A glue coating detection method and device based on laser projection video recognition

Abstract

The present invention discloses a glue coating detection method and device based on laser projection video recognition. The detection method includes the following steps: S1, laser projection line acquisition and recognition; S2, determination of the recognition area; S3, binarization of the recognition area; S4, detection of the glue state. Through video image processing and automatic algorithms, the present invention realizes automatic monitoring and analysis of the glue coating process, breaking away from the high cost and low efficiency of traditional manual methods and significantly improving work efficiency. Utilizing the principle of glue refraction, the continuity and shape of the glue can be accurately detected, avoiding errors that may arise from human judgment, achieving a high degree of automation and increasing reliability. Through real-time video monitoring and image processing, it can promptly detect whether the glue line is abnormally interrupted, and automatically identify the interruption location and distance, so that glue can be quickly replenished to ensure product quality.

Description

Technical Field

[0001] The present invention relates to the field of gluing technology, and in particular to a gluing detection method and device based on laser projection video recognition. Background Art

[0002] In electronic devices, the screen is an important interactive interface, but it needs to be waterproof in harsh environments. Transparent glass is usually covered in front of the LCD screen and sealed with glue to ensure stability and waterproofness. During production, CNC dispensing machines are used for dispensing, but there may be problems with bubbles or impurities in the glue, resulting in voids in the glue lines and affecting the waterproof effect. Because the glue is transparent and colorless, the glue coating can only be observed manually with the naked eye. However, manual methods are costly, inefficient, unreliable, and may leave marks. For devices with gaps, such as glass panels, water ingress may damage the equipment. Therefore, an automatic and accurate glue continuity detection method is needed.

[0003] Chinese patent publication number CN108896545A discloses a method, device, and computer-readable storage medium for gluing inspection. The method includes the following steps: capturing a first image of a gluing product and a second image of the gluing product at the same inspection position; fusing the first and second images to obtain a corresponding third image; and, based on the third image, inspecting the gluing product to determine if it is qualified.

[0004] The above-mentioned patented technical solution has some shortcomings. When the shape of the sample to be tested changes, the lighting plan needs to be re-determined, which reduces real-time performance and applicability. The high requirements for equipment increase the detection cost and it is difficult to cope with real-time changes in the dynamic gluing process. These limitations affect its practical application and stability. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and device for gluing detection based on laser projection video recognition to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a method for detecting glue coating based on laser projection video recognition, comprising the following steps:

[0008] S1. Laser projection line acquisition and recognition: By tilting the laser projection device and adding a refractive lens, the light path is adjusted to achieve the acquisition and recognition of the laser projection line;

[0009] S2. Determine the recognition area: Calculate the ratio of the effective segmentation distance of each image to the camera pixel and the actual distance, and define the recognition area;

[0010] S3. Binarization of the identified area: A narrowband filter highlights the light stripes formed by the laser light source, performs binary processing on the highlighted light stripes, and uses brightness thresholds to distinguish between light and darkness, generating multiple area images.

[0011] S4. Detect glue status: Analyze the laser line refraction and line distance ratio to determine whether the glue is broken, and determine whether the glue strip is interrupted based on the number of consecutive suspected broken glue slices.

[0012] As a further solution of the present invention, the specific method of adjusting the optical path to achieve the collection and recognition of the laser projection line is as follows:

[0013] T1. By adjusting the relative position of the two refractive lenses, the angle of the refractive lenses can be adjusted, thereby adjusting the distance between the laser projection lines L1 and L3;

[0014] T2. Install mapping line L2 between L1 and L3 to ensure that L1, L2 and L3 remain parallel;

[0015] T3. Set the ratio of the distance between the line L1 and L2 to the distance between the line L2 and L3 to PP, where PP is a constant not equal to 1. Adjust PP to ensure that the projection line spacing meets the specific requirements of different strip widths;

[0016] T4. By installing a camera equipped with a narrow-band filter, the laser projection lines formed on the product casing can be collected and identified.

[0017] As a further solution of the present invention, the specific method of determining the identification area is as follows:

[0018] A1. Calculate the time required for the entire gluing process and the total number of frames captured by the camera based on the dispensing speed, total length of the dispensing line, and the camera frame rate.

[0019] A2. Calculate the effective segmentation distance for each image based on the total length of the dispensing line and the total number of frames captured by the camera.

