Appearance defect detection device for light-emitting logo of automobile

Abstract

The invention discloses an automobile luminous logo appearance defect detection device, and relates to the technical field of automobile logo appearance detection tools. Through the cooperation of the damage detection mechanism and the brightness detection mechanism, the appearance, damage and brightness defects of the logo can be conveniently and comprehensively detected; the analysis module processes the gray block, draws the contour and marks the abnormity, accurately positions the areas where defects possibly exist on the surface of the logo, transmits the position information of the areas to the execution module, and when the execution module controls laser detection, the abnormal areas are scanned preferentially, so that omission or misjudgment caused by blind laser scanning is avoided, and the detection efficiency is improved. Laser detection is more targeted, and the accuracy of defect detection is improved; according to the technical scheme, the cleaning operation is performed firstly, and then the laser scanning is performed, so that the bulge caused by the attachment or the bulge formed by the material of the logo can be effectively distinguished, the accuracy of bulge defect judgment is ensured, and unnecessary loss caused by misjudgment is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile logo appearance detection tools, and in particular to an automobile luminous logo appearance defect detection device. Background Art

[0002] As an important component of vehicle brand identification and exterior decoration, the appearance quality of automotive luminous logos directly affects the brand image and market competitiveness of the vehicle. With consumers' increasing demands for vehicle quality and the automotive industry's increasingly stringent quality control measures, higher standards are being set for the inspection of appearance defects of automotive luminous logos. This requires not only the detection of physical defects such as surface damage and scratches, but also the precise detection of optical defects such as uneven brightness and abnormal luminous areas.

[0003] To improve inspection efficiency, some companies have introduced simple automated inspection equipment. However, existing equipment still has significant limitations. For example, traditional inspection equipment often uses a single inspection method and can only complete single-item inspections for surface scratches or brightness. It cannot simultaneously detect multi-dimensional defects, requiring multiple handling and inspections, which increases process complexity and inspection time. The positioning mechanism of some equipment has poor adaptability, and the socket cannot be accurately inserted during brightness inspection. In addition, the sorting function of existing equipment is imperfect, and qualified and unqualified products need to be manually sorted after inspection. This fails to achieve integrated inspection and sorting, increasing labor costs and the risk of errors.

[0004] Therefore, the above problems need to be improved. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a device for detecting appearance defects of automobile luminous logos.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solutions: a device for detecting appearance defects of a car luminous logo, comprising a cabinet, an entrance and an exit being formed on one side of the cabinet, a first conveyor belt and a second conveyor belt being installed in the entrance and the exit, respectively, a robotic arm being installed on one side of each of the first and second conveyor belts, a gripping mechanism being installed on the other end of the robotic arm, the robotic arm being located inside the cabinet, and a turntable being installed inside the cabinet, a rotating mechanism being installed at the lower end of the turntable, and a plurality of positioning mechanisms being equidistantly installed on the top surface of the turntable, an industrial camera, a damage detection mechanism, a brightness detection mechanism, and a support rod being sequentially installed around the turntable, and a sorting mechanism being installed on the support rod;

[0007] An intelligent control component is provided inside the cabinet, and the intelligent control component includes an analysis module;

[0008] The analysis module receives the collected data, analyzes the image data, obtains an analysis image based on the abnormal grayscale blocks, compares the grayscale values ​​in the analysis image, draws the outline of the logo, and analyzes the abnormalities within the logo outline. If it is dirty, a cleaning signal is generated and the cleaning signal is passed to the execution module; if it is a concave defect, a concave warning signal is generated and the concave warning signal is passed to the execution module; if it is a convex defect, it is cleaned and tested again. If it is still determined to be abnormal, a convex warning signal is generated and the convex warning signal is passed to the execution module.

[0009] Preferably, the analysis module performs the following steps to analyze the image data:

[0010] S1: collected at the same time The image data is gray-scale processed and segmented according to the size of the pixel block into a number of gray-scale blocks of the same size, and the gray-scale blocks are numbered according to the row and column numbers of the segmented image blocks;

[0011] S2: Get the gray value data of each gray block and calculate The mean grayscale value of the corresponding grayscale block in the image data and standard deviation , to calculate the mean and standard deviation Detect the fluctuation range of gray value data The setting of the grayscale value data that is not within the fluctuation range is marked as abnormal grayscale, and the number of abnormal grayscales detected is Perform statistics, if , then it is determined that the image detected at the corresponding moment is abnormal. is the preset proportional coefficient; after removing abnormal grayscale, the remaining grayscale value data is averaged Calculation of the mean value As the grayscale value data corresponding to the grayscale block;

[0012] S3: collected at the corresponding time The image with the least abnormal grayscale is selected from the image data as the analysis image, and the grayscale value of the corresponding numbered grayscale block in the analysis image is compared with the preset logo grayscale value fluctuation range. If the grayscale value of the corresponding numbered grayscale block is within the preset logo grayscale value fluctuation range, the corresponding grayscale block is determined to be the logo grayscale block;

[0013] S4: Set the grayscale value of the logo grayscale block Grayscale values ​​of adjacent grayscale blocks For comparison, if , then the corresponding logo grayscale block is determined to be the logo boundary grayscale block, and the grayscale difference dividing line between the logo boundary grayscale block and the adjacent grayscale block is the logo contour line. is a preset difference threshold; the logo outline is connected and the drawn outline is compared with the logo outline. If the two are proportional, the outline is determined to be accurate;

[0014] S5: Mark the grayscale blocks with abnormal grayscale values ​​inside the contour line, randomly select a marked grayscale block, and calculate the distance data between the marked grayscale block and the nearest contour line , and the distance data and the number of its nearest contour are passed to the execution module, is the number of the corresponding marked grayscale block.

