A fast classification of a cake defect detection integrated device
Patent Information
- Application Number
- CN202310828745.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-07
- Publication Date
- 2026-10-27
- Estimated Expiration
- 2043-07-07
AI Technical Summary
[0005]为解决现有技术存在的对于单一固定丝饼采用丝饼的检测装置存在效率低、速度慢、相机图像采样重叠、缺陷丝饼分类的技术问题,本发明提供一种快速分类的丝饼缺陷检测一体化设备
[0038]与现有技术相比,本发明的有益效果至少包括:
Smart Images

Figure CN116990312B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of chemical fiber filament production technology, specifically relating to an integrated device for rapid classification and defect detection of filament filaments. Background Technology
[0002] Synthetic fiber is a man-made fiber characterized by its lightweight, durability, and ease of dyeing. Therefore, it is widely used in the textile industry in daily life and light industrial production to manufacture various garments, home textiles, and industrial products. A synthetic fiber cake is a form of synthetic fiber product. It is made by winding synthetic fiber onto a cylindrical paper support tube, forming a disc-shaped object. The diameter of a synthetic fiber cake is generally 300-450mm, and its weight is 8-15kg.
[0003] During the manufacturing process of chemical fiber yarn cakes, various defects can occur on the surface of the yarn cakes due to factors such as dust contamination during spinning and handling. Currently, there are 13 common defects in production, including tangled yarns, stiff yarns, oil stains, shiny yarns, double-strand yarns, oily yarns, waste yarns, fuzzy yarns, tangled yarns, poor forming, uneven yarns, fuzzy defects, and paper tube defects. These defects directly affect the quality of the yarn cakes and the quality control of subsequent textile processing.
[0004] Currently, domestic equipment for detecting defects in silk cakes mainly falls into two categories. The first type uses a conveyor belt combined with a rotary vision inspection mechanism to acquire images of the silk cake's side surface, as described in patent publication number CN 109142369 A. However, this type of structure suffers from drawbacks such as slow visual acquisition cycle time of the rotary mechanism, high precision requirements for the rotary mechanism, and complex execution actions. The second type involves setting up multiple inspection stations on the inspection conveyor belt to collect image information of the silk cake surface step-by-step, as described in patent publication number CN214894922U. However, this device is prone to issues such as rotational deviation of the silk cake disc angle during roller conveyor belt movement, leading to missed detections and overlapping detections of some silk cake surface images. Furthermore, it suffers from problems such as a large number of inspection stations on the conveyor belt, low equipment concentration, low utilization of sampling light sources, and large space threshold occupancy. Summary of the Invention
[0005] To address the technical problems of low efficiency, slow speed, overlapping camera image sampling, and defective silk cake classification in existing technologies that use silk cake detection devices for single fixed silk cakes, this invention provides an integrated device for rapid classification of silk cake defects.
[0006] The technical solution adopted in this invention is:
[0007] An integrated device for rapid classification and defect detection of silk cakes includes a silk cake surface pretreatment device, a silk cake defect detection device, a silk cake rotation limiting device, and a silk cake diversion device, which are sequentially arranged on the roller production line of the silk cake transmission device and controlled by an embedded PLC system. The silk cake defect detection device acquires an image of the silk cake surface and outputs it to an edge computing node. The edge computing node identifies and detects the image and outputs it to a PC-based host computer. The PC-based host computer performs real-time silk cake image display, defect statistics, and storage. Its key feature is:
[0008] The yarn cake defect detection device includes a first defect detection mechanism and a second defect detection mechanism. The first and second defect detection mechanisms are located on opposite sides of the roller conveyor line and are mirror images of each other relative to the longitudinal axis of the roller conveyor line. The first defect detection mechanism is located upstream of the second defect detection mechanism and performs image acquisition on the opposite side of the yarn cake. A yarn cake rotation limiting device is installed on the roller conveyor line, close to the upstream of the first and second defect detection mechanisms, to limit the rotation phase of the yarn cake relative to the roller conveyor line.
[0009] Furthermore, the silk cake rotation limiting device includes two parallel rows of rollers. The two rows of rollers are respectively connected to their respective three-phase asynchronous motors via a transmission chain. The control terminals of the three-phase asynchronous motors are connected to a PLC. When the PLC controls the two three-phase asynchronous motors to rotate in the same direction, the silk cake is transported forward. When the PLC controls the two three-phase asynchronous motors to rotate in opposite directions, the silk cake is rotated.
[0010] It also includes a limit pin mounted on the lifting cylinder, and a first infrared sensor for detecting when the yarn cake enters the yarn cake rotation limit device. The output of the first infrared sensor is connected to the PLC. When the yarn cake enters the yarn cake rotation limit device, the first infrared sensor sends a signal to the PLC. The PLC instructs two three-phase asynchronous motors to rotate in opposite directions to make the yarn cake rotate. At the same time, it instructs the lifting cylinder to move so that the limit pin protrudes from the surface of the roller assembly line and is embedded in the limit groove on the yarn cake support base to position the yarn cake. Then, the PLC instructs the lifting cylinder to lower the limit pin and instructs the two three-phase asynchronous motors to rotate in the same direction to send the yarn cake into the yarn cake defect detection device.
[0011] Furthermore, a strain gauge pressure sensor is mounted on the front side of the limiting pin. The output of the strain gauge pressure sensor is connected to a PLC to transmit the signal that the limiting pin is embedded in the limiting groove.
[0012] Furthermore, the first defect detection agency collects defect information of the silk cake paper tube, as well as information of the left circumferential surface of the silk cake, information of the upper and lower right end surfaces of the silk cake, and images of the fuzz on the upper and lower right end surfaces of the silk cake; the second defect detection agency collects defect information of the silk cake paper tube, as well as information of the right circumferential surface of the silk cake, information of the upper and lower left end surfaces of the silk cake, and images of the fuzz on the upper and lower left end surfaces of the silk cake; the plane that is equidistant from and parallel to the upper and lower end surfaces of the silk cake is defined as the center plane of the silk cake.
[0013] Furthermore, the first linear array camera and the first line light source of the first defect detection mechanism are mounted on the first linear array camera mounting bracket. The first linear motor drives the position movement of the first linear array camera and the first line light source. The first linear array camera is installed below the wire cake in the detection station and shoots vertically upwards. The second linear array camera and the second line light source are mounted on the second linear array camera mounting bracket. The second linear motor drives the position movement of the second linear array camera and the second line light source. The second linear array camera is installed directly above the wire cake in the detection station and shoots vertically downwards. The first and second linear array cameras are arranged symmetrically with respect to the center plane of the wire cake. In addition, a drag chain is provided to protect the transmission cables of the linear array camera and the line light source, and a linear motor is provided for adjusting the position of the linear array camera and the light source before detection.