[0020] A3. Based on the principle of monocular imaging, the ratio of each camera pixel to the actual distance is inferred;

[0021] A4. Based on the dispensing head's forward direction and the effective segmentation distance of each image, the pixel values ​​within a certain range above and below the midpoint are intercepted as the recognition area.

[0022] As a further solution of the present invention, the ratio of each pixel to the actual distance can also be obtained by a reference object calibration method.

[0023] As a further solution of the present invention, the specific method of detecting the glue state is as follows:

[0024] B1. Calculate the linear distance ratio of L3 relative to L2 and L1 in each slice based on the overall image area captured by the camera, and create a standard reference value ZZ template.

[0025] B2. During the detection process, calculate the current distance ratio of L3 to L2 and L1 and the reference value ZZ;

[0026] B3. Compare the current line distance ratio of L3 relative to L2 and L1 with the reference value ZZ. If the difference exceeds the preset threshold, it is determined to be a suspected glue break;

[0027] B4. Count the number of consecutive suspected glue-broken slices. If the number exceeds the set threshold, it is considered that the glue strip is broken.

[0028] As a further solution of the present invention, the specific method of creating the standard reference value ZZ template is as follows:

[0029] Determine the overall area and direction of the camera's image capture based on the motion trajectory and speed of the dispensing equipment as it runs the entire dispensing path;

[0030] Slice the entire image area and calculate the line distance ratio of L3 relative to L2 and L1 in each slice;

[0031] Remove p% of the extreme value slices and retain 1-2*p% of the slices in the middle part. Calculate the sum of the line distance ratios of this part and divide it by the number of remaining slices to obtain the average value ZZ of these values ​​to form the reference ZZ template.

[0032] The present invention also provides a glue coating detection device based on laser projection video recognition, which is used to implement the glue coating detection method based on laser projection video recognition described in the above item, including a slide, a shell, a glue dispensing moving mechanism, a glue dispensing device and a glue strip detection module; the slide includes an X slide and a Y slide, which are used to position and move the glue dispensing mechanism and place the shell; the shell provides a protection and support structure to ensure the safe operation of the entire device; the glue dispensing moving mechanism is used to drive the glue dispensing module to move to the set glue coating position; the glue dispensing module includes a glue dispensing controller and a glue dispensing nozzle, which are used to apply glue to the workpiece; the glue strip detection module is used to monitor and analyze the glue strips formed during the glue dispensing process in real time.

[0033] As a further solution of the present invention, the rubber strip detection module includes a line laser projection device, a narrow-band filter, a camera, a refractive lens, a servo mechanism and a control unit; the line laser projection device is used to emit light to the surface of the rubber strip; the narrow-band filter is installed on the camera lens to filter out light other than the laser line; the camera is used to capture the projection image of the laser line on the rubber strip; the refractive lens is used to change the direction of the light path; the servo mechanism includes a servo motor and a servo gear, which are used to control the line laser projection device to rotate around the dispensing nozzle; the control unit is responsible for the control and coordination of the entire system.

[0034] As a further solution of the present invention, the servo gear can be adjusted between 0° and 180°.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1) Through video image processing and automatic algorithms, the glue coating process is automatically monitored and analyzed, eliminating the high cost and low efficiency of traditional manual methods and significantly improving work efficiency;

[0037] 2) Using the principle of glue refraction, it can accurately detect the continuity and shape of glue, avoiding the errors that may be caused by human judgment, achieving high automation and increasing reliability;

[0038] 3) Through real-time video monitoring and image processing, it can promptly detect whether the glue line is abnormally interrupted, and automatically identify the interruption location and distance so that glue can be quickly replenished to ensure product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 The following is a flowchart of a method for gluing detection based on laser projection video recognition;

[0040] Figure 2 This is a structural diagram of a glue coating detection device based on laser projection video recognition;

[0041] Figure 3 It is a structural diagram of the dispensing module;

[0042] Figure 4 This is a structural diagram of the rubber strip detection module.

