A material counting system based on vision recognition

Abstract

This invention provides a visual recognition-based material counting system, comprising a material counting device, a service management module, a human-machine interaction module, a network control module, a counting mechanism, and a visual recognition module. The service management module sends outbound tasks to the human-machine interaction module and control commands to the network control module. The human-machine interaction module operates the counting mechanism. The network control module determines the real-time material quantity based on the real-time distance of the material conveyor belt movement and the material spacing sent by the visual recognition module. The visual recognition module obtains the material spacing from a processed image obtained after processing the original image. The counting mechanism drives the external material counting device to transport materials and provides feedback on its own movement steps. This visual recognition-based material counting system features high intelligence, accurate counting, and minimal human intervention, and can be widely applied in the field of material inventory.

Description

Technical Field

[0001] This invention relates to visual recognition technology, and in particular to a material counting system based on visual recognition. Background Technology

[0002] Currently, all electronic material counting devices on the market require manual intervention. Specifically, the receipt and dispatch of materials, as well as the modification of parameters such as the material spacing during the operation of the electronic material counting device, all require manual intervention. Moreover, the actual material spacing is judged by manual visual inspection or experience, which has a relatively large operational error. Sometimes, incorrect judgments can lead to incorrect counting, which affects the efficiency of material management.

[0003] Therefore, it is evident that, in the current technology, there is no intelligent material counting system with high counting accuracy and minimal human intervention. Summary of the Invention

[0004] In view of this, the main objective of the present invention is to provide a material counting system based on vision recognition that has high counting accuracy, requires less human intervention, and is intelligent.

[0005] To achieve the above objectives, the technical solution proposed by this invention is as follows:

[0006] A material counting system based on vision recognition includes a material counting device. The material counting system further includes: a service management module, a human-computer interaction module, a network control module, a counting mechanism, a vision recognition module, a display module, and a power supply module; wherein,

[0007] The service management module is used to send outbound tasks to the human-computer interaction module based on the work tasks, and at the same time, send control commands to the network control module.

[0008] The human-machine interaction module is used to operate the counting mechanism according to the outbound task sent by the service management module: after adjusting the width of the conveyor track of the counting mechanism, it starts counting the materials and installs the tape-type materials on the running conveyor track.

[0009] The network control module is used to send counting instructions to the counting mechanism based on the control instructions sent by the service management module and the number of movement steps sent by the counting mechanism, and to obtain the real-time distance of material tape movement; to determine the real-time material quantity based on the real-time distance of material tape movement and the material spacing sent by the vision recognition module; and to send display information to the display module and the real-time material tape movement distance to the vision recognition module.

[0010] The visual recognition module is used to acquire original images of the material tape being transported based on the real-time material tape movement distance sent by the network control module, preprocess the original images to obtain processed images, and send the material spacing obtained from the processed images to the network control module.

[0011] The counting mechanism is used to send counting commands from the network control module to drive the operation of external material counting equipment and send the movement steps to the network control module.

[0012] The display module displays the quantity of task materials, material spacing, and system operating status based on the display information sent by the network control module.

[0013] The power supply module is used to supply power to the network control module, counting mechanism, vision recognition module, and display module.

[0014] In summary, in the visual recognition-based material counting system of this invention, the network control module operates according to the control commands from the service management module and the operations from the human-computer interaction module. The counting mechanism installed on the external material counting device operates under the network control module and sends the stepper motor's movement steps to the network control module. The stepper motor's movement steps, as a negative feedback signal, together with the control commands sent by the service management module, serve as the input for the network control module to control the counting mechanism, making the network control module's control of the counting mechanism more precise. Simultaneously, the network control module determines the real-time material quantity based on the material spacing and real-time movement distance of the material tape sent by the visual recognition module, improving counting accuracy and intelligence while significantly reducing human intervention. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall structure of the material counting system based on vision recognition described in this invention.

[0016] Figure 2 This is a front view of the overall structural composition of the counting mechanism described in this invention.