[0014] The first array camera and the second array camera are positioned above the paper tube and shooting downwards at an angle. The first array camera is used to collect information about the outside of the paper tube, and the second array camera is used to collect information about the inside of the paper tube. The first array camera and the second array camera are at an angle of 90-120 degrees.
[0015] The third, fourth, and fifth array cameras are positioned on the left side of the silk cake and at the same horizontal level as the center plane of the silk cake; the third and fourth array cameras form an angle of 40-60 degrees; the fourth and fifth array cameras form an angle of 40-60 degrees; the third, fourth, and fifth array cameras are used to acquire information about the left circumferential surface of the silk cake.
[0016] The sixth and seventh array cameras are positioned on the right side of the silk disc and at the same horizontal level as the top surface of the disc. They are symmetrically arranged on the right side of the disc at an angle of 60-120 degrees to each other, and are used to acquire images of the feathers on the upper left side of the disc. The eighth and ninth array cameras are positioned on the right side of the silk disc and at the same horizontal level as the bottom surface of the disc. They are symmetrically arranged on the right side of the disc at an angle of 60-120 degrees to each other, and are used to acquire images of the feathers on the lower left side of the disc. The sixth and eighth array cameras are mounted on the same support column. The seventh and ninth array cameras are mounted on the same support column.
[0017] A ring light source is positioned directly above the silk cake and parallel to its upper surface. A first linear light source is installed below the silk cake in the inspection station, illuminating it obliquely upwards in front of the first linear array camera. A second linear light source is installed above the silk cake in the inspection station, illuminating it obliquely downwards in front of the second linear array camera. The light sources are symmetrically arranged with respect to the central plane of the silk cake. A first arc-shaped light source is positioned to the lower left of the silk cake, and a second arc-shaped light source is positioned to the upper left of the silk cake. The first and second arc-shaped light sources are arranged at a 60-90 degree angle and are symmetrically arranged with respect to the central plane of the silk cake. A first backlight source and a second backlight source are positioned to the upper right of the silk cake and symmetrically arranged parallel to its upper surface.
[0018] The first line light source provides a detection light source environment for the first line scan camera, the second line light source provides a detection light source environment for the second line scan camera, the first arc light source and the second arc light source provide the light source environment required for image acquisition for the detection of the third, fourth, and fifth area scan cameras, the ring light source, the first arc light source and the second arc light source provide a detection light source environment for the first and second area scan cameras, and the first arc light source, the second arc light source, the first back light source and the second back light source provide the light source environment required for image acquisition for the detection of the sixth, seventh, eighth, and ninth area scan cameras.
[0019] The control terminals of the first linear scan camera, the second linear scan camera, the first area scan camera, the second area scan camera, the third area scan camera, the fourth area scan camera, the fifth area scan camera, the sixth area scan camera, the seventh area scan camera, the eighth area scan camera, and the ninth area scan camera are respectively connected to a PLC. The output terminals of the first linear scan camera and the second linear scan camera are connected to the first edge node. The output terminals of the first area scan camera and the second area scan camera are connected to the second edge node. The output terminals of the third area scan camera, the fourth area scan camera, and the fifth area scan camera are connected to the third edge node. The output terminals of the sixth area scan camera, the seventh area scan camera, the eighth area scan camera, and the ninth area scan camera are connected to the fourth edge node. At the same time, the first edge node, the second edge node, the third edge node, and the fourth edge node are connected to a PC host computer.
[0020] The wire cake enters the first rotary limiting device. The first infrared sensor sends a signal to the PLC. The PLC instructs the lifting cylinder to lift the limiting pin. The limiting pin rises and contacts the wire cake head. The strain gauge pressure sensor on the limiting pin transmits a signal to the PLC. Based on the pressure sensor signal, the PLC instructs the first rotary limiting device to operate. Then, the PLC instructs the lifting cylinder to lower the limiting pin. Simultaneously, the PLC instructs the first and second linear array cameras to perform detection for 0.1s-2s. The wire cake enters the first defect detection mechanism. The second infrared sensor sends a signal to the PLC. The PLC instructs the first, second, third, fourth, fifth, and sixth area array cameras... Seven-face array cameras, eighth-face array cameras, and ninth-face array cameras acquire images of the silk pancake. Then, the first and second linear array cameras transmit images to the first edge node for defect detection. The first and second array cameras transmit images to the second edge node for defect detection. The third, fourth, and fifth array cameras transmit images to the third edge node for defect detection. The sixth, seventh, eighth, and ninth array cameras transmit images to the fourth edge node for defect detection. Then, the first, second, third, and fourth edge nodes transmit defect image data to the PC host computer for display, defect statistics, and storage.
[0021] The control method of the second rotary limiting device and the second defect detection mechanism is the same as that of the first rotary limiting device and the first defect detection mechanism.
[0022] Furthermore, the first edge node is an FPGA-ZYNQ7020 core board, while the second, third, and fourth edge nodes are MAIXI K210 core boards.
[0023] Furthermore, a servo motor is installed on the gantry bracket in the silk cake surface pretreatment device. The servo motor drives a rolling brush through a shaft. The rolling brush is located 20mm to 40mm above the upper surface of the silk cake and is used to remove the silk ends and impurities from the upper surface of the silk cake.
[0024] Furthermore, guide rails are provided on both sides of the cylinder of the silk cake diverting device, a rubber pad is provided at the cylinder extension and retraction end, and a third infrared sensor is provided.
[0025] A method for inspecting the appearance of silk cakes using the aforementioned integrated rapid classification and defect detection equipment includes the following steps:
[0026] S1, based on the position and size of the chemical fiber cake and cake tray on the conveyor belt, the position of the industrial camera and light source is adjusted using the adjustment device inside the equipment to ensure that the industrial camera does not block the transport line of the chemical fiber cake, so that each industrial camera is aimed at its own object to be photographed.
[0027] S2, Place the chemical fiber cake on the cake tray, and both enter the testing equipment together with the cake conveyor.
[0028] S3, at the silk cake surface pretreatment device, the upper surface of the silk cake is treated to remove impurities and silk ends;
[0029] S4, position limitation at the first rotation limit device: When the first infrared sensor detects that the yarn cake has entered the first rotation limit device and receives the signal, the lifting cylinder lifts the limit pin. The pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the first rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motors on both sides of the device rotate in opposite directions, driving the rollers to rotate in the opposite direction so that the yarn cake support base rotates and the limit groove and limit pin are embedded in the limit. Then the PLC determines whether the limit is completed according to the pressure sensor signal. After the limit is completed, the PLC instructs the lifting cylinder to lower the limit pin.
