Product surface gluing detection method and detection device

Abstract

The invention provides a product surface gluing detection method and device, and the method is used for detecting the gluing quality of the product surface, and comprises the following steps: photographing a plurality of to-be-detected surfaces of a product, and enabling the plurality of to-be-detected surfaces to be imaged on a detection image; dividing the detection image into a plurality of areas, wherein the plurality of areas are in one-to-one correspondence with the plurality of surfaces to be detected; respectively selecting a position needing to be glued in each area on the detection image; the gray average value of the position needing to be glued in each area is detected; the average gray value is compared with a first preset gray value range value, whether the average gray value is within the first preset gray value range value or not is judged, and if the average gray value is within the first preset gray value range value, it is determined that dispensing is completed in the area; and if the gray average value is not within the first preset gray range value, it is determined that dispensing is not completed in the area, so that shooting and detection of the multiple to-be-detected surfaces are completed at a time, and the detection efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of quality inspection technology, and in particular to a method and apparatus for detecting adhesive coating on product surfaces. Background Technology

[0002] Automatic optical inspection has been widely used in industrial production, such as the quality inspection of adhesive coatings on product surfaces. In related technologies, when inspecting the quality of adhesive coatings on product surfaces, each surface to be tested is usually photographed separately and then inspected individually. If there are a large number of surfaces to be tested, the inspection efficiency will be relatively low. Summary of the Invention

[0003] This application discloses a method and apparatus for detecting adhesive coating on product surfaces, which can improve the detection efficiency of adhesive coating quality on product surfaces.

[0004] To achieve the above objectives, this application discloses a method for detecting adhesive coating quality on product surfaces, comprising the following steps:

[0005] Take pictures of multiple surfaces of the product to be tested, and image the multiple surfaces to be tested onto a single test image;

[0006] The detected image is divided into multiple regions, and each of the multiple regions corresponds one-to-one with the multiple surfaces to be tested;

[0007] Select the locations in each of the aforementioned regions that require adhesive application on the detection image;

[0008] The average grayscale value of the location requiring adhesive application in each of the aforementioned regions is detected separately;

[0009] The average grayscale value is compared with a first preset grayscale range value to determine whether the average grayscale value is within the first preset grayscale range value.

[0010] If the average grayscale value is within the first preset grayscale range, then it is determined that the dispensing in that area has been completed.

[0011] If the average grayscale value is not within the first preset grayscale range, it is determined that the dispensing in that area has not been completed.

[0012] Optionally, if it is determined that dispensing has been completed in multiple areas, the method includes:

[0013] Determine whether glue has overflowed in the non-glued area at the boundary between two adjacent areas.

[0014] Optionally, determining whether glue overflow occurs in the non-adhesive area at the boundary between two adjacent areas includes:

[0015] On the detection image, the outer contour lines of two glue images in two adjacent regions are determined respectively;

[0016] Extend the outline segments of the two glue images that are perpendicular to the boundary lines of the two adjacent regions and form an intersection point;

[0017] The portion of the detection image that is close to the intersection point and located outside the outline of the glue image is identified as the non-glue area at the intersection.

[0018] Detect the grayscale value of the non-coated area at the boundary of the detected image;

[0019] The grayscale value of the non-coated area at the boundary is compared with the second preset grayscale range value to determine whether the grayscale value of the non-coated area at the boundary falls within the second preset grayscale range value.

[0020] If the gray value of the non-glued area at the junction is not within the second preset gray value range, it is determined that glue has overflowed in that area.

[0021] If the gray value of the non-adhesive area at the boundary is within the second preset gray value range, it is determined that no glue overflow has occurred in that area.

[0022] Optionally, determining the outer contour lines of two glue images in two adjacent regions on the detected image includes:

[0023] Based on the grayscale difference between the glue and the background in the detected image, the outer contour lines of the glue images in the two adjacent regions are extracted respectively.

[0024] Optionally, if multiple areas are determined to have been dispensed, the method includes:

[0025] The glue image is determined based on the grayscale difference between the glue and the background in the detected image;

[0026] The glue image is checked for broken glue or holes.

[0027] Optionally, detecting whether the glue image has breaks or holes includes:

[0028] Calculate the center point set of the glue image;

[0029] Based on the distribution pattern of the center point set, screen for abrupt change anomalies in the glue image.