[0015] Preferably, the analysis module performs the following steps to analyze the laser scanning distance data:

[0016] K1: If , it is determined that there is an abnormality at the detection position, is the preset difference threshold; if , it is determined that the grayscale abnormality at the detection position is caused by dirt, a cleaning signal is generated, and the cleaning signal is transmitted to the execution module. The distance data fed back by scanning, is the distance data at the corresponding position of the standard part;

[0017] K2: If , it is determined that there is a depression at the abnormal detection position; otherwise, it is determined that there is a bulge; when the laser scans the grayscale block with abnormal grayscale value inside the contour line, the laser moves the position within the corresponding range of the grayscale block until Stop moving and store the distance data detected during the movement Record; divide the defect location area into several laser-sized columns and record the number of columns corresponding to the defect , according to the recorded distance data Calculate the defect volume data of the defect location ;

[0018] K3: The total number of columns corresponding to defects on the logo and total defect volume data Perform calculations. If the defect is a depression, compare the corresponding data of the defect. If or , it is determined that the dent defect affects the normal use of the logo, a dent warning signal is generated, and the dent warning signal is transmitted to the execution module;

[0019] K4: If the defect is a bulge, perform a cleaning operation first, then perform laser scanning again, and then compare the corresponding data of the defect after the scanning is completed; if or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the preset defect quantity threshold, is the preset defect volume threshold;

[0020] K5: If and , then perform laser scanning on the position of the logo that is not marked with abnormalities, and then compare the total number and total volume of the corresponding columns of convex and concave again. If or or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the total volume of the secondary detection bulge, is the preset secondary detection volume threshold.

[0021] Preferably, transparent windows are hinged on the four sides of the cabinet, a display screen is mounted on one side of the cabinet via a fixing rod, and an alarm is mounted on one end of the top surface of the cabinet.

[0022] Preferably, the rotating mechanism includes a first motor located at the lower end of the turntable, a first gear is installed on the output shaft of the first motor, a second gear is meshed and transmitted on one side of the first gear, the second gear is fixedly connected to the turntable shaft, and the turntable is made of transparent and highly light-transmitting material.

[0023] Preferably, the grasping mechanism includes a mounting platform installed at one end of the robotic arm and a second motor installed on one side of the mounting platform, the output end of the second motor is installed with a first bidirectional screw rod through a coupling, two clamping rods are threadedly connected to the first bidirectional screw rod, and a position sensor is installed on the bottom surface of the mounting platform.

[0024] Preferably, the damage detection mechanism includes two laser scanners and two light sources, wherein one of the laser scanners is located above the light source, and the other laser scanner is located below the light source.

[0025] Preferably, the positioning mechanism includes a first positioning plate and a second positioning plate installed on the turntable, the top surface of the first positioning plate is fixedly connected to the limiting plate, one end of the first positioning plate is installed with a third motor, the output end of the third motor is installed with a one-way screw through a coupling, the front end of the one-way screw is rotatably connected to the first positioning plate, the rear end of the one-way screw is threadedly connected to the second positioning plate, and the other end of the first positioning plate is fixed with two guide rods, both of the two guide rods are inserted into the second positioning plate, a fourth motor is installed on one side of the second positioning plate, the output end of the fourth motor is installed with a second bidirectional screw through a coupling, two splints are threadedly connected to the second bidirectional screw, and the first positioning plate and the second positioning plate are made of transparent and highly light-transmitting materials, an inverted trapezoidal guide groove is provided on the limit plate, a guide angle is provided at the notch of the guide groove, and a limit ring is installed in the guide groove, and a plurality of pressure sensors are installed on the limit ring.

[0026] Preferably, the brightness detection mechanism includes a mounting plate fixed to the inner wall of the cabinet, a sliding groove is provided on the mounting plate, a mounting block is slidably installed in the sliding groove, a power socket is provided on the mounting block, and a first electric push rod is installed on the top surface of the mounting plate, and the telescopic end of the first electric push rod is connected to the top of the mounting block.

[0027] Preferably, the sorting mechanism includes a paint bucket installed at the top of the support rod, a pressure pump is installed at one end of the paint bucket, an atomizer is installed at the other end of the pressure pump, the atomizer is located inside the top of the support rod, and a second electric push rod is installed on one side of the top of the support rod, and a plastic-sprayed plate is installed at the telescopic end of the second electric push rod, different digital slots are opened through the plastic-sprayed plate, and the plastic-sprayed plate is slidably installed inside the top of the support rod.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. The coordination of an industrial camera, a damage detection mechanism, and a brightness detection mechanism facilitates comprehensive detection of logo appearance, damage, and brightness defects, improving the comprehensiveness and accuracy of detection and enabling multi-dimensional defect detection. Furthermore, the coordination of a sorting mechanism, a turntable, and a second conveyor belt facilitates classification, marking, and transportation of logos with different test results, improving sorting efficiency and enabling automatic sorting. Furthermore, the installation of a positioning mechanism facilitates securing the plug within the limiting ring and facilitating insertion into the power socket for power supply. This ultimately resolves the issues of poor detection accuracy, subsequent troublesome sorting, and difficulty aligning the plug during brightness detection associated with existing devices.