[0030] S5, image acquisition is performed at the first defect detection mechanism: the PLC commands the lifting cylinder limit pin to descend, and then the PLC controls the first and second line array cameras to start image acquisition for 0.1S-2S; when the second infrared sensor at the first defect detection mechanism detects the passing of the yarn cake tray, after receiving the signal, the PLC commands the industrial camera group to take pictures of the upper and lower half of the chemical fiber yarn cake, the hairy image of the upper and lower half of the half of the half of the half of the half of the half of the half of the half of the half of the circumference, the paper tube color, and the paper tube defects, so as to realize the acquisition of half of the image of the entire surface information of the yarn cake.
[0031] S6, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOv5 model is applied to detect the image;
[0032] S7. The acquired defect image of the silk roll is transmitted to the PC host computer: the host computer displays the silk roll image and performs defect statistics in real time. If no defect label is detected, the silk roll is determined to be OK; otherwise, the silk roll is determined to be NG, and the label signal is transmitted to the PLC control terminal.
[0033] S8, position limitation at the second rotation limit device: When the infrared sensor at the second defect detection mechanism detects the silk cake entering the silk cake rotation limit device, after receiving the signal, the lifting cylinder lifts the limit pin, and the pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the second rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motor in the device rotates in the opposite direction, and the driving roller drives in the opposite direction so that the silk cake support base rotates and the limit groove and the limit pin are embedded in the limit, suppressing the difference in the rotation angle of the silk cake support base when entering the first defect detection mechanism and the second defect detection mechanism;
[0034] S9, Image acquisition is performed at the second defect detection mechanism: The PLC commands the lifting cylinder limit pin to descend. When the limit pin in the second rotary limit device descends, the PLC controls the first and second line array cameras to start image acquisition for 0.1s-2s. When the infrared sensor at the second defect detection mechanism detects the passing of the yarn cake tray, the PLC receives the infrared sensor signal and commands the industrial camera group to take pictures of the remaining half of the upper and lower end faces of the chemical fiber yarn cake, the half of the upper and lower end face feathers, the half of the circumferential surface, the paper tube color, and the paper tube defects, thereby obtaining half of the image acquisition of the full surface information of the entire chemical fiber yarn cake, thus completing the full surface information acquisition of the entire yarn cake.
[0035] S10, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOV5 model is applied to detect the image;
[0036] S11, the acquired defect image of the silk cake is transmitted to the PC host computer: the host computer displays the silk cake image and performs defect statistics in real time. If no defect label is detected, the silk cake is determined to be OK; otherwise, the silk cake is determined to be NG, and the label signal is transmitted to the PLC control terminal.
[0037] S12, at the silk cake separating device: when the third infrared sensor detects that the silk cake has entered the pushing and separating area, the PLC control terminal, according to the silk cake label transmitted by the host computer, pushes the silk cake separating device to separate and classify good and bad silk cakes.
[0038] Compared with the prior art, the beneficial effects of the present invention include at least the following:
[0039] 1. The integrated equipment for rapid classification and defect detection of silk cake provided by the present invention can be applied to large-scale production lines and has the advantages of easy control of detection cycle, fast detection speed, high space utilization efficiency, high utilization rate of light source setting, and low defect missed detection rate.
[0040] 2. Compared with some existing automated silk cake detection equipment, the present invention has two detection stations set up in a centralized manner, which has a high degree of concentration of detection equipment, a small space threshold, and a high light source utilization rate.
[0041] 3. By setting a rotation limiting mechanism, the present invention controls the detection cycle and adjusts the detection angle of the silk cake during detection, making the detection cycle of the equipment easier to control and reducing the rate of missed detection and repeated detection on the surface of the silk cake.
[0042] 4. This invention uses edge computing nodes to assist in defect detection, which can significantly improve detection efficiency and reduce the computing power requirements of a large amount of image data on the PC host computer. Attached Figure Description
[0043] Figure 1 This is a schematic diagram of the overall structure of the present invention. Figure 1 ;
[0044] Figure 2 This is a schematic diagram of the overall structure of the present invention. Figure 2 (Top view);
[0045] Figure 3 This is a schematic diagram of the silk cake surface pretreatment device of the present invention;
[0046] Figure 4 This is a schematic diagram of the defect detection device of the present invention. Figure 1 ;
[0047] Figure 5 This is a schematic diagram of the defect detection device of the present invention. Figure 2 ;
[0048] Figure 6 This is a schematic diagram of a single-detection station structure of the defect detection device of the present invention;
[0049] Figure 7 This is a schematic diagram of the single-station camera position in the defect detection device of the present invention. Figure 1 ;
[0050] Figure 8 This is a schematic diagram of the single-station camera position in the defect detection device of the present invention. Figure 2 ;
[0051] Figure 9 This is a schematic diagram of the structure of the silk cake rotation limiting device of the present invention. Figure 1 ;
[0052] Figure 10 This is a schematic diagram of the structure of the silk cake rotation limiting device of the present invention. Figure 2 ;
[0053] Figure 11 This is a schematic diagram of the structure of the silk cake rotation limiting device of the present invention. Figure 3 ;
[0054] Figure 12 This is a schematic diagram of the silk cake support base structure of the present invention;
[0055] Figure 13 This is a schematic diagram of the structure of the shredded cake diversion device of the present invention;
[0056] Figure 14 This is a schematic diagram of the control data flow of the defect detection device of the present invention;
[0057] Figure 15 The flowchart illustrates the steps of the integrated equipment for rapid classification of defects in silk cakes and its control method according to the present invention.