[0030] If there is an abrupt change point, it is determined that there is a break in the glue or a hole in the glue image. The abrupt change point is the location of the break in the glue or the hole.

[0031] If no abrupt change points are found, it is determined that there are no broken glue or holes in the glue image.

[0032] Optionally, detecting whether the glue image has breaks or holes includes:

[0033] Identify whether there are areas of alternating light and dark gray levels in the glue image.

[0034] If the alternating light and dark areas exist, it is determined that there are breaks or holes in the glue image, and the dark areas are the locations of the breaks or holes.

[0035] If there are no areas of alternating light and dark, then the glue image is determined to be free of breaks or holes.

[0036] Optionally, if multiple areas are determined to have been dispensed, the method includes:

[0037] The glue image is determined based on the grayscale difference between the glue and the background in the detected image;

[0038] Calculate the glue width of the glue image;

[0039] The adhesive width is compared with a preset adhesive width range to determine whether the adhesive width falls within the preset range.

[0040] If the adhesive width is within the preset adhesive width range, then the adhesive width is determined to meet the requirements;

[0041] If the adhesive width is not within the preset adhesive width range, then the adhesive width is determined to not meet the requirements.

[0042] Optionally, the method for detecting adhesive coating on the product surface includes:

[0043] If the surface to be tested has a flexible board adhesive reinforcement structure, the adhesive coating quality of the flexible board adhesive reinforcement structure is tested.

[0044] Optionally, the quality of the adhesive coating on the flexible board reinforcement structure is inspected, including:

[0045] Select the area where the flexible board adhesive reinforcement structure is located from among the multiple regions;

[0046] The image of the flexible board adhesive reinforcement structure is located in the area where the flexible board adhesive reinforcement structure is located.

[0047] The dimensional parameters of the image of the flexible board adhesive reinforcement structure are measured, and the dimensional parameters are compared with a preset dimensional parameter range to determine whether the dimensional parameters are within the preset dimensional parameter range.

[0048] If the dimensional parameters are within the preset dimensional parameter range, then the adhesive coating quality of the flexible board adhesive reinforcement structure is determined to be qualified.

[0049] If the dimensional parameters are not within the preset dimensional parameter range, the adhesive coating quality of the flexible board adhesive reinforcement structure is determined to be unqualified.

[0050] Optionally, the detection of the adhesive coating quality of the flexible board reinforcement structure includes:

[0051] The image of the flexible board adhesive reinforcement structure is checked to see if there are any breaks in the adhesive or holes.

[0052] Optionally, after taking pictures of the product's test surfaces to image multiple test surfaces onto a single detection image, and before dividing the detection image into multiple regions that correspond one-to-one with the multiple test surfaces, the above method includes:

[0053] Distortion correction is performed on the detection image obtained by taking the picture.

[0054] Optionally, dividing the image into multiple regions includes:

[0055] Determine the size of the selection box for each of the aforementioned regions;

[0056] Each of the selected regions is selected on the detection image based on pixel coordinates using the selected boxes;

[0057] Images of each region within each selected box are acquired to complete the region division.

[0058] This application also provides a detection device, which is applied to the above-mentioned product surface adhesive detection method, and the detection device includes:

[0059] Mounting rack;

[0060] An image acquisition module, comprising a camera and a lens, wherein the camera and lens are mounted on the mounting bracket, and the image acquisition module is used to take pictures of multiple surfaces of the product to be tested.

[0061] A prism module includes a prism mounting base and at least three reflectors mounted on the prism mounting base. The prism mounting base is mounted on the mounting frame along the optical axis of the lens. The at least three reflectors are arranged around the optical axis of the lens and have a preset angle with the optical axis of the lens, so that the at least three reflectors can reflect images of multiple surfaces to be measured to the image acquisition module respectively.

[0062] Optionally, the prism mount is rotatably mounted on the mounting bracket along a direction perpendicular to the optical axis of the lens.

[0063] Optionally, along the optical axis of the lens, the distance between the end face of the lens facing the prism module and the prism module is 4mm to 8mm.

[0064] Optionally, the detection device further includes a light source disposed on the mounting bracket, along the optical axis of the lens, and the light source is disposed between the image acquisition module and the prism module. The light source is a coaxial light source or a strip light source.

[0065] Optionally, any one of at least three of the reflectors is used to reflect an image of the flexible board adhesive reinforcement structure of the product to the image acquisition module.