[0043] In the figure: 1. X slide; 2. Y slide; 3. Housing; 4. Glue dispensing movement mechanism and control module; 5. Glue dispensing module; 6. Glue dispensing controller; 7. Glue dispensing nozzle; 8. Glue strip detection module; 9. Servo motor; 10. Servo gear; 11. Line laser projection device; 12. Refractive lens; 13. Camera; 14. Control unit. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. Example

[0045] like Figure 1 As shown, a method for detecting glue coating based on laser projection video recognition according to an embodiment of the present invention includes the following steps:

[0046] S1. Laser projection line acquisition and recognition: By tilting the laser projection device and adding a refractive lens, the light path is adjusted to achieve the acquisition and recognition of the laser projection line;

[0047] S2. Determine the recognition area: Calculate the ratio of the effective segmentation distance of each image to the camera pixel and the actual distance, and define the recognition area;

[0048] S3. Binarization of the identified area: A narrowband filter highlights the light stripes formed by the laser light source, performs binary processing on the highlighted light stripes, and uses brightness thresholds to distinguish between light and darkness, generating multiple area images.

[0049] S4. Detect glue status: Analyze the laser line refraction and line distance ratio to determine whether the glue is broken, and determine whether the glue strip is interrupted based on the number of consecutive suspected broken glue slices.

[0050] In an embodiment of the present invention, the specific manner of adjusting the optical path to achieve the collection and recognition of the laser projection line is as follows:

[0051] T1. By adjusting the relative position of the two refractive lenses, the angle of the refractive lenses can be adjusted, and then the distance between the laser projection lines L1 and L3 can be adjusted; it can adapt to different widths of rubber strips,

[0052] T2. Install a mapping line L2 in the middle of L1 and L3 to ensure that L1, L2, and L3 remain parallel; this can ensure that a certain projection line spacing is maintained at a specific distance.

[0053] T3. Set the ratio of the distance between the line L1 and L2 to the distance between the line L2 and L3 to PP, where PP is a constant not equal to 1. Adjust PP to ensure that the projection line spacing meets the specific requirements of different strip widths;

[0054] T4. By adding a camera equipped with a narrowband filter, the laser projection lines formed on the product housing can be collected and identified. This can be used for subsequent data analysis, measurement, or inspection.

[0055] The above solution can achieve accurate laser projection positioning, flexibly adjust the projection line spacing to ensure parallel and clear projection lines, adapt to the requirements of different rubber strip widths, and efficiently identify laser projection lines to provide accurate data for subsequent processing, which helps to improve production efficiency, accuracy and product quality.

[0056] In the embodiment of the present invention, the specific method of demarcating the identification area is as follows:

[0057] A1. Calculate the time required for the entire gluing process and the total number of frames captured by the camera based on the dispensing speed, total length of the dispensing line, and the camera frame rate.

[0058] A2. Calculate the effective segmentation distance for each image based on the total length of the dispensing line and the total number of frames captured by the camera.

[0059] A3. Based on the principle of monocular imaging, the ratio of each camera pixel to the actual distance is inferred;

[0060] A4. Based on the dispensing head's forward direction and the effective segmentation distance of each image, the pixel values ​​within a certain range above and below the midpoint are intercepted as the recognition area.

[0061] The above solutions can improve the accuracy, efficiency and automation level of the automated dispensing system, thereby optimizing the production process and improving production efficiency.

[0062] In the embodiment of the present invention, the ratio of each pixel to the actual distance can also be obtained through a reference object calibration method.

[0063] In an embodiment of the present invention, the specific method of detecting the glue status is as follows:

[0064] B1. Calculate the linear distance ratio of L3 relative to L2 and L1 in each slice based on the overall image area captured by the camera, and create a standard reference value ZZ template.

[0065] B2. During the detection process, calculate the current distance ratio of L3 to L2 and L1 and the reference value ZZ;

[0066] B3. Compare the current line distance ratio of L3 relative to L2 and L1 with the reference value ZZ. If the difference exceeds the preset threshold, it is determined to be a suspected glue break;

[0067] B4. Count the number of consecutive suspected glue-broken slices. If the number exceeds the set threshold, it is considered that the glue strip is interrupted at this moment.

[0068] The above solution can realize accurate and real-time monitoring of the glue status, timely detect whether the glue line is abnormally interrupted, and automatically identify the interruption position and distance so as to quickly replenish the glue and ensure product quality.