[0058] The labels in the diagram represent:
[0059] 1-Silk cake surface pretreatment device; 2-Silk cake; 3-Silk cake support base; 4-Silk cake defect detection device; 5-First silk cake rotation limiting device; 6-Second silk cake rotation limiting device; 7-Silk cake diversion device; 8-Silk cake transmission device; 101-Rolling brush; 102-Servo motor; 103-Motor support frame; 104-Gantry bracket; 105-Support foot; 301-Limiting groove; 4101-First defect detection mechanism; 4102-Second defect detection mechanism; 401-First line scan camera; 402-Second line scan camera; 403-First area scan camera. 404 - Second array camera, 405 - Third array camera, 406 - Fourth array camera, 407 - Fifth array camera, 408 - Sixth array camera, 409 - Seventh array camera, 410 - Eighth array camera, 411 - Ninth array camera, 412 - Ring light source connector, 413 - Ring light source, 414 - First line light source, 415 - Second line light source, 416 - First curved light source, 417 - Second curved light source, 418 - First backlight source, 419 - Second backlight source, 420 - First line array camera mount, 421 - Second line array camera Fixed frame, 422-First linear motor, 424-Second linear motor, 423-First cable chain, 425-Second cable chain, 426-Second infrared sensor, 427-Camera light source assembly fixing plate, 428-Aluminum profile bracket, 429-Fixed support foot; 501-First three-phase asynchronous motor, 502-Second three-phase asynchronous motor, 503-First motor fixing connector, 504-Second motor fixing connector, 505-First fixing plate connector, 506-Roller center fixing plate, 507-Double chain roller, 508-First roller fixing plate, 509-The first linear motor, 421-First linear motor, 422-Second linear motor, 423-First cable chain, 425-Second cable chain, 426-Second infrared sensor, 427-Camera light source assembly fixing plate, 428-Aluminum profile bracket, 429-Fixed support foot; 501-First three-phase asynchronous motor, 502-Second three-phase asynchronous motor, 503-First motor fixing connector, 504-Second motor fixing connector, 505-First fixing plate connector, 506-Roller center fixing plate, 507-Double chain roller, 508-First roller fixing plate, 509-The second linear motor, 421-Second linear motor, 422-First linear motor, 424-Second linear motor, 425-Second linear motor, 426-Second linear motor, 427-Camera light source assembly fixing plate, 428-Aluminum profile bracket, 429-Fixed support foot, 420-Second linear motor, 420-Second linear motor, 421-First linear motor, 422-Second linear motor, 423 Two roller fixing plate, 510-Second fixing plate connector, 511-Limit pin, 512-Lifting fixing plate, 513-First lifting mechanism fixing component, 514-Second lifting mechanism fixing component, 515-Lifting cylinder, 516-First rubber pad, 517-First infrared sensor, 518-Second reducer, 519-First reducer, 520-Strain gauge pressure sensor, 521-Sprocket; 701-Second rubber pad, 702-Push mechanism fixing base, 703-Fixed slider, 704-Cylinder, 705-Guide rail, 706-Third infrared sensor. Detailed Implementation
[0060] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.
[0061] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.
[0062] The present invention will now be described in detail with reference to the accompanying drawings and exemplary embodiments.
[0063] refer to Figure 1 and Figure 2 The present invention provides an integrated device for rapid classification and defect detection of silk cakes, comprising a silk cake surface pretreatment device 1, a silk cake defect detection device 4, a silk cake rotation limiting device, and a silk cake diversion device 7, which are sequentially arranged on the roller production line of the silk cake transmission device 8 and controlled by an embedded system PLC. The silk cake defect detection device 4 acquires an image of the surface of the silk cake 2 and outputs it to an edge computing node. The edge computing node identifies and detects the image and outputs it to a PC host computer. The PC host computer performs real-time silk cake image display, defect statistics, and storage.
[0064] The yarn cake defect detection device includes a first defect detection machine 4101 and a second defect detection mechanism 4102. The first defect detection mechanism 4101 and the second defect detection mechanism 4102 are located on both sides of the roller conveyor line and are mirror images of each other with respect to the longitudinal axis of the roller conveyor line. The first defect detection mechanism 4101 is located upstream of the second defect detection mechanism 4102 and performs image acquisition on the opposite side of the yarn cake 2. A yarn cake rotation limiting device is installed on the roller conveyor line close to the upstream of the first defect detection mechanism 4101 and the second defect detection mechanism 4102, namely a first yarn cake rotation limiting device 5 and a second yarn cake rotation limiting device 6. The first and second yarn cake rotation limiting devices limit the rotation phase of the yarn cake 2 with respect to the roller conveyor line.
[0065] like Figure 9 , Figure 10 , Figure 11 and Figure 12 As shown, it should be noted that the structure and control method of the first yarn cake rotation limiting device 5 and the second yarn cake rotation limiting device 6 are the same. The first yarn cake rotation limiting device 5 includes two rows of parallel rollers 507. The two rows of rollers 507 are respectively connected to a first three-phase asynchronous motor 501 and a second three-phase asynchronous motor 502 through a transmission chain 521. The control terminals of the first and second three-phase asynchronous motors are connected to a PLC. When the PLC controls the first and second three-phase asynchronous motors to rotate in the same direction, the yarn cake is transported forward. When the PLC controls the first and second three-phase asynchronous motors to rotate in opposite directions, the yarn cake is rotated.
[0066] It also includes a limit pin 511 set on the lifting cylinder 515, and a first infrared sensor 517 for detecting the entry of the yarn cake into the yarn cake rotation limit device. The output of the first infrared sensor 517 is connected to the PLC. When the yarn cake 2 enters the yarn cake rotation limit device, the first infrared sensor 517 sends a signal to the PLC. The PLC instructs the two three-phase asynchronous motors to rotate in opposite directions to make the yarn cake rotate. At the same time, it instructs the lifting cylinder 515 to move so that the limit pin 511 is exposed on the surface of the roller production line and embedded in the limit groove 301 on the yarn cake support base 3 to position the yarn cake. Then, the PLC instructs the lifting cylinder 515 to lower the limit pin 511 and instructs the two three-phase asynchronous motors to rotate in the same direction to send the yarn cake 2 into the yarn cake defect detection device 4.
[0067] like Figure 10 As shown, a strain gauge pressure sensor 520 is mounted on the front side of the limiting pin 511. The output of the strain gauge pressure sensor 520 is connected to a PLC to transmit the signal that the limiting pin 511 is embedded in the limiting groove 301.
[0068] Specifically, the first and second three-phase asynchronous motors are respectively mounted on the first roller fixing plate 509 and the second roller fixing plate 508 via the first motor fixing connector 503 and the second motor fixing connector 504; one end of one row of rollers is mounted on the first roller fixing plate, one end of the other row of rollers is mounted on the second roller fixing plate, and the other ends of the two rows of rollers are respectively mounted on a roller center fixing plate 506. The two ends of the two roller center fixing plates 506 are respectively provided with a first fixing plate connector 505 and a second fixing plate connector 506. The fixed plate connector 510 connects the two rows of rollers together; the limiting pin is located between the center fixed plates of the two rollers and its lower end is connected to the lifting cylinder 515 through a lifting fixed plate 512. The end of the lifting cylinder 515 is provided with a first rubber pad 516. The two ends of the lifting cylinder 515 are fixed on the first lifting mechanism fixing member 513 and the second lifting fixing member 514, and the upper end surfaces of the first lifting mechanism fixing member 513 and the second lifting fixing member 514 are respectively connected to the first roller fixing plate 509 and the second roller fixing plate 508.
[0069] Specifically, a first reducer 519 and a second reducer 518 are respectively installed on the first and second three-phase asynchronous motors. The input end of the first reducer 519 is connected to the first three-phase asynchronous motor 501, and the input end of the second reducer 518 is connected to the output end of the second three-phase asynchronous motor 502.
[0070] like Figure 5 and Figure 6 As shown, it should be noted that the structure and control method of the first defect detection mechanism 4101 and the second defect detection mechanism 4102 are the same.