[0030] 2. The analysis module processes grayscale blocks, draws outlines, and marks anomalies to accurately locate areas on the logo surface where defects may exist. The location information of these areas is then passed to the execution module. When the execution module controls laser detection, it prioritizes scanning these abnormal areas, avoiding omissions or misjudgments caused by blind laser scanning. This makes laser detection more targeted and improves the accuracy of defect detection.

[0031] 3. By performing a cleaning operation before laser scanning, it is possible to effectively distinguish whether the protrusion is caused by an attached object or a protrusion formed by the logo's own material. If the protrusion disappears after cleaning, it indicates that it is a protrusion caused by the attached object, avoiding misjudgment of it as a quality defect of the logo itself. If the protrusion still exists after cleaning, it can be determined to be a protrusion of the same material as the logo, ensuring the accuracy of the protrusion defect judgment and reducing unnecessary losses caused by misjudgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0033] Figure 1 This is a schematic diagram of the overall three-dimensional structure of the device proposed in the present invention;

[0034] Figure 2 This is a schematic diagram of the internal structure of the device proposed by the present invention;

[0035] Figure 3 This is a schematic diagram of the rotating mechanism structure proposed by the present invention;

[0036] Figure 4 This is a schematic diagram of the structure of the gripping mechanism proposed in the present invention;

[0037] Figure 5 This is a schematic structural diagram of the damage detection mechanism proposed in the present invention;

[0038] Figure 6 This is a schematic diagram of the positioning mechanism structure proposed by the present invention;

[0039] Figure 7 This is a schematic diagram of the cross-sectional structure of the positioning mechanism proposed by the present invention;

[0040] Figure 8 This is a schematic structural diagram of the brightness detection mechanism proposed in the present invention;

[0041] Figure 9 This is a schematic diagram of the sorting mechanism structure proposed by the present invention;

[0042] Figure 10 This is a schematic diagram of the cross-sectional structure of the sorting mechanism proposed by the present invention;

[0043] Figure 11 The present invention proposes Figure 4 A magnified schematic diagram of the structure of part A in the middle;

[0044] Figure 12 This is a flow chart of the system proposed in the present invention.

[0045] Serial numbers in the figure: 1. cabinet; 2. first conveyor belt; 3. second conveyor belt; 4. robotic arm; 5. industrial camera; 6. support rod; 7. transparent window door; 8. alarm; 9. display screen; 10. turntable; 11. first motor; 12. first gear; 13. second gear; 14. second motor; 15. first bidirectional lead screw; 16. clamping rod; 17. laser scanner; 18. light source; 19. first positioning plate; 20. second positioning plate; 21. limit plate; 22. third motor; 23. one-way lead screw; 24. fourth motor; 25. second bidirectional lead screw; 26. clamping plate; 27. limit ring; 28. pressure sensor; 29. ​​mounting block; 30. power socket; 31. first electric push rod; 32. paint bucket; 33. pressure pump; 34. atomizer; 35. spray plate; 36. second electric push rod. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0047] Example 1: See Figure 1-11The present invention provides a device for detecting defects in the appearance of a car luminous logo, comprising a cabinet 1, through which the logo can be detected; an entrance and an exit are provided on one side of the cabinet 1, and a first conveyor belt 2 and a second conveyor belt 3 are installed in the entrance and the exit respectively, and the first conveyor belt 2 and the second conveyor belt 3 facilitate loading and unloading with a robotic arm 4; a robotic arm 4 is installed on one side of the first conveyor belt 2 and the second conveyor belt 3, and a gripping mechanism is installed at the other end of the robotic arm 4, and the robotic arm 4 is located inside the cabinet 1, and a turntable 10 is installed inside the cabinet 1, and the turntable 10 facilitates the detection parts to reach different detection links respectively; a rotating mechanism is installed at the lower end of the turntable 10, and a plurality of positioning mechanisms are installed at equal intervals on the top surface of the turntable 10, and industrial cameras 5 are installed in sequence on the circumference of the turntable 10. , damage detection mechanism, brightness detection mechanism and support rod 6, a sorting mechanism is installed on the support rod 6, and transparent windows 7 are hinged on the four sides of the cabinet 1, through which the internal situation of the cabinet 1 can be easily observed; a display screen 9 is installed on one side of the cabinet 1 through a fixed rod, and the display screen 9 is convenient for displaying detection data; and an alarm 8 is installed at one end of the top surface of the cabinet 1, and the alarm 8 is convenient for warning faults; the rotating mechanism includes a first motor 11 located at the lower end of the turntable 10, which is convenient for driving the first gear 12 to rotate through the first motor 11; a first gear 12 is installed on the output shaft of the first motor 11, and the first gear 12 is convenient for cooperating with the second gear 13 to drive the turntable 10 to rotate; a second gear 13 is meshed with the first gear 12 on one side, and the second gear 13 is meshed with the turntable 10 is fixed to the rotating shaft, and the turntable 10 is made of a transparent and highly light-transmitting material. The grasping mechanism includes a mounting platform installed at one end of the robotic arm 4 and a second motor 14 installed on one side of the mounting platform. The second motor 14 is used to drive the first bidirectional screw 15 to rotate. The output end of the second motor 14 is installed with a first bidirectional screw 15 through a coupling. The first bidirectional screw 15 is used to drive the clamping rod 16 to move inward. Two clamping rods 16 are threadedly connected to the first bidirectional screw 15, and the clamping rod 16 is used to clamp the detection part. A position sensor is installed on the bottom surface of the mounting platform, and the position sensor is used to accurately locate the clamping position. The damage detection mechanism includes two laser scanners 17 and two light sources 18. The light source 18 and the light transmittance of the detection part make the damaged part more obvious. One of the laser scanners 17 is located at the upper end of the light source 18, and the other laser scanner 17 is located at the lower end of the light source 18. The positioning mechanism includes a first positioning plate 19 and a second positioning plate 20 installed on the turntable 10, and the first positioning plate 19 and the second positioning plate 20 are respectively used to facilitate the installation of the third motor 22 and the fourth motor 24; the top surface of the first positioning plate 19 is fixedly connected to the limiting plate 21, and the limiting plate 21 is used to facilitate the installation of the limiting ring 27; a third motor 22 is installed at one end of the first positioning plate 19, and the third motor 22 is used to facilitate the driving of the one-way screw rod 23 to rotate; a one-way screw rod 23 is installed at the output end of the third motor 22 through a coupling, and the one-way screw rod 23 is used to facilitate the movement of the second positioning plate 20; the front end of the one-way screw rod 23 is rotatably connected to the first positioning plate 19.