[0017] Figure 3 This is a top-down right rear view of the counting mechanism described in this invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] Figure 1 This is a schematic diagram of the overall structure of the visual recognition-based material counting system described in this invention. Figure 1 As shown, the material counting system based on vision recognition according to the present invention includes a material counting device, and further includes: a service management module 1, a human-computer interaction module 2, a network control module 3, a counting mechanism 4, a vision recognition module 5, a display module 6, and a power supply module 7; wherein,

[0020] Service management module 1 is used to send outbound tasks to human-computer interaction module 2 according to work tasks, and at the same time, send control commands to network control module 3.

[0021] The human-machine interaction module 2 is used to operate the counting mechanism 4 according to the outbound task sent by the service management module 1: after adjusting the width of the conveyor track of the counting mechanism 4, it starts counting the materials and installs the tape-type materials on the running conveyor track.

[0022] In practical applications, when the visual recognition-based material counting system described in this invention is operating in non-intelligent mode, operators can manually input the material quantity and spacing through the human-computer interaction module 2, enabling the system to perform offline counting. When the system is operating in intelligent mode, operators are not allowed to modify the relevant parameters through the human-computer interaction module 2.

[0023] The network control module 3 is used to send a counting command to the counting mechanism 4 based on the control command sent by the service management module 1 and the number of movement steps sent by the counting mechanism 4, and to obtain the real-time distance of material tape movement; to determine the real-time material quantity based on the real-time distance of material tape movement and the material spacing sent by the vision recognition module 5; and to send display information to the display module 6 and the real-time material tape movement distance to the vision recognition module 5.

[0024] The visual recognition module 5 is used to collect the original image of the material tape being transported based on the real-time material tape movement distance sent by the network control module 3, and to preprocess the original image to obtain a processed image; and to send the material spacing obtained based on the processed image to the network control module 3.

[0025] The counting mechanism 4 is used to send counting commands from the network control module 3, drive the external material counting equipment to operate, and send the movement steps to the network control module 3.

[0026] Display module 6 displays the quantity of task materials, material spacing, and system working status based on the display information sent by network control module 3.

[0027] In practical applications, the system's working status is the start-up, running, and stop status of the visual recognition-based material counting system.

[0028] Power module 7 is used to supply power to network control module 3, counting mechanism 4, vision recognition module 5, and display module 6.

[0029] In practical applications, the visual recognition-based material counting system described in this invention requires three power supplies: 12V, 5V, and 3.3V. Therefore, a 12V adapter or a 12V lithium battery pack is used as the power source. To meet all the needs of the visual recognition-based material counting system, the power module 7 also converts the 12V voltage to 5V and 3.3V. Converting 12V to 5V and 3.3V is existing technology and will not be elaborated upon here.

[0030] In summary, in the visual recognition-based material counting system of this invention, the network control module operates according to the control commands from the service management module and the operations from the human-computer interaction module. The counting mechanism installed on the external material counting device operates under the network control module and sends the stepper motor's movement steps to the network control module. The stepper motor's movement steps serve as a negative feedback signal, which, together with the control commands sent by the service management module, constitutes the input for the network control module to control the counting mechanism, making the network control module's control of the counting mechanism more precise. Simultaneously, the network control module determines the real-time material quantity based on the material spacing and real-time movement distance of the material tape sent by the visual recognition module, improving counting accuracy and intelligence while significantly reducing human intervention.

[0031] In this invention, the network control module 3 includes: a counting unit and a motor control unit; wherein,

[0032] The counting unit is used to obtain the rotational angular velocity of the stepper motor based on the number of steps the stepper motor moves sent by the counting mechanism 4; after converting the rotational angular velocity of the stepper motor into the running speed of the material tape, it obtains the real-time material quantity corresponding to the real-time distance the belt moves within a certain time based on the material spacing sent by the vision recognition module 5.

[0033] The motor control unit is used to send the generated counting command to the counting mechanism 4 based on the control command sent by the service management module 1 and the number of steps the stepper motor moves sent by the counting mechanism 4.

[0034] In practical applications, the stepper motor movement count sent by the counting mechanism 4 is used as a feedback signal to the motor control unit, so that the counting mechanism 4 and the motor control unit form a closed loop with negative feedback, thereby improving the control accuracy of the motor control unit over the stepper motor in the counting mechanism 4.