[0071] The first defect detection unit 4101 collects defect information of the silk cake paper tube, as well as information of the left circumferential surface of the silk cake, information of the upper and lower right end surfaces of the silk cake, and images of the fuzz on the upper and lower right end surfaces of the silk cake; the second defect detection unit 4102 collects defect information of the silk cake paper tube, as well as information of the right circumferential surface of the silk cake, information of the upper and lower left end surfaces of the silk cake, and images of the fuzz on the upper and lower left end surfaces of the silk cake; the plane that is equidistant from the upper and lower end surfaces of the silk cake 2 is defined as the center plane of the silk cake.
[0072] The first line array camera 401 and the first line light source 414 of the first defect detection mechanism 4101 are mounted on the first line array camera mounting bracket 420. The first linear motor 422 drives the first line array camera 401 and the first line light source 414 to move their positions. The first line array camera 401 is installed below the wire cake in the detection station and shoots vertically upwards. The second line array camera 402 and the second line light source 415 are mounted on the second line array camera mounting bracket 421. The second linear motor 424 drives the second line array camera 402 and the second line light source 415 to move their positions. The second line array camera 402 is installed directly above the wire cake in the detection station and shoots vertically downwards. The first line array camera 401 and the second line array camera 402 are arranged symmetrically with respect to the center plane of the wire cake. In addition, the first drag chain 423 and the second drag chain 425 are provided to protect the transmission cables of the line array camera and the line light source. The second and second linear motors are provided for adjusting the positions of the line array camera and the light source before detection.
[0073] The first array camera 403 and the second array camera 404 are positioned above the paper tube and take pictures at an angle downwards. The first array camera 403 is used to collect information on the outside of the paper tube, and the second array camera 404 is used to collect information on the inside of the paper tube. The first array camera 403 and the second array camera 404 form an angle of 90-120 degrees.
[0074] The third array camera 405, the fourth array camera 406, and the fifth array camera 407 are positioned on the left side of the silk cake and are at the same horizontal height as the center plane of the silk cake; the third array camera 405 and the fourth array camera 406 form an angle of 40-60 degrees; the fourth array camera 406 and the fifth array camera 407 form an angle of 40-60 degrees; the third array camera 405, the fourth array camera 406, and the fifth array camera 407 are used to collect information about the left circumferential surface of the silk cake;
[0075] The sixth and seventh array cameras 408 and 409 are positioned on the right side of the silk cake and at the same horizontal level as the upper surface of the silk cake. The sixth and seventh array cameras 408 and 409 are symmetrically arranged on the right side of the silk cake at an angle of 60-120 degrees to each other, and are used to acquire images of the feathers on the upper left side of the silk cake. The eighth and ninth array cameras 410 and 411 are positioned on the right side of the silk cake and at the same horizontal level as the lower surface of the silk cake. The eighth and ninth array cameras 410 and 411 are symmetrically arranged on the right side of the silk cake at an angle of 60-120 degrees to each other. A 0-degree angle is symmetrically arranged on the right side of the silk cake to collect images of the feathers on the lower left side of the silk cake; the sixth array camera 408 and the eighth array camera 410 are set on the same support column; the seventh array camera 409 and the ninth array camera 411 are set on the same support column; the support columns are respectively mounted on their respective camera light source assembly fixing plates 427, the camera light source assembly fixing plates 427 are mounted on the aluminum profile bracket 428, and the bottom of the aluminum profile bracket 428 is equipped with a fixed support foot 429.
[0076] A ring light source 413 is positioned directly above the silk cake and parallel to its upper surface via a ring light source connector 412. A first line light source 414 is installed below the silk cake in the inspection station, illuminating it obliquely upwards from the front of the first line array camera 401. A second line light source 415 is installed above the silk cake in the inspection station, illuminating it obliquely downwards from the front of the second line array camera 402. The light sources are symmetrically arranged with respect to the central plane of the silk cake. A first arc-shaped light source 416 is positioned to the left below the silk cake, and a second arc-shaped light source 417 is positioned to the left above the silk cake. The first arc-shaped light source 416 and the second arc-shaped light source 417 are arranged at a 60-90 degree angle and are symmetrically arranged with respect to the central plane of the silk cake. A first backlight source 418 and a second backlight source 419 are positioned to the right above the silk cake and are symmetrically arranged parallel to its upper surface.
[0077] The first line light source 414 provides a detection light source environment for the first line scan camera 401; the second line light source 415 provides a detection light source environment for the second line scan camera 402; the first arc light source 416 and the second arc light source 417 provide the light source environment required for image acquisition for the detection of the third area scan camera 405, the fourth area scan camera 406, and the fifth area scan camera 407; the ring light source 413, the first arc light source 416, and the second arc light source 417 provide a detection light source environment for the first area scan camera 403 and the second area scan camera 404; and the first arc light source 416, the second arc light source 417, the first back light source 418, and the second back light source 419 provide the light source environment required for image acquisition for the detection of the sixth area scan camera 408, the seventh area scan camera 409, the eighth area scan camera 410, and the ninth area scan camera 411.
[0078] The control terminals of the first line array camera 401, the second line array camera 402, the first area array camera 403, the second area array camera 404, the third area array camera 405, the fourth area array camera 406, the fifth area array camera 407, the sixth area array camera 408, the seventh area array camera 409, the eighth area array camera 410, and the ninth area array camera 411 are respectively connected to a PLC. The output terminals of the first line array camera 401 and the second line array camera 402 are connected to the first edge node. The output terminals of the first area array camera 403 and the second area array camera 404 are connected to the second edge node. The output terminals of the third area array camera 405, the fourth area array camera 406, and the fifth area array camera 407 are connected to the third edge node. The output terminals of the sixth area array camera 408, the seventh area array camera 409, the eighth area array camera 410, and the ninth area array camera 411 are connected to the fourth edge node. At the same time, the first edge node, the second edge node, the third edge node, and the fourth edge node are connected to a PC host computer.
[0079] The wire cake enters the first rotary limiting device 5. The first infrared sensor 517 sends a signal to the PLC. The PLC instructs the lifting cylinder 515 to lift the limiting pin 511. The limiting pin 511 rises and contacts the wire cake head. The strain gauge pressure sensor 520 on the limiting pin 511 transmits a signal to the PLC. Based on the pressure sensor signal, the PLC instructs the first rotary limiting device to operate. Then, the PLC instructs the lifting cylinder to lower the limiting pin 511. Simultaneously, the PLC instructs the first linear array camera 401 and the second linear array camera 402 to perform detection for 0.1s-2s. The wire cake 2 enters the first defect detection mechanism 4101. The second infrared sensor 426 sends a signal to the PLC. The PLC instructs the first area array camera 403, the second area array camera 404, the third area array camera 405, the fourth area array camera 406, the fifth area array camera 407, and the sixth area array camera 408 to perform detection. Cameras 408, 409, 410, and 411 acquire images of the silk pancake. Then, cameras 401 and 402 transmit images to the first edge node for defect detection. Cameras 403 and 404 transmit images to the second edge node for defect detection. Cameras 405, 406, and 407 transmit images to the third edge node for defect detection. Cameras 408, 409, 410, and 411 transmit images to the fourth edge node for defect detection. Finally, the first, second, third, and fourth edge nodes transmit defect image data to a PC for display, defect statistics, and storage.