[0069] In an embodiment of the present invention, the specific method of creating a standard reference value ZZ template is as follows:

[0070] Determine the overall area and direction of the camera's image capture based on the motion trajectory and speed of the dispensing equipment as it runs the entire dispensing path;

[0071] Slice the entire image area and calculate the line distance ratio of L3 relative to L2 and L1 in each slice;

[0072] Remove p% of the extreme value slices and retain 1-2*p% of the slices in the middle part. Calculate the sum of the line distance ratios of this part and divide it by the number of remaining slices to obtain the average value ZZ of these values ​​to form the reference ZZ template.

[0073] The above solution can accurately capture and analyze the line distance ratio of L3 relative to L2 and L1 in each slice during the dispensing process. At the same time, by removing extreme values ​​and retaining the middle slices, the influence of outliers is reduced, and a more reliable reference value ZZ template is obtained. This helps to optimize the accuracy and stability of the dispensing process and improve product quality and production efficiency. Example

[0074] like Figure 2 and Figure 3 As shown, an embodiment of the present invention is a glue coating detection device based on laser projection video recognition, the detection device includes a slide, a shell, a glue dispensing movement mechanism, a glue dispensing device and a glue strip detection module; the slide includes an X slide and a Y slide, which are used to position and move the glue dispensing mechanism and place the shell; the shell provides a protection and support structure to ensure the safe operation of the entire device; the glue dispensing movement mechanism is used to drive the glue dispensing module to move to the set glue coating position; the glue dispensing module includes a glue dispensing controller and a glue dispensing nozzle, which are used to apply glue to the workpiece; the glue strip detection module is used to monitor and analyze the glue strips formed during the glue dispensing process in real time.

[0075] The Y slide is used to support the X slide and place the shell. The dispensing mechanism installed on the X slide pushes the dispensing module to move to the set gluing position. The dispensing module controls the flow of glue through the dispensing controller. The glue flows out through the dispensing nozzle to form a semi-cylindrical glue strip. The glue strip detection module installed on the dispensing module monitors and analyzes the semi-cylindrical glue strip formed during the dispensing process in real time.

[0076] like Figure 4As shown, in an embodiment of the present invention, the rubber strip detection module includes a line laser projection device, a narrow-band filter, a camera, a refractive lens, a servo mechanism and a control unit; the line laser projection device is used to emit light to illuminate the surface of the rubber strip; the narrow-band filter is installed on the camera lens to filter out light other than the laser line; the camera is used to capture the projection image of the laser line on the rubber strip; the refractive lens is used to change the direction of the light path; the servo mechanism includes a servo motor and a servo gear, which are used to control the line laser projection device to rotate around the dispensing nozzle; the control unit is responsible for the control and coordination of the entire system, controls the operation of the laser projection device, narrow-band filter, camera, servo mechanism, etc., and performs image processing, data analysis and result output.

[0077] The line laser projection device emits light of a specific shape to illuminate the surface of the rubber strip. The narrow-band filter installed on the camera lens filters the light. The camera captures the projection image of the laser line on the rubber strip. The refractive lens is used to change the direction of the light path and project the laser line light and shadow projected on the rubber strip onto the camera. The servo mechanism controls the line laser projection device to rotate around the dispensing nozzle to ensure that the detection direction of the equipment is consistent with the direction of the dispensing head. The control unit processes the video images captured by the camera and controls the servo mechanism to make corresponding movements.

[0078] The above solution can achieve accurate and real-time monitoring of the rubber strips, provide data support for quality control, and realize automated adjustment, thereby improving production efficiency and product quality.

[0079] In an embodiment of the present invention, the servo gear can be adjusted between 0°-180°; in the motion trajectory of dispensing, the dispensing device swings back and forth between -90° and 90° to keep the direction always parallel to the motion direction of the dispensing head.

[0080] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A glue coating detection method based on laser projection video recognition, characterized in that: The following steps are involved: S1. Laser projection line acquisition and recognition: By tilting the laser projection device and adding a refractive lens, the light path is adjusted to achieve the acquisition and recognition of the laser projection line; S2. Determine the recognition area: Calculate the ratio of the effective segmentation distance of each image to the camera pixel and the actual distance, and define the recognition area. The specific method of defining the recognition area is as follows: A1. Calculate the time required for the entire gluing process and the total number of frames captured by the camera based on the dispensing speed, total length of the dispensing line, and the camera frame rate. A2. Calculate the effective segmentation distance for each image based on the total length of the dispensing line and the total number of frames captured by the camera. A3. Based on the principle of monocular imaging, the ratio of each camera pixel to the actual distance is inferred; A4. Based on the dispensing head's forward direction and the effective segmentation distance of each image, the pixel values ​​within a certain range above and below the midpoint are intercepted as the recognition area; S3. Binarization of the identified area: A narrowband filter highlights the light stripes formed by the laser light source, performs binary processing on the highlighted light stripes, and uses brightness thresholds to distinguish between light and darkness, generating multiple area images. S4. Detect glue status: Analyze the laser line refraction and line distance ratio to determine whether the glue is broken, and determine whether the glue strip is interrupted based on the number of consecutive suspected broken glue slices.

