protocol. The image analysis and processing module 3 uses LBP texture feature extraction to compare the gray values of the neighboring pixels of each pixel in the image. If the gray value of the neighboring pixel is greater than or equal to the central pixel, it is marked as 1, otherwise it is marked as 0. The formula is , the comparison results are arranged in a clockwise order to form a binary number, and then the 8-bit binary number is converted into a decimal number as the LBP value of the pixel. The formula is , at the same time, visual algorithms are used to detect and track key points on the face such as eyes, mouth, eyebrows, etc. Based on the position coordinates of the key points in the geometric features, the formula for the feature vector F =

X1, Y1, X2, Y2.....Xn, Yn

[0026] The purpose of the present invention is to provide an intelligent treadmill control system based on facial expression analysis. This system can realize manual control or control the running state of the treadmill through facial expressions, etc. The treadmill is controlled by various methods, adding interactivity, fun and a sense of technological intelligence to the treadmill, and improving the user experience.

Claims

1. A treadmill control system for facial expression applications, comprising a CPU control module, a mobile phone terminal, an image analysis and processing module, a motor drive system module, an EMC module, a power supply, a wireless module, and a Bluetooth module, characterized in that: The mobile device (4) includes a mobile APP and a mobile camera; The signal input port of the CPU control module (2) is connected to the signal output port of the image analysis and processing module (3). The control signal output port of the CPU control module (2) is connected to the control signal input port of the motor drive system module (5). The power input port of the motor drive system module (5) is connected to the power filter output port of the EMC module (6). The power filter input port of the EMC module (6) is connected to the output port of the power supply (7). The mobile signal port of the CPU control module (2) and the mobile device (4) are connected and paired through a Bluetooth module (14) or a wireless module (13) to transmit data information; It includes the following steps: Step 1, the system is initialized to detect whether there is an access request at the external port. If not, it enters the sleep state after waiting for a period of time. If there is, the signal is accessed, and it is detected whether the fitness program on the device side is the main interface program. If so, an instruction is called, and the system enters different mode selections of "real - scene running mode", "target mode", and "quick start mode". After obtaining the manually selected mode and detecting again that it is not the main interface, continue to Step 2; Step 2, determine whether it is a running mode program. If it is a running mode program, an instruction is called, and the system enters the exercise target and exercise program interface, and calls the instructions for manually setting time, distance, and calories to be consumed. If it is not a running mode program, continue to Step 3; Step 3, determine whether there is an access to wireless local area network or Bluetooth data transmission. If there is an access request, the signal is accessed, and the user mode selection of the mobile APP is obtained. Determine whether the expression recognition mode is selected. If it is the expression recognition mode, the mobile camera image is collected. The image analysis and processing module detects and extracts features from the face image collected by the mobile camera to judge the user state. The CPU control module judges whether the treadmill continues to operate or stops according to the feedback of the image analysis and processing module, and controls the motor drive system module to control the actions of the treadmill; If it is not the expression recognition mode, the manual mode is started, and the "start", "pause", "increase speed", and "decrease speed" signals of the manual operation in the mobile APP are obtained to control the speed and slope of the treadmill. If there is no access request, continue to Step 4; Step 4, determine whether it is an exercise data program. If it is an exercise data program, the display function instruction is called to display the historical exercise parameter data, and it can be shared. If it is not an exercise data program, continue to Step 5; Step 5, determine whether it is a system setting. If it is a system setting, an instruction is called, and the system enters the setting mode interface to obtain the set brightness and language switching parameters. If it is not a system setting, return to Step 1.

2. The treadmill control system for facial expression application according to claim 1, characterized in that: The CPU control module (2) is arranged inside the treadmill control platform. The treadmill control platform is installed at the front end of the treadmill, and there is a handle paddle on the treadmill control platform.

3. The treadmill control system for facial expression application according to claim 1, characterized in that: The power supply terminal of the CPU control module (2) is connected to the power output port of the 12V power supply module (10). The CPU control module (2) is provided with a wireless charging interface (11) and a USB interface (12).

4. The treadmill control system for facial expression application according to claim 1, characterized in that: The operating voltage of the power supply (7) is 220V, and the frequency is 50Hz or 60Hz.

5. The treadmill control system for facial expression application according to claim 1, wherein: The model of the wireless module (13) is MTK7981B, and the model of the Bluetooth module (14) is CC2640R2F.

6. The treadmill control system for facial expression application according to claim 1, wherein: The two signal output terminals of the motor drive system module (5) are respectively connected to the drive end of the AC motor (8) and the drive end of the lifting motor (9).

7. The treadmill control system for human face expression application according to claim 1, wherein: The described image analysis and processing module (3) uses the LBP texture feature extraction method to compare the gray values of neighboring pixels for each pixel in the image. If the gray value of the neighboring pixel is greater than or equal to the central pixel, it is marked as 1; otherwise, it is marked as 0. The formula is , arrange the comparison results in a clockwise order to form a binary number, and then convert the 8-bit binary number into a decimal number as the LBP value of the pixel. The formula is .

Images