[0080] The control method of the second rotation limiting device 6 and the second defect detection mechanism 4102 is the same as that of the first rotation limiting device 5 and the first defect detection mechanism 4101.
[0081] In one embodiment, the first edge node is an FPGA-ZYNQ7020 core board, and the second, third, and fourth edge nodes are MAIXI K210 core boards.
[0082] In one embodiment, a servo motor 102 is installed on the gantry bracket 104 in the silk cake surface pretreatment device 1. The servo motor 102 drives a rolling brush 101 through a shaft. The rolling brush 101 is located within a range of 20mm to 40mm above the upper end face of the silk cake and is used to remove the silk ends and impurities from the upper end face of the silk cake.
[0083] Specific examples Figure 3 As shown, the bottom of the gantry bracket 104 is provided with support feet 105, and the servo motor 102 is mounted on the gantry bracket 104 through the motor support bracket 103.
[0084] like Figure 13 As shown, the cylinder 704 of the yarn cake diverting device is provided with guide rails 705 on both sides, and a second rubber pad 701 and a third infrared sensor 706 are provided at the telescopic end of the cylinder 704. Specifically, the cylinder 704 is fixed on the fixed base 702 of the pushing mechanism, and a fixed slider 703 that cooperates with the guide rail is provided on the guide rail 705.
[0085] A method for inspecting the appearance of silk cakes using the aforementioned integrated rapid classification and defect detection equipment includes the following steps:
[0086] S1, based on the position and size of the chemical fiber cake and cake tray on the conveyor belt, the position of the industrial camera and light source is adjusted using the adjustment device inside the equipment to ensure that the industrial camera does not block the transport line of the chemical fiber cake, so that each industrial camera is aimed at its own object to be photographed.
[0087] S2, Place the chemical fiber cake on the cake tray, and both enter the testing equipment together with the cake conveyor.
[0088] S3, at the silk cake surface pretreatment device, the upper surface of the silk cake is treated to remove impurities and silk ends;
[0089] S4, position limitation at the first rotation limit device: When the first infrared sensor detects that the yarn cake has entered the first rotation limit device and receives the signal, the lifting cylinder lifts the limit pin. The pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the first rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motors on both sides of the device rotate in opposite directions, driving the rollers to rotate in the opposite direction so that the yarn cake support base rotates and the limit groove and limit pin are embedded in the limit. Then the PLC determines whether the limit is completed according to the pressure sensor signal. After the limit is completed, the PLC instructs the lifting cylinder to lower the limit pin.
[0090] S5, image acquisition is performed at the first defect detection mechanism: the PLC commands the lifting cylinder limit pin to descend, and then the PLC controls the first and second line array cameras to start image acquisition for 0.1S-2S; when the second infrared sensor at the first defect detection mechanism detects the passing of the yarn cake tray, after receiving the signal, the PLC commands the industrial camera group to take pictures of the upper and lower half of the chemical fiber yarn cake, the hairy image of the upper and lower half of the half of the half of the half of the half of the half of the half of the half of the half of the circumference, the paper tube color, and the paper tube defects, so as to realize the acquisition of half of the image of the entire surface information of the yarn cake.
[0091] S6, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOv5 model is applied to detect the image;
[0092] S7. The acquired defect image of the silk roll is transmitted to the PC host computer: the host computer displays the silk roll image and performs defect statistics in real time. If no defect label is detected, the silk roll is determined to be OK; otherwise, the silk roll is determined to be NG, and the label signal is transmitted to the PLC control terminal.
[0093] S8, position limitation at the second rotation limit device: When the infrared sensor at the second defect detection mechanism detects the silk cake entering the silk cake rotation limit device, after receiving the signal, the lifting cylinder lifts the limit pin, and the pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the second rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motor in the device rotates in the opposite direction, and the driving roller drives in the opposite direction so that the silk cake support base rotates and the limit groove and the limit pin are embedded in the limit, suppressing the difference in the rotation angle of the silk cake support base when entering the first defect detection mechanism and the second defect detection mechanism;
[0094] S9, Image acquisition is performed at the second defect detection mechanism: The PLC commands the lifting cylinder limit pin to descend. When the limit pin in the second rotary limit device descends, the PLC controls the first and second line array cameras to start image acquisition for 0.1s-2s. When the infrared sensor at the second defect detection mechanism detects the passing of the yarn cake tray, the PLC receives the infrared sensor signal and commands the industrial camera group to take pictures of the remaining half of the upper and lower end faces of the chemical fiber yarn cake, the half of the upper and lower end face feathers, the half of the circumferential surface, the paper tube color, and the paper tube defects, thereby obtaining half of the image acquisition of the full surface information of the entire chemical fiber yarn cake, thus completing the full surface information acquisition of the entire yarn cake.
[0095] S10, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOV5 model is applied to detect the image;
[0096] S11, the acquired defect image of the silk cake is transmitted to the PC host computer: the host computer displays the silk cake image and performs defect statistics in real time. If no defect label is detected, the silk cake is determined to be OK; otherwise, the silk cake is determined to be NG, and the label signal is transmitted to the PLC control terminal.
[0097] S12, at the silk cake separating device: when the third infrared sensor detects that the silk cake has entered the pushing and separating area, the PLC control terminal, according to the silk cake label transmitted by the host computer, pushes the silk cake separating device to separate and classify good and bad silk cakes.
[0098] The integrated equipment for rapid classification and defect detection of fiber cakes provided by this invention can be applied to large-scale production lines, and the detection process is non-contact, highly efficient, and highly adaptable. Compared with traditional manual inspection methods, it avoids the errors that are easily made by workers. Furthermore, because it has two inspection stations, images of each inspection surface of the fiber cake can be acquired, enabling simultaneous detection of defects such as stains, strands, fuzz, and delamination. Moreover, by using edge nodes for auxiliary calculations, the computational requirements on the PC are reduced due to excessive image data. The model uses the YOLOv5 target detection model, so changes to the detection target require minimal modifications to the equipment, making it easy to adjust. In addition, when integrated with other automated production lines, it can significantly improve efficiency.
[0099] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.
[0100] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0101] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0102] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.
[0103] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0104] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.