2. The method for gluing detection based on laser projection video recognition according to claim 1, characterized in that: The specific method of adjusting the optical path to realize the collection and recognition of the laser projection line is as follows: T1. By adjusting the relative position of the two refractive lenses, the angle of the refractive lenses can be adjusted, thereby adjusting the distance between the laser projection lines L1 and L3; T2. Install mapping line L2 between L1 and L3 to ensure that L1, L2 and L3 remain parallel. T3. Set the ratio of the distance between the line L1 and L2 to the distance between the line L2 and L3 to PP, where PP is a constant not equal to 1. Adjust PP to ensure that the projection line spacing meets the specific requirements of different strip widths; T4. By installing a camera equipped with a narrow-band filter, the laser projection lines formed on the product casing can be collected and identified.

3. The method for gluing detection based on laser projection video recognition according to claim 1, characterized in that: The ratio of each pixel to the actual distance can also be obtained through a reference object calibration method.

4. The method for gluing detection based on laser projection video recognition according to claim 2, characterized in that: The specific method of detecting the glue status is as follows: B1. Calculate the linear distance ratio of L3 relative to L2 and L1 in each slice based on the overall image area captured by the camera, and create a standard reference value ZZ template. B2. During the detection process, calculate the current distance ratio of L3 to L2 and L1 and the reference value ZZ; B3. Compare the current line distance ratio of L3 relative to L2 and L1 with the reference value ZZ. If the difference exceeds the preset threshold, it is determined to be a suspected glue break; B4. Count the number of consecutive suspected glue-broken slices. If the number exceeds the set threshold, it is considered that the glue strip is broken.

5. The method for gluing detection based on laser projection video recognition according to claim 4, characterized in that: The specific method of creating the standard reference value ZZ template is as follows: Determine the overall area and direction of the camera's image capture based on the motion trajectory and speed of the dispensing equipment as it runs the entire dispensing path; Slice the entire image area and calculate the line distance ratio of L3 relative to L2 and L1 in each slice; Remove p% of the extreme value slices and retain 1-2*p% of the slices in the middle part. Calculate the sum of the line distance ratios of this part and divide it by the number of remaining slices to obtain the average value ZZ of these values ​​to form the reference ZZ template.

6. A glue coating detection device based on laser projection video recognition, used to implement the glue coating detection method based on laser projection video recognition according to any one of claims 1 to 5, characterized in that: The detection device includes a slide, a housing, a dispensing movement mechanism, a dispensing device, and a glue strip detection module; the slide, including an X slide and a Y slide, is used to position and move the dispensing mechanism and place the housing; the housing provides a protective and support structure to ensure the safe operation of the entire device; the dispensing movement mechanism is used to drive the dispensing module to move to the set glue application position; The dispensing module includes a dispensing controller and a dispensing nozzle, which are used to apply glue to the workpiece; the glue strip detection module is used to monitor and analyze the glue strips formed during the dispensing process in real time.

7. The glue coating detection device based on laser projection video recognition according to claim 6 is characterized in that: The rubber strip detection module includes a line laser projection device, a narrowband filter, a camera, a refractive lens, a servo mechanism and a control unit; the line laser projection device is used to emit light to the surface of the rubber strip; the narrowband filter is installed on the camera lens to filter out light other than the laser line; the camera is used to capture the projected image of the laser line on the rubber strip; the refractive lens is used to change the direction of the light path; the servo mechanism includes a servo motor and a servo gear, which are used to control the line laser projection device to rotate around the dispensing nozzle; the control unit is responsible for the control and coordination of the entire system.

8. The glue coating detection device based on laser projection video recognition according to claim 7 is characterized in that: The servo gear can be adjusted between 0° and 180°.

Images