[0066] Compared with the prior art, the beneficial effects of this application are as follows:

[0067] By capturing images of multiple test surfaces simultaneously into a single inspection image, the need for separate imaging and processing of each surface is eliminated, significantly reducing inspection time. For example, in large-scale product surface adhesive quality inspection, where previously it was necessary to repeatedly set imaging parameters and perform operations for each test surface, this can now be done in one go, significantly accelerating the inspection process and improving efficiency. Furthermore, capturing images of multiple test surfaces simultaneously ensures that all test surfaces are imaged under the same lighting, shooting angle, and environmental conditions, reducing errors caused by variations in shooting conditions and making the inspection results more consistent and comparable. In addition, dividing the inspection image into regions corresponding to each of the multiple test surfaces helps to more accurately analyze and judge each surface, avoiding confusion and error propagation between different images. For example, when detecting minute defects, the specific surface location of the defect can be more accurately located and assessed. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0069] Figure 1 This is a flowchart of a product surface adhesive coating detection method provided in an embodiment of this application;

[0070] Figure 2 This is a flowchart provided in an embodiment of this application after determining that multiple areas have been dispensed;

[0071] Figure 3 This is a flowchart of the process for detecting whether glue overflow has occurred, provided in an embodiment of this application;

[0072] Figure 4This is a flowchart provided in another embodiment of the present application after determining that multiple areas have been dispensed;

[0073] Figure 5 This is a flowchart of the detection process for the presence of broken glue or holes provided in the embodiments of this application;

[0074] Figure 6 This is a flowchart of another embodiment of the present application for detecting the presence of broken glue or holes;

[0075] Figure 7 This is a flowchart of the process for detecting whether the adhesive width meets the requirements, provided in an embodiment of this application.

[0076] Figure 8 This is a flowchart provided in an embodiment of the present application for detecting whether the adhesive coating quality of the flexible board reinforcement structure is qualified;

[0077] Figure 9 This is a flowchart provided in an embodiment of the present application after imaging multiple surfaces to be tested onto a single detection image;

[0078] Figure 10 This is a flowchart of dividing an image into multiple regions provided in an embodiment of this application;

[0079] Figure 11 This is a schematic diagram of the product to be tested provided in an embodiment of this application;

[0080] Figure 12 This is a schematic diagram of the detection device provided in the embodiments of this application;

[0081] Figure 13 This is a top view of the prism module provided in the embodiments of this application.

[0082] Explanation of main figure symbols

[0083] 1-Detection device;

[0084] 20 - Product; 21 - Surface to be tested; 22 - Flexible board adhesive reinforcement structure;

[0085] 100 - Mounting bracket;

[0086] 200 - Image acquisition module; 210 - Camera; 220 - Lens;

[0087] 300 - Prism module; 310 - Prism mount; 320 - Reflector;

[0088] 400 - Light source. Detailed Implementation

[0089] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0090] In this application, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing this application and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation.

[0091] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0092] Furthermore, the terms "installation," "setup," "equipped with," "connection," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral structure; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0093] Furthermore, the terms "first," "second," etc., are primarily used to distinguish different devices, elements, or components whose specific types and structures may be the same or different, and are not intended to indicate or imply the relative importance or quantity of the indicated devices, elements, or components. Unless otherwise stated, "a plurality of" means two or more.

[0094] As mentioned in the background section, in the relevant technologies for inspecting the quality of adhesive coating on product surfaces, each surface to be tested is usually photographed separately and then inspected individually. If there are a large number of surfaces to be tested, the inspection efficiency will be relatively low.

[0095] To address the aforementioned issues, this application provides a method for detecting adhesive coating on product surfaces. By capturing images of multiple test surfaces simultaneously into a single detection image, the method avoids the need for separate imaging and processing of each surface, thus saving detection time.

[0096] The technical solution of this application will be further described below with reference to specific embodiments and accompanying drawings.

[0097] See Figure 1 and Figure 11 This embodiment provides a method for detecting adhesive coating on a product surface, including the following steps:

[0098] S10 takes pictures of multiple test surfaces of the product, so that the multiple test surfaces are imaged on a single test image.