[0048] The second positioning plate 19 and the second positioning plate 20 are connected by a screw thread to each other, and the first positioning plate 19 and the second positioning plate 20 are connected by a screw thread to each other, and the second ... At the accurate position; the brightness detection mechanism includes a mounting plate fixed to the inner wall of the cabinet 1, a slide groove is provided on the mounting plate, a mounting block 29 is slidably installed in the slide groove, a power socket 30 is provided on the mounting block 29, and a first electric push rod 31 is installed on the top surface of the mounting plate, which facilitates the movement of the mounting block 29 by the first electric push rod 31; the telescopic end of the first electric push rod 31 is connected to the top of the mounting block 29, and the sorting mechanism includes a paint bucket 32 ​​installed at the top of the support rod 6, and the paint bucket 32 ​​is A pressure pump 33 is installed at one end, which facilitates the cooperation with the atomizer 34 for spraying paint; an atomizer 34 is installed at the other end of the pressure pump 33, and the atomizer 34 is located inside the top end of the support rod 6, and a second electric push rod 36 is installed on one side of the top end of the support rod 6, which facilitates changing the shape of the paint spraying; a plastic spraying plate 35 is installed at the telescopic end of the second electric push rod 36, and different digital slots are opened on the plastic spraying plate 35, and the plastic spraying plate 35 is slidably installed inside the top end of the support rod 6.

[0049] Example 2: See Figure 12 , the cabinet 1 is provided with an intelligent control component, which includes an acquisition module, an analysis module and an execution module;

[0050] An acquisition module acquires the image data acquired by the industrial camera 5 and the distance data acquired by the laser scanner 17, and transmits the acquired data to the analysis module;

[0051] The analysis module receives the collected data, analyzes the image data, obtains an analysis image based on the abnormal grayscale blocks, compares the grayscale values ​​in the analysis image, draws the outline of the logo, and analyzes the abnormalities within the logo outline. If it is dirty, a cleaning signal is generated and transmitted to the execution module; if it is a concave defect, a concave warning signal is generated and transmitted to the execution module; if it is a convex defect, it is cleaned and tested again. If it is still determined to be abnormal, a convex warning signal is generated and transmitted to the execution module;

[0052] The image data collected by industrial camera 5 are arranged in the order of collection time, and the images collected at the same time are sorted. The image data is gray-scale processed and divided into several gray-scale blocks of the same size according to the size of the pixel block. The gray-scale blocks are numbered according to the number of rows and columns where the image blocks are located after segmentation. The gray-scale value data of each gray-scale block is obtained and calculated. The mean grayscale value of the corresponding grayscale block in the image data and standard deviation , to calculate the mean and standard deviation Detect the fluctuation range of gray value data The setting of the grayscale value data that is not within the fluctuation range is marked as abnormal grayscale, and the number of abnormal grayscales detected is Perform statistics, if , then it is determined that the image detected at the corresponding moment is abnormal, and the acquisition moment is marked as an abnormal moment. is the preset proportional coefficient; after removing abnormal grayscale, the remaining grayscale value data is averaged Calculation of the mean value As the grayscale value data corresponding to the grayscale block;