[0035] Figure 2 This is a front view of the overall structural composition of the counting mechanism described in this invention. Figure 3 This is a top-down, right-rear view of the counting mechanism described in this invention. Figure 2 , Figure 3As shown, the counting mechanism 4 of the present invention is mounted on the fixed frame 40 of the external material counting equipment, and includes: a stepper motor 43, a counting ratchet 41, a photoelectric encoder 42, three synchronous pulleys 44, and a synchronous belt 45; wherein, the three synchronous pulleys 44 are respectively mounted on the rotating shaft of the stepper motor 43, the rotating shaft of the counting ratchet 41, and the output shaft of the photoelectric encoder 42, and the synchronous belt 45 is mounted on the three synchronous pulleys 44;

[0036] The stepper motor 43 operates according to the counting command sent by the motor control unit, and transmits the rotational driving force to the synchronous pulley 44 corresponding to the counting ratchet 41 and the synchronous pulley 44 corresponding to the photoelectric encoder 42 in sequence through the synchronous pulley 44 and the synchronous belt 45 installed on its output shaft.

[0037] The counting ratchet 41 is used to rotate under the drive of its corresponding synchronous pulley 44 and synchronous belt 45, thereby driving the material tape of the external material counting device to operate.

[0038] In practical applications, since the edges of the tape-type tray material have fixed-specification, equally spaced holes, the teeth of the counting ratchet mesh with the edge holes, driving the material tape forward or backward.

[0039] The photoelectric encoder 42 is used to rotate under the drive of its corresponding synchronous pulley 44 and synchronous belt 45, and sends the number of steps of the stepper motor 43 obtained by converting its own rotation angle to the motor control unit.

[0040] In practical applications, the aforementioned counting mechanism 4 and the motor control unit form a closed loop with negative feedback. Specifically, the motor control unit, the stepper motor 43, and the photoelectric encoder 42 form a closed loop with negative feedback to prevent inaccurate material tape movement distance caused by stepper motor stalling, step loss, or overshoot.

[0041] In this invention, the visual recognition module 5 includes a camera, a coaxial light source, a preprocessing unit, a first distance acquisition unit, a second distance acquisition unit, and a comparison unit; wherein,

[0042] The camera is used to acquire raw images of the material tape under the illumination provided by the coaxial light source, and send the raw images of the material tape to the preprocessing unit.

[0043] In practical applications, the original image of the material tape changes as the real-time material tape moves a certain distance.

[0044] A coaxial light source is used to provide sufficient illumination for the camera.

[0045] The preprocessing unit performs differential binarization on the original image of the material tape from the camera, then performs erosion filtering to obtain a processed image; the processed image is then sent to the recognition unit.

[0046] The first spacing acquisition unit is used to identify the number of rectangles in the processed image using a rectangle recognition method; and to obtain the first spacing of the material based on the real-time material tape movement distance corresponding to the number of rectangles sent by the network control module 3. Real-time material conveyor travel distance / number of rectangles.

[0047] The second spacing acquisition unit is used to acquire the positioning of two adjacent materials in the processed image using a machine vision model, and to acquire the second spacing between the materials. ,in, , The coordinates of the center points of two adjacent materials.

[0048] In practical applications, machine vision models are existing technologies and will not be discussed further here.

[0049] Comparison unit, used to determine Validity: If valid, the first or second spacing is sent to the network control module 3 as the material spacing; if invalid, a warning is issued that the spacing identification is unavailable.

[0050] In this invention, the comparison unit in the visual recognition module 5 compares the first material spacing generated by the first spacing acquisition unit with the second material spacing generated by the second spacing acquisition unit. Only when the two are the same is the material spacing sent to the network control module 3. This greatly improves the accuracy of the material spacing and further enhances the accuracy of the visual recognition-based material counting system described in this invention.