[0099] Specifically, select a suitable photographic device, such as a high-definition camera, and ensure that its resolution and clarity meet the testing requirements. Place and position the product 20 on a specific workbench so that multiple test surfaces 21 are within the camera's field of view. Then, use the camera to take a single photograph of the multiple test surfaces 21 of the product, so that they are imaged in one image.

[0100] S20 divides the detection image into multiple regions, with each region corresponding to a different surface to be tested.

[0101] Specifically, the acquired detection images are analyzed to determine the approximate positions and boundaries of each test surface 21 in the images. Based on the characteristics and requirements of the images, appropriate image editing software or tools, such as Adobe Photoshop or MATLAB, are selected to determine whether to use manual, automatic, or semi-automatic division methods to divide different regions. Each divided region is clearly marked, such as with a number or name.

[0102] Therefore, by capturing images of multiple test surfaces 21 at once and combining them into a single inspection image, the need to photograph and process each test surface 21 individually is avoided, greatly saving inspection time. For example, in large-scale product surface adhesive quality inspection, it was originally necessary to repeatedly set shooting parameters and perform operations for each test surface 21. Now, this can be done in one go, significantly speeding up the inspection process and improving inspection efficiency. Moreover, capturing images of multiple test surfaces 21 at once ensures that all test surfaces 21 are imaged under the same lighting, shooting angle, and environmental conditions, reducing errors caused by changes in shooting conditions and making the inspection results more consistent and comparable. In addition, dividing the inspection image into regions that correspond one-to-one with multiple test surfaces helps to analyze and judge each surface more accurately, avoiding confusion and error transmission between different images. For example, when detecting minor defects, the specific surface location of the defect can be more accurately located and evaluated.

[0103] S30: Select the locations in each region of the detection image where adhesive needs to be applied.

[0104] Specifically, the detection image is enhanced by adjusting contrast and brightness to highlight the features of the areas where glue needs to be applied. Based on the previously defined region markings, the corresponding areas to be selected are accurately located. The features of the areas where glue needs to be applied in each region are studied, such as specific shapes, colors, and textures. The key features used to identify the areas where glue needs to be applied are determined. The areas where glue needs to be applied are selected based on the key features of the areas where glue needs to be applied. The selected areas are carefully checked to ensure that no omissions or errors are made.

[0105] S40, detect the average grayscale value of the area where glue needs to be applied in each region;

[0106] Ensure that the locations where glue needs to be applied in each area have been accurately selected and that the pixel data of these locations can be accurately acquired. Then, convert the color image data of the selected locations into grayscale image data, extract the grayscale values ​​of all pixels corresponding to the locations where glue needs to be applied from the grayscale image, sum all the extracted grayscale values, and count the total number of pixels at the locations where glue needs to be applied. Finally, divide the sum of the grayscale values ​​by the total number of pixels to obtain the average grayscale value.

[0107] S50: Compare the average grayscale value with a first preset grayscale range value to determine whether the average grayscale value is within the first preset grayscale range value.

[0108] If the average grayscale value is within the first preset grayscale range, then it is determined that the dispensing in that area has been completed.

[0109] If the average grayscale value is not within the first preset grayscale range, it is determined that the dispensing in that area has not been completed.

[0110] The first preset grayscale range value is selected based on the actual situation of product 20. The first preset grayscale value is the grayscale value of the area where the qualified product has completed the glue coating.

[0111] Specifically, the average grayscale value calculated for each area where adhesive needs to be applied is compared with the lower and upper limits of a first preset grayscale range. If the average grayscale value is greater than or equal to the lower limit and less than or equal to the upper limit, it is determined that the average grayscale value is within the first preset grayscale range. If the average grayscale value is less than the lower limit or greater than the upper limit, it is determined that the average grayscale value is not within the first preset grayscale range. When the average grayscale value is within the first preset grayscale range, the area is marked as "adhesive application completed". When the average grayscale value is not within the first preset grayscale range, the area is marked as "adhesive application incomplete".

[0112] In one possible embodiment, see Figure 2 If multiple areas need to be covered with adhesive, the methods include:

[0113] S60, determine whether glue has overflowed in the non-glued area at the boundary of two adjacent areas.