[0053] In practical applications, It is usually set between 0.3 and 0.5. The specific value needs to be calibrated according to the logo type and the detection environment. For example, for a metal logo with a smooth surface, the local abnormal grayscale caused by reflection is less. It can be set to 0.3; for plastic logos, the surface is easily contaminated with dust, and the abnormal grayscale ratio may be higher. It can be increased to 0.4-0.5; the setting is based on a large number of sample statistics: by analyzing the image data of thousands of qualified logos, it is found that under normal circumstances, the proportion of abnormal grayscale rarely exceeds 30%, so " The judgment criteria of " can effectively filter out a small amount of anomalies caused by accidental noise (such as camera sensor error, ambient light interference); if If the value is too low (such as 0.2), a normal image with slight noise may be misjudged as abnormal; if the value is too high (such as 0.6), a large number of abnormal grayscales caused by severe dirt or defects may be missed, affecting the accuracy of subsequent detection. The number of images to be collected Linkage adjustment, when Increase (such as from 5 to 10), can be appropriately reduced To balance detection efficiency and avoid triggering overall judgment due to accidental anomalies in a single image;

[0054] Collected at the corresponding time The image with the least abnormal grayscale is selected from the image data as the analysis image, and the grayscale value of the corresponding numbered grayscale block in the analysis image is compared with the preset logo grayscale value fluctuation range. If the grayscale value of the corresponding numbered grayscale block is within the preset logo grayscale value fluctuation range, the corresponding grayscale block is determined to be the logo grayscale block; the grayscale value of the logo grayscale block is Grayscale values ​​of adjacent grayscale blocks For comparison, if , then the corresponding logo grayscale block is determined to be the logo boundary grayscale block, and the grayscale difference dividing line between the logo boundary grayscale block and the adjacent grayscale block is the logo contour line. is a preset difference threshold; the logo outline is connected and the drawn outline is compared with the logo outline. If the two are proportional, the outline is determined to be accurate;

[0055] The grayscale value fluctuation range of the preset logo is determined through strict sample statistics and calibration. Image data of at least 500 qualified logos are collected, and through grayscale value distribution analysis, the grayscale mean of all samples is taken as the basic range ± 3 times the standard deviation. For example, if the grayscale mean of a certain type of chrome-plated logo is 180 and the standard deviation is 20, the fluctuation range is set to 120-240.

[0056] The range needs to be dynamically adjusted for different materials: Metal logos have a higher grayscale range (e.g., 150-255) due to their strong reflectivity; matte plastic logos have a lower grayscale range (e.g., 80-180); logos with patterns require region-specific range settings (e.g., 60-100 for a red patterned area, 150-220 for a silver bordered area). Furthermore, changes in ambient light can affect grayscale values, so the system requires adaptive calibration: the fluctuation range is automatically updated after every 100 logos are inspected to avoid biased judgments due to light source attenuation. For new logo types, a "learning mode" is used to collect 100 samples and recalculate the range to ensure coverage of their grayscale characteristics.

[0057] Preset difference threshold Used to distinguish the logo boundary and non-border area. Its value must match the grayscale characteristics of the logo and background. For example, the grayscale difference at the boundary of a metal logo on a white background can reach 60-80, so It can be set to 40; while the black plastic logo has a smaller grayscale difference (about 30-40) against a dark background. The threshold value needs to be reduced to 25-30. The core function of this threshold is to capture "grayscale mutation": the grayscale value of the logo boundary will change dramatically due to material or color differences, while the grayscale value of internal defects (such as scratches) will change more slowly (usually less than 20). By setting The maximum grayscale difference of the internal defects can be greater than that of the internal defects to avoid misjudging the internal defects as boundaries. For example, if the maximum grayscale difference of the surface scratches on a certain car model logo is 15, then Setting it to 20 can effectively distinguish; in addition, Need to adapt to the image resolution: The grayscale changes in high-resolution images are more delicate. It can be appropriately reduced (e.g. from 40 to 30) to ensure the accuracy of boundary recognition; low-resolution images need to be increased To reduce misjudgments;

[0058] Mark the grayscale blocks with abnormal grayscale values ​​inside the contour line, randomly select a marked grayscale block, and calculate the distance data between the marked grayscale block and the nearest contour line , and the distance data and the number of its nearest contour are passed to the execution module, is the number of the corresponding marked grayscale block;

[0059] When executing module control for laser detection, the distance data transmitted is given priority. Scan the logo surface at the corresponding position and the number of its nearest contour line, and use the distance data fed back by the scan Passed to the analysis module; the analysis module feeds back the distance data Distance data at the corresponding position with the standard part If the comparison , it is determined that there is an abnormality at the detection position, is the preset difference threshold; if , it is determined that the grayscale abnormality at the detection position is caused by dirt, a cleaning signal is generated, and the cleaning signal is transmitted to the execution module;