Claims
1. A rapid classification and defect detection integrated device for silk cakes, comprising a silk cake surface pretreatment device, a silk cake defect detection device, a silk cake rotation limiting device, and a silk cake diversion device, sequentially arranged on the roller conveyor of the silk cake transmission device and controlled by an embedded PLC system; the silk cake defect detection device acquires an image of the silk cake surface and outputs it to an edge computing node; the edge computing node identifies and detects the image and outputs it to a PC-based host computer; the PC-based host computer performs real-time silk cake image display, defect statistics, and storage; characterized in that: The yarn cake defect detection device includes a first defect detection mechanism and a second defect detection mechanism, which are located on opposite sides of the roller conveyor line and are mirror images of each other relative to the longitudinal axis of the roller conveyor line. The first defect detection mechanism is located upstream of the second defect detection mechanism and performs image acquisition on the opposite side of the yarn cake. A yarn cake rotation limiting device is installed on the roller conveyor line, close to the upstream of the first and second defect detection mechanisms, to limit the rotation phase of the yarn cake relative to the roller conveyor line. The silk cake rotation limiting device includes two parallel rollers. The two rollers are connected to their respective three-phase asynchronous motors via a transmission chain. The control terminals of the three-phase asynchronous motors are connected to a PLC. When the PLC controls the two three-phase asynchronous motors to rotate in the same direction, the silk cake is transported forward. When the PLC controls the two three-phase asynchronous motors to rotate in opposite directions, the silk cake is rotated. It also includes a limit pin mounted on the lifting cylinder, and a first infrared sensor for detecting when the yarn cake enters the yarn cake rotation limit device. The output of the first infrared sensor is connected to the PLC. When the yarn cake enters the yarn cake rotation limit device, the first infrared sensor sends a signal to the PLC. The PLC instructs two three-phase asynchronous motors to rotate in opposite directions to make the yarn cake rotate. At the same time, it instructs the lifting cylinder to move so that the limit pin protrudes from the surface of the roller assembly line and is embedded in the limit groove on the yarn cake support base to position the yarn cake. Then, the PLC instructs the lifting cylinder to lower the limit pin and instructs the two three-phase asynchronous motors to rotate in the same direction to send the yarn cake into the yarn cake defect detection device.
2. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 1, characterized in that: A strain gauge pressure sensor is mounted on the front side of the limit pin. The output of the strain gauge pressure sensor is connected to the PLC to transmit the signal that the limit pin is embedded in the limit groove.
3. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 1, characterized in that: The first defect detection agency collects defect information of the silk cake paper tube, as well as information of the left circumferential surface of the silk cake, information of the upper and lower right end surfaces of the silk cake, and images of the fuzz on the upper and lower right end surfaces of the silk cake; the second defect detection agency collects defect information of the silk cake paper tube, as well as information of the right circumferential surface of the silk cake, information of the upper and lower left end surfaces of the silk cake, and images of the fuzz on the upper and lower left end surfaces of the silk cake; the plane that is equidistant from the upper and lower end surfaces of the silk cake is defined as the center plane of the silk cake.
4. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 3, characterized in that: The first linear array camera and the first linear light source of the first defect detection mechanism are mounted on a first linear array camera mount. A first linear motor drives the position of the first linear array camera and the first linear light source. The first linear array camera is installed below the wire cake in the detection station, shooting vertically upwards. The second linear array camera and the second linear light source are mounted on a second linear array camera mount. A second linear motor drives the position of the second linear array camera and the second linear light source. The second linear array camera is installed directly above the wire cake in the detection station, shooting vertically downwards. The first and second linear array cameras are arranged symmetrically with respect to the center plane of the wire cake. Furthermore, a drag chain is provided to protect the transmission cables of the linear array camera and the linear light source, and a linear motor is provided for adjusting the position of the linear array camera and the light source before detection. The first array camera and the second array camera are positioned above the paper tube and shooting downwards at an angle. The first array camera is used to collect information about the outside of the paper tube, and the second array camera is used to collect information about the inside of the paper tube. The first array camera and the second array camera are at an angle of 90-120 degrees. The third, fourth, and fifth array cameras are positioned on the left side of the silk cake and at the same horizontal level as the center plane of the silk cake; the third and fourth array cameras form an angle of 40-60 degrees; the fourth and fifth array cameras form an angle of 40-60 degrees; the third, fourth, and fifth array cameras are used to acquire information about the left circumferential surface of the silk cake. The sixth and seventh array cameras are positioned on the right side of the silk disc and at the same horizontal level as the top surface of the disc. They are symmetrically arranged on the right side of the disc at an angle of 60-120 degrees to each other, and are used to acquire images of the feathers on the upper left side of the disc. The eighth and ninth array cameras are positioned on the right side of the silk disc and at the same horizontal level as the bottom surface of the disc. They are symmetrically arranged on the right side of the disc at an angle of 60-120 degrees to each other, and are used to acquire images of the feathers on the lower left side of the disc. The sixth and eighth array cameras are mounted on the same support column. The seventh and ninth array cameras are mounted on the same support column. A ring light source is positioned directly above the silk cake and parallel to its upper surface. A first linear light source is installed below the silk cake in the inspection station, illuminating it obliquely upwards in front of the first linear array camera. A second linear light source is installed above the silk cake in the inspection station, illuminating it obliquely downwards in front of the second linear array camera. The light sources are symmetrically arranged with respect to the central plane of the silk cake. A first arc-shaped light source is positioned to the lower left of the silk cake, and a second arc-shaped light source is positioned to the upper left of the silk cake. The first and second arc-shaped light sources are arranged at a 60-90 degree angle and are symmetrically arranged with respect to the central plane of the silk cake. A first backlight source and a second backlight source are positioned to the upper right of the silk cake and symmetrically arranged parallel to its upper surface. The first line light source provides a detection light source environment for the first line scan camera, the second line light source provides a detection light source environment for the second line scan camera, the first arc light source and the second arc light source provide the light source environment required for image acquisition for the detection of the third, fourth, and fifth area scan cameras, the ring light source, the first arc light source and the second arc light source provide a detection light source environment for the first and second area scan cameras, and the first arc light source, the second arc light source, the first back light source and the second back light source provide the light source environment required for image acquisition for the detection of the sixth, seventh, eighth, and ninth area scan cameras. The control terminals of the first linear scan camera, the second linear scan camera, the first area scan camera, the second area scan camera, the third area scan camera, the fourth area scan camera, the fifth area scan camera, the sixth area scan camera, the seventh area scan camera, the eighth area scan camera, and the ninth area scan camera are respectively connected to a PLC. The output terminals of the first linear scan camera and the second linear scan camera are connected to the first edge node. The output terminals of the first area scan camera and the second area scan camera are connected to the second edge node. The output terminals of the third area scan camera, the fourth area scan camera, and the fifth area scan camera are connected to the third edge node. The output terminals of the sixth area scan camera, the seventh area scan camera, the eighth area scan camera, and the ninth area scan camera are connected to the fourth edge node. At the same time, the first edge node, the second edge node, the third edge node, and the fourth edge node are connected to a PC host computer. The wire cake enters the first rotary limiting device. The first infrared sensor sends a signal to the PLC. The PLC instructs the lifting cylinder to lift the limiting pin. The limiting pin rises and contacts the wire cake head. The strain gauge pressure sensor on the limiting pin transmits a signal to the PLC. Based on the pressure sensor signal, the PLC instructs the first rotary limiting device to operate. Then, the PLC instructs the lifting cylinder to lower the limiting pin. Simultaneously, the PLC instructs the first and second linear array cameras to perform detection for 0.1s-2s. The wire cake enters the first defect detection mechanism. The second infrared sensor sends a signal to the PLC. The PLC instructs the first, second, third, fourth, fifth, and sixth area array cameras... Seven-face array cameras, eighth-face array cameras, and ninth-face array cameras acquire images of the silk pancake. Then, the first and second linear array cameras transmit images to the first edge node for defect detection. The first and second array cameras transmit images to the second edge node for defect detection. The third, fourth, and fifth array cameras transmit images to the third edge node for defect detection. The sixth, seventh, eighth, and ninth array cameras transmit images to the fourth edge node for defect detection. Then, the first, second, third, and fourth edge nodes transmit defect image data to the PC host computer for display, defect statistics, and storage. The control method of the second rotary limiting device and the second defect detection mechanism is the same as that of the first rotary limiting device and the first defect detection mechanism.
5. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 4, characterized in that: The first edge node is an FPGA-ZYNQ7020 core board, while the second, third, and fourth edge nodes are MAIXIK210 core boards.
6. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 1, characterized in that: A servo motor is installed on the gantry support in the silk cake surface pretreatment device. The servo motor drives a rolling brush through a shaft. The rolling brush is located 20mm to 40mm above the upper surface of the silk cake and is used to remove the silk ends and impurities from the upper surface of the silk cake.
7. The integrated equipment for rapid classification and defect detection of silk cakes as described in claim 1, characterized in that: The cylinder of the silk cake diverting device is equipped with guide rails on both sides, a rubber pad at the cylinder extension end, and a third infrared sensor.
8. A method for inspecting the appearance of a silk cake using the integrated equipment for rapid classification and defect detection of silk cakes as described in claim 1, comprising the following steps: S1, based on the position and size of the chemical fiber cake and cake tray on the conveyor belt, the position of the industrial camera and light source is adjusted using the adjustment device inside the equipment to ensure that the industrial camera does not block the transport line of the chemical fiber cake, so that each industrial camera is aimed at its own object to be photographed. S2, Place the chemical fiber cake on the cake tray, and both enter the testing equipment together with the cake conveyor. S3, at the silk cake surface pretreatment device, the upper surface of the silk cake is treated to remove impurities and silk ends; S4, position limitation at the first rotation limit device: When the first infrared sensor detects that the yarn cake has entered the first rotation limit device and receives the signal, the lifting cylinder lifts the limit pin. The pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the first rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motors on both sides of the device rotate in opposite directions, and the driving rollers drive in reverse to make the yarn cake support base rotate and the limit groove and limit pin are embedded in the limit. Then the PLC determines whether the limit is completed according to the pressure sensor signal. After the limit is completed, the PLC instructs the lifting cylinder to lower the limit pin. S5, Image acquisition is performed at the first defect detection mechanism: The PLC commands the lifting cylinder limit pin to descend, and then the PLC controls the first and second line array cameras to start image acquisition for 0.1S-2S; When the second infrared sensor at the first defect detection mechanism detects the passing of the yarn cake tray, after receiving the signal, the PLC commands the industrial camera group to take pictures of the upper and lower half of the chemical fiber yarn cake, the hairy image of the upper and lower half of the half of the half of the half of the half of the half of the half of the half of the half of the circumference, the paper tube color, and the paper tube defects, so as to realize the acquisition of half of the image of the entire surface information of the yarn cake. S6, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOv5 model is applied to detect the image; S7, transmit the acquired defect image of the silk cake to the PC host computer: the host computer performs real-time silk cake image display and defect statistics. If no defect label is detected, the silk cake is determined to be OK; otherwise, the silk cake is determined to be NG, and the label signal is transmitted to the PLC control terminal. S8, position limitation at the second rotation limit device: When the infrared sensor at the second defect detection mechanism detects the silk cake entering the silk cake rotation limit device, after receiving the signal, the lifting cylinder lifts the limit pin, and the pressure sensor on the limit pin receives the signal and transmits it to the PLC. The PLC instructs the second rotation limit device whether to perform rotation limit according to the pressure sensor signal. If it needs to be performed, the motor in the device rotates in the opposite direction, and the driving roller drives in the opposite direction so that the silk cake support base rotates and the limit groove and the limit pin are embedded in the limit, suppressing the difference in the rotation angle of the silk cake support base when entering the first defect detection mechanism and the second defect detection mechanism; S9, Image acquisition is performed at the second defect detection mechanism: The PLC commands the lifting cylinder limit pin to descend. When the limit pin in the second rotary limit device descends, the PLC controls the first and second line array cameras to start image acquisition for 0.1s-2s. When the infrared sensor at the second defect detection mechanism detects the passing of the yarn cake tray, the PLC receives the infrared sensor signal and commands the industrial camera group to take pictures of the remaining half of the upper and lower end faces of the chemical fiber yarn cake, the half of the upper and lower end face feathers, the half of the circumferential surface, the paper tube color, and the paper tube defects, thereby obtaining half of the image acquisition of the full surface information of the entire chemical fiber yarn cake, thus completing the full surface information acquisition of the entire yarn cake. S10, Detect the acquired image: The acquired image is fed into the edge computing node, and the trained YOLOV5 model is applied to detect the image; S11, the acquired defect image of the silk cake is transmitted to the PC host computer: the host computer displays the silk cake image and performs defect statistics in real time. If no defect label is detected, the silk cake is determined to be OK; otherwise, the silk cake is determined to be NG, and the label signal is transmitted to the PLC control terminal. S12, at the silk cake separating device: when the third infrared sensor detects that the silk cake has entered the pushing and separating area, the PLC control terminal, according to the silk cake label transmitted by the host computer, pushes the silk cake separating device to separate and classify good and bad silk cakes.
Citation Information
Patent Citations
Appearance detection device for chemical fiber spinning cake
CN109142369A
Detection device used for detecting cylindrical batteries
CN105964553A
Automatic detection system for appearances of chemical fiber spinning cakes
CN108362701A
Automatic positioning and assembling integrated production line
CN109531136A