[0114] Specifically, the boundary positions of two adjacent regions are clearly defined, and the extent of the boundary area is precisely defined. For example, the boundary area can be determined by coordinate marking or image marking. Then, the local image of the boundary between the two adjacent regions is extracted from the overall detection image. The extracted boundary area image is processed, such as denoising and contrast enhancement, to highlight possible glue overflow features. The characteristics that glue overflow may present, such as color changes, brightness differences, and irregular shapes, are studied. Then, it is decided whether to use a threshold-based detection method, an edge detection algorithm, or a machine learning-based image recognition method. Based on the output results of the selected detection method, it is determined whether glue overflow occurs in the non-glue area at the boundary between the two adjacent regions.

[0115] In one possible embodiment, see Figure 3 Determine whether glue has overflowed in the non-glued area at the boundary between two adjacent areas, including:

[0116] S601, determine the outer contour lines of two glue images in two adjacent regions on the detection image respectively.

[0117] Specifically, the detection image is first processed, such as smoothing, to reduce noise interference, or adjusting the contrast and brightness of the image to make the glue image clearer and more distinguishable. Then, a contour detection algorithm is selected, such as the Canny edge detection algorithm, Sobel operator, Laplacian operator, and other common edge detection algorithms. For the selected edge detection algorithm, appropriate parameters are set, such as threshold and gradient strength. The two adjacent regions are processed in turn, focusing on the glue image in each region. In the selected region, the edge detection algorithm is run to obtain the contour line.

[0118] S602, extend the outline segments of the two glue images that are perpendicular to the boundary line of the two adjacent areas and form an intersection point;

[0119] The boundary line is determined by drawing straight lines or calculating the distribution of pixels in the boundary area. For the outer contour line of each glue image, contour segments perpendicular to the boundary line are selected (the angle between the line segment and the boundary line can be calculated; angles close to 90 degrees are considered perpendicular contour segments). The selected contour segments are extended along the direction perpendicular to the boundary line using a linear extrapolation method. The two extended contour segments may intersect. The coordinates of the intersection point are determined by mathematical calculations (such as solving the equation of a straight line). The calculated intersection point is checked to see if it is reasonable, for example, whether it is within the effective range of the detection image.

[0120] S603, determine the non-adhesive area at the intersection of the detection image and the part that is close to the intersection and outside the outline of the glue image.

[0121] Specifically, taking the intersection point as the center, a circular or rectangular area of ​​appropriate size is determined as the key observation area according to the size and accuracy requirements of the glue image. For each pixel in the defined area, it is determined whether it is outside the outline of the glue image (this can be determined by comparing the pixel coordinates with the coordinate range of the area enclosed by the outline). Among the pixels outside the outline, pixels closer to the intersection point are further filtered out. The set of pixels that are close to the intersection point and outside the outline is determined as the non-glue area at the intersection. The determined non-glue area is marked, for example, by using a specific color or symbol.

[0122] S604, detects the gray value of the non-coated area at the boundary of the detection image;

[0123] Extract all pixels located in the non-coated area from the detection image. If the detection image is a color image, convert the extracted pixels to grayscale values ​​and read the grayscale values ​​of the pixels in the non-coated area one by one.

[0124] S605, compare the grayscale value of the non-coated area at the boundary with the second preset grayscale range value, and determine whether the grayscale value of the non-coated area at the boundary is within the second preset grayscale range value.

[0125] If the gray value of the non-glued area at the boundary is not within the second preset gray value range, it is determined that glue has overflowed in that area.

[0126] If the gray value of the non-adhesive area at the boundary is within the second preset gray value range, it is determined that no glue has overflowed in that area.

[0127] The second preset grayscale range value is the grayscale value of the non-glued area of ​​the qualified product.

[0128] For each grayscale value in the non-adhesive area, it is compared with the lower limit and upper limit of the second preset grayscale range. If the grayscale value is greater than or equal to the lower limit and less than or equal to the upper limit, it is determined that the grayscale value is within the second preset grayscale range. If the grayscale value is less than the lower limit or greater than the upper limit, it is determined that the grayscale value is not within the second preset grayscale range. When there are grayscale values ​​in the non-adhesive area that are not within the second preset grayscale range, it is determined that glue has overflowed in that area. When all grayscale values ​​in the non-adhesive area are within the second preset grayscale range, it is determined that glue has not overflowed in that area.

[0129] In one possible embodiment, determining the outer contour lines of two glue images in two adjacent regions on the detection image includes:

[0130] S6011, extracts the outer contour lines of the glue images of two adjacent regions based on the grayscale difference between the glue and the background in the detected image.