[0060] The value of needs to be combined with the accuracy of the laser scanner and the critical size of the logo defect; the typical accuracy of the laser scanner is ±0.05mm, so It needs to be greater than twice the precision value (i.e. 0.1mm) to filter out equipment errors; for different defect types, Different settings are required: the depth of the scratch is usually small (0.05-0.2mm), Set to 0.1mm; if the depth of the concave defect is large (above 0.2mm), It can be set to 0.2mm; the setting is based on industry standards: the allowable depth of surface defects of car logos is usually 0.15mm, so Setting it to 0.15mm can be directly associated with qualified judgment; for example, when the difference between the scanning distance and the standard value exceeds 0.15mm, it is judged as an excessive defect; in addition, Part of the logo that needs to be adapted: A slight deviation in the edge area (such as 0.2mm) may not affect the appearance. It can be relaxed to 0.25mm; the central pattern area has strict requirements. Maintain 0.15mm; the system also needs to have temperature compensation function, because the metal logo will expand and contract due to temperature (about 0.03mm / ℃), and needs to be adjusted according to the real-time temperature , avoid misjudgment caused by environmental factors;

[0061] After receiving the cleaning signal, the execution module controls the cleaning robot arm provided in the detection device to clean the logo according to the corresponding position of the abnormality on the logo;

[0062] like , it is determined that there is a depression at the abnormal detection position; otherwise, it is determined that there is a bulge; when the laser scans the grayscale block with abnormal grayscale value inside the contour line, the laser moves the position within the corresponding range of the grayscale block until Stop moving and store the distance data detected during the movement Record; divide the defect location area into several laser-sized columns and record the number of columns corresponding to the defect , according to the recorded distance data Calculate the defect volume data of the defect location ;

[0063] The total number of bars corresponding to defects on the logo and total defect volume data Calculate (the total number of convexities and concavities). If the defect is a concavity, compare the corresponding data of the defect. If or , it is determined that the dent defect affects the normal use of the logo, a dent warning signal is generated, and the dent warning signal is transmitted to the execution module;

[0064] It is a counting standard based on the defect "column unit", where "laser-sized column" refers to a cylinder with a diameter equal to the laser spot (usually 0.5mm), and each column corresponds to a single scanning point of the laser; The setting should be combined with the logo size and defect visibility: small logos (such as 50mm diameter) have a small surface area, Set to 3 (i.e. the total defect coverage area does not exceed 3×0.5²×π≈2.35mm²); for large logos (e.g. 100mm in diameter), this limit can be relaxed to 5-6. The setting logic is based on the visual characteristics of the human eye: studies have shown that when the total defect area exceeds 0.5% of the logo surface area, it can be clearly perceived by the naked eye, so The total columnar area must be less than or equal to 0.5% of the logo area. For example, if the logo area is 1000mm², 0.5% is 5mm². 5÷(0.5²×π)≈6; In addition, the defect distribution also affects : Concentrated defects (such as three columns in one place) are more noticeable than scattered defects (such as one in three locations). Therefore, the system needs to give additional weight to concentrated defects (for example, two concentrated defects are equivalent to three scattered defects) to ensure that the judgment is consistent with visual perception;

[0065] (First detection) and The difference in (secondary inspection) reflects the flexibility of the inspection strategy; the volume is calculated as "single column volume × number", where single column volume = π × (0.25mm)² × defect depth (determined by the difference in laser scanning distance); Need to be slightly larger (such as 0.12mm³), because the secondary inspection is for unmarked areas, most of which are minor defects, appropriate relaxation can reduce unnecessary rework; for key areas, The threshold should be strictly 0.05mm³, and can be relaxed to 0.15mm³ in non-critical areas. In addition, the volume weights of different defect types are different: convex defects are easy to wear, and the volume threshold is 20% lower than that of concave defects (such as convex 0.08, concave 0.1), ensuring key control of high-risk defects;

[0066] After receiving the dent warning signal, the execution module controls the buzzer module in the intelligent control component to sound a buzzer warning, and displays "dent warning" on the display screen 9, and displays the number of dents, the corresponding location of the dents, and the corresponding volume of the dents, so that the staff can accurately conduct manual inspection again;

[0067] If the defect is a bulge, perform a cleaning operation first, then perform laser scanning again, and then compare the corresponding data of the defect after the scanning is completed; if or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the preset defect quantity threshold, is the preset defect volume threshold;

[0068] After receiving the dent warning signal, the execution module controls the buzzer module in the intelligent control component to sound a buzzer warning, and displays "protrusion warning" on the display screen 9, and displays the number of protrusions, the corresponding protrusion positions and the corresponding protrusion volumes, so that the staff can accurately conduct manual inspection again;

[0069] like and , then perform laser scanning on the position of the logo that is not marked with abnormalities, and then compare the total number and total volume of the corresponding columns of convex and concave again. If or or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the total volume of the secondary detection bulge, is the preset secondary detection volume threshold.