[0051] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A material counting system based on vision recognition, comprising a material counting device, characterized in that, The material counting system further includes: a service management module, a human-machine interaction module, a network control module, a counting mechanism, a vision recognition module, a display module, and a power supply module; among which, The service management module is used to send outbound tasks to the human-computer interaction module based on the work tasks, and at the same time, send control commands to the network control module. The human-machine interaction module is used to operate the counting mechanism according to the outbound task sent by the service management module: after adjusting the width of the conveyor track of the counting mechanism, it starts counting the materials and installs the tape-type materials on the running conveyor track. The network control module is used to send counting instructions to the counting mechanism based on the control instructions sent by the service management module and the number of movement steps sent by the counting mechanism, and to obtain the real-time movement distance of the material tape; to determine the real-time material quantity based on the real-time movement distance of the material tape and the material spacing sent by the vision recognition module; and to send display information to the display module and the real-time movement distance of the material tape to the vision recognition module. The visual recognition module is used to acquire original images of the material tape being transported based on the real-time material tape movement distance sent by the network control module, preprocess the original images to obtain processed images, and send the material spacing obtained from the processed images to the network control module. The counting mechanism is used to drive the operation of external material counting equipment by sending counting commands from the network control module, and sends its own movement steps to the network control module. The display module displays the quantity of task materials, material spacing, and system operating status based on the display information sent by the network control module. The power supply module is used to power the network control module, the counting mechanism, the vision recognition module, and the display module; The visual recognition module includes a camera, a coaxial light source, a preprocessing unit, a first distance acquisition unit, a second distance acquisition unit, and a comparison unit; wherein, The camera is used to acquire raw images of the material tape under the illumination provided by the coaxial light source, and send the raw images of the material tape to the preprocessing unit. A coaxial light source is used to provide sufficient illumination for the camera; The preprocessing unit performs differential binarization on the original image of the material tape from the camera, then performs erosion filtering to obtain the processed image; and sends the processed image to the recognition unit. The first spacing acquisition unit is used to identify the number of rectangles in the processed image using a rectangle recognition method; and to acquire the first spacing of the material based on the identified number and the real-time material tape movement distance corresponding to the number of rectangles sent by the network control module. ; The second spacing acquisition unit is used to acquire the positioning of two adjacent materials in the processed image using a machine vision model, and to acquire the second spacing between the materials. ,in, , These are the coordinates of the center points of two adjacent materials; Comparison unit, used to determine Validity: If valid, the first or second spacing is sent to the network control module as the material spacing; if invalid, a warning is issued that the spacing identification is unavailable.

2. The material counting system based on visual recognition according to claim 1, characterized in that, The network control module includes: a counting unit and a motor control unit; wherein... The counting unit is used to obtain the rotational angular velocity of the stepper motor based on the number of steps the stepper motor moves sent by the counting mechanism; after converting the rotational angular velocity of the stepper motor into the running speed of the material tape, it obtains the real-time material quantity corresponding to the real-time distance the belt moves within a certain time based on the material spacing sent by the visual recognition module. The motor control unit is used to send the generated counting command to the counting mechanism based on the control command sent by the service management module and the number of steps the stepper motor moves sent by the counting mechanism.

3. The material counting system based on visual recognition according to claim 2, characterized in that, The counting mechanism is mounted on the fixed frame of the external material counting equipment and includes: a stepper motor, a counting ratchet, a photoelectric encoder, three synchronous pulleys, and a synchronous belt; wherein, the three synchronous pulleys are respectively mounted on the rotating shaft of the stepper motor, the rotating shaft of the counting ratchet, and the output shaft of the photoelectric encoder, and the synchronous belt is mounted on the three synchronous pulleys; The stepper motor operates according to the counting command sent by the motor control unit, and transmits the rotational driving force to the synchronous pulley corresponding to the counting ratchet and the synchronous pulley corresponding to the photoelectric encoder in sequence through the synchronous pulley and synchronous belt installed on its output shaft; The counting ratchet is used to rotate under the drive of its corresponding synchronous pulley and synchronous belt, thereby driving the material tape of the external material counting equipment to operate; The photoelectric encoder is used to rotate under the drive of its corresponding synchronous pulley and synchronous belt, and sends the number of steps of the stepper motor obtained by converting its own rotation angle to the motor control unit.

Images