[0070] Working principle: When the present invention is used, the car luminous logo to be detected is transported to the entrance of the cabinet 1 through the first conveyor belt 2, and the mechanical arm 4 drives the grabbing mechanism to move to the top of the logo. The position sensor in the grabbing mechanism assists in positioning to ensure accurate identification of the logo position. The second motor 14 is started to drive the first bidirectional screw rod 15 to rotate, so that the two clamping rods 16 approach each other and clamp the logo. The mechanical arm 4 transfers the clamped logo to the positioning mechanism of the turntable 10. In the positioning mechanism, the third motor 22 drives the unidirectional screw rod 23 to rotate, driving the second The positioning plate 20 moves along the guide rod, adjusts the distance with the first positioning plate 19, and pulls the logo power cord. At this time, the socket at one end of the power cord contacts the limit ring 27 under the action of the guide groove. At this time, the pressure sensor 28 senses the signal, the third motor 22 stops, and the fourth motor 24 drives the second bidirectional screw 25 to rotate, so that the two clamps 26 clamp the logo. In the rotating mechanism, the first motor 11 starts, driving the first gear 12 to rotate, and through the meshing transmission with the second gear 13, the turntable 10 rotates. The turntable 10 transmits the positioned logo to the industrial camera 5, the damage detection mechanism, the brightness detection mechanism and the sorting mechanism in sequence. The industrial camera 5 captures the image of the logo appearance and preliminarily detects surface defects. The two laser scanners 17, in cooperation with the light source 18, scan the logo from top and bottom to detect physical defects such as surface damage and scratches. The first electric push rod 31 pushes the mounting block 29 to move along the slide groove, so that the power socket 30 is connected to the logo for power supply. The industrial camera 5 captures the brightness data in the illuminated state to detect optical defects such as uneven brightness and abnormal luminous area. After the detection is completed The turntable 10 sends the logo to the sorting mechanism, the pressure pump 33 pressurizes the paint in the paint bucket 32, and atomizes it through the atomizer 34. The second electric push rod 36 pushes the spray plate 35 to slide, so that the corresponding digital slot is aligned with the atomizer 34, and the number is sprayed on the logo to mark different test results. The robotic arm 4 grabs the marked logo to the second conveyor belt 3 and sends it out of the cabinet 1. The display screen 9 displays the test process and results in real time to facilitate operator monitoring. If a serious defect or equipment failure is detected, the alarm 8 will sound an alarm to remind timely processing.

[0071] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A device for detecting appearance defects of a car luminous logo, comprising a cabinet (1), characterized in that: An entrance and an exit are provided on one side of the cabinet (1), and a first conveyor belt (2) and a second conveyor belt (3) are installed in the entrance and the exit respectively. A mechanical arm (4) is installed on one side of each of the first conveyor belt (2) and the second conveyor belt (3), and a gripping mechanism is installed at the other end of the mechanical arm (4). The mechanical arm (4) is located inside the cabinet (1), and a turntable (10) is installed inside the cabinet (1). A rotation mechanism is installed at the lower end of the turntable (10), and a plurality of positioning mechanisms are installed at equal intervals on the top surface of the turntable (10). An industrial camera (5), a damage detection mechanism, a brightness detection mechanism, and a support rod (6) are installed in sequence around the turntable (10), and a sorting mechanism is installed on the support rod (6); An intelligent control component is provided inside the cabinet (1), and the intelligent control component includes an analysis module; The analysis module receives the collected data, analyzes the image data, obtains an analysis image based on the abnormal grayscale blocks, compares the grayscale values ​​in the analysis image, draws the outline of the logo, and analyzes the abnormalities within the logo outline. If it is dirty, a cleaning signal is generated and the cleaning signal is passed to the execution module; if it is a concave defect, a concave warning signal is generated and the concave warning signal is passed to the execution module; if it is a convex defect, it is cleaned and tested again. If it is still determined to be abnormal, a convex warning signal is generated and the convex warning signal is passed to the execution module.

2. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The analysis module performs the following steps to analyze image data: S1: collected at the same time The image data is gray-scale processed and segmented according to the size of the pixel block into a number of gray-scale blocks of the same size, and the gray-scale blocks are numbered according to the row and column numbers of the segmented image blocks; S2: Get the gray value data of each gray block and calculate The mean grayscale value of the corresponding grayscale block in the image data and standard deviation , to calculate the mean and standard deviation Detect the fluctuation range of gray value data The setting of the grayscale value data that is not within the fluctuation range is marked as abnormal grayscale, and the number of abnormal grayscales detected is Perform statistics, if , then it is determined that the image detected at the corresponding moment is abnormal. is the preset proportional coefficient; after removing abnormal grayscale, the remaining grayscale value data is averaged Calculation of the mean value As the grayscale value data corresponding to the grayscale block; S3: collected at the corresponding time The image with the least abnormal grayscale is selected from the image data as the analysis image, and the grayscale value of the corresponding numbered grayscale block in the analysis image is compared with the preset logo grayscale value fluctuation range. If the grayscale value of the corresponding numbered grayscale block is within the preset logo grayscale value fluctuation range, the corresponding grayscale block is determined to be the logo grayscale block; S4: Set the grayscale value of the logo grayscale block Grayscale values ​​of adjacent grayscale blocks For comparison, if , then the corresponding logo grayscale block is determined to be the logo boundary grayscale block, and the grayscale difference dividing line between the logo boundary grayscale block and the adjacent grayscale block is the logo contour line. is a preset difference threshold; the logo outline is connected and the drawn outline is compared with the logo outline. If the two are proportional, the outline is determined to be accurate; S5: Mark the grayscale blocks with abnormal grayscale values ​​inside the contour line, randomly select a marked grayscale block, and calculate the distance data between the marked grayscale block and the nearest contour line , and the distance data and the number of its nearest contour are passed to the execution module, is the number of the corresponding marked grayscale block.

3. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The analysis module performs the following steps to analyze the laser scanning distance data: K1: If , it is determined that there is an abnormality at the detection position, is the preset difference threshold; if , it is determined that the grayscale abnormality at the detection position is caused by dirt, a cleaning signal is generated, and the cleaning signal is transmitted to the execution module. The distance data fed back by scanning, is the distance data at the corresponding position of the standard part; K2: If , it is determined that there is a depression at the abnormal detection position; otherwise, it is determined that there is a bulge; when the laser scans the grayscale block with abnormal grayscale value inside the contour line, the laser moves the position within the corresponding range of the grayscale block until Stop moving and store the distance data detected during the movement Record; divide the defect location area into several laser-sized columns and record the number of columns corresponding to the defect , according to the recorded distance data Calculate the defect volume data of the defect location ; K3: The total number of columns corresponding to defects on the logo and total defect volume data Perform calculations. If the defect is a depression, compare the corresponding data of the defect. If or , it is determined that the dent defect affects the normal use of the logo, a dent warning signal is generated, and the dent warning signal is transmitted to the execution module; K4: If the defect is a bulge, perform a cleaning operation first, then perform laser scanning again, and then compare the corresponding data of the defect after the scanning is completed; if or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the preset defect quantity threshold, is the preset defect volume threshold; K5: If and , then perform laser scanning on the position of the logo that is not marked with abnormalities, and then compare the total number and total volume of the corresponding columns of convex and concave again. If or or , it is determined that the concave defect affects the normal use of the logo, a raised warning signal is generated, and the raised warning signal is transmitted to the execution module. is the total volume of the secondary detection bulge, is the preset secondary detection volume threshold.

4. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: Transparent windows (7) are hingedly connected to the four sides of the cabinet (1), a display screen (9) is mounted on one side of the cabinet (1) via a fixing rod, and an alarm (8) is mounted on one end of the top surface of the cabinet (1).

5. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The rotating mechanism comprises a first motor (11) located at the lower end of the turntable (10), a first gear (12) is mounted on the output shaft of the first motor (11), a second gear (13) is meshed and driven on one side of the first gear (12), the second gear (13) is fixedly connected to the rotating shaft of the turntable (10), and the turntable (10) is made of a transparent and highly light-transmitting material.

6. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The gripping mechanism comprises a mounting platform mounted on one end of the robotic arm (4) and a second motor (14) mounted on one side of the mounting platform, wherein a first bidirectional screw rod (15) is mounted on the output end of the second motor (14) via a coupling, two clamping rods (16) are threadedly connected to the first bidirectional screw rod (15), and a position sensor is mounted on the bottom surface of the mounting platform.

7. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The damage detection mechanism comprises two laser scanners (17) and two light sources (18), wherein one of the laser scanners (17) is located above the light source (18), and the other laser scanner (17) is located below the light source (18).

8. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The positioning mechanism comprises a first positioning plate (19) and a second positioning plate (20) mounted on the turntable (10), the top surface of the first positioning plate (19) is fixedly connected to a limit plate (21), one end of the first positioning plate (19) is mounted with a third motor (22), the output end of the third motor (22) is mounted with a one-way screw rod (23) via a coupling, the front end of the one-way screw rod (23) is rotatably connected to the first positioning plate (19), the rear end of the one-way screw rod (23) is threadedly connected to the second positioning plate (20), and the other end of the first positioning plate (19) is fixedly connected to two guide rods, both of which are plugged into the first Inside the second positioning plate (20), a fourth motor (24) is installed on one side of the second positioning plate (20), and a second bidirectional screw (25) is installed on the output end of the fourth motor (24) through a coupling. Two clamping plates (26) are threadedly connected to the second bidirectional screw (25), and the first positioning plate (19) and the second positioning plate (20) are both made of transparent and highly light-transmitting materials. An inverted trapezoidal guide groove is provided on the limiting plate (21), and a guide angle is provided at the notch of the guide groove. A limiting ring (27) is installed in the guide groove, and a plurality of pressure sensors (28) are installed on the limiting ring (27).

9. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The brightness detection mechanism comprises a mounting plate fixed to the inner wall of the cabinet (1), a sliding groove is provided on the mounting plate, a mounting block (29) is slidably installed in the sliding groove, a power socket (30) is provided on the mounting block (29), and a first electric push rod (31) is installed on the top surface of the mounting plate, and the telescopic end of the first electric push rod (31) is connected to the top of the mounting block (29).

10. The device for detecting appearance defects of a car luminous logo according to claim 1, characterized in that: The sorting mechanism comprises a paint bucket (32) mounted on the top of a support rod (6), a pressure pump (33) being mounted on one end of the paint bucket (32), an atomizer (34) being mounted on the other end of the pressure pump (33), the atomizer (34) being located inside the top of the support rod (6), and a second electric push rod (36) being mounted on one side of the top of the support rod (6), a plastic spraying plate (35) being mounted on the telescopic end of the second electric push rod (36), different digital slots being provided through the plastic spraying plate (35), and the plastic spraying plate (35) being slidably mounted inside the top of the support rod (6).

Images