[0131] The detection image is initially observed and analyzed to understand the approximate distribution and differences in grayscale values ​​between the glue and the background. A grayscale histogram of the detection image is generated to more intuitively observe the distribution of grayscale values. Based on the grayscale histogram and the characteristics of the grayscale differences between the glue and the background, a suitable grayscale threshold is determined through experimentation or experience to distinguish between the glue and the background. The outer contour lines of the glue images of two adjacent regions are extracted from the edge detection results.

[0132] In one possible embodiment, see Figure 4 If multiple areas need to be covered with adhesive, the methods include:

[0133] S70, determine the glue image based on the grayscale difference between the glue and the background in the detected image.

[0134] The detection image is initially observed and analyzed to understand the approximate distribution and differences in grayscale values ​​between the glue and the background. A grayscale histogram of the detection image is generated to more intuitively observe the distribution of grayscale values. Based on the grayscale histogram and the characteristics of the grayscale differences between the glue and the background, a suitable grayscale threshold is determined through experimentation or experience to distinguish between the glue and the background, thereby identifying the glue image.

[0135] S80 detects whether there are broken glue or holes in the glue image.

[0136] Extract the outline of the glue image, calculate the length, area and other parameters of the outline, check whether there are discontinuous parts of the outline to determine whether there is a break in the glue, and mark the small and irregularly shaped connected regions in the glue area as possible holes.

[0137] In one possible embodiment, see Figure 5 Detecting whether there are broken glue or holes in the glue image, including:

[0138] S801a, calculates the center point set of the glue image;

[0139] The glue image is segmented into small regions. For example, a uniform grid can be used or an adaptive segmentation can be performed based on the shape and features of the glue image. For each small region, the coordinates of the centroid are calculated based on the distribution of its pixels. Centroids located near the boundaries of the glue image or that do not conform to specific rules are removed. The selected centroid coordinates are combined to form the set of center points of the glue image.

[0140] S802a, based on the distribution pattern of the center point set, screen for abrupt change anomalies in the glue image.

[0141] If there are abrupt change points, it is determined that there are broken glue or holes in the glue image. The abrupt change points are the locations of broken glue or holes.

[0142] If no abrupt change points are found, it is determined that there are no broken glue or holes in the glue image.

[0143] Calculate the mean, variance, median, and other statistical measures of the center point set to understand its central tendency and dispersion. Observe the approximate distribution shape of the center point set in the image coordinate system, such as whether it is linear, circular, or other regular distribution. Based on the distribution characteristics obtained from the analysis, select an appropriate mathematical model to describe the distribution law of the center point set. For example, a normal distribution, uniform distribution, or other suitable probability distribution model can be used. Based on the distribution model and statistical measures, determine a reasonable threshold to determine whether the center point is an outlier. This threshold can be a certain multiple of the standard deviation from the mean, or a certain deviation range from the median. For each point in the center point set, calculate its deviation from the distribution model and compare the deviation with the set threshold. If the deviation of a center point exceeds the threshold, mark it as a jump outlier. If a jump outlier exists, it is determined that there is glue breakage or hole in the glue image, and the jump outlier is recorded as the location of glue breakage or hole. If no jump outlier exists, it is determined that there is no glue breakage or hole in the glue image.

[0144] In one possible embodiment, see Figure 6 Detecting whether there are broken glue or holes in the glue image, including:

[0145] S801b identifies whether there are alternating areas of grayscale brightness in an adhesive image.

[0146] If there are areas of alternating light and dark, it indicates that there are breaks or holes in the glue image, and the dark areas are the locations of the breaks or holes.

[0147] If there are no areas of alternating light and dark, then the glue image is confirmed to be free of breaks or holes.

[0148] Determine the scanning direction in the image, such as horizontal, vertical, or diagonal. Read the grayscale values ​​of pixels row by row or column by column according to the selected direction. For two adjacent pixels, calculate the grayscale difference between them. Based on the characteristics of the image and the expected degree of brightness alternation, set a reasonable grayscale difference threshold. If the grayscale difference between adjacent pixels exceeds the set threshold, it is considered that there is a break in the glue or a hole, and this position is marked. If the grayscale difference between adjacent pixels does not exceed the set threshold, it is considered that there is no break in the glue or a hole.

[0149] In one possible embodiment, see Figure 7 If it is confirmed that adhesive has been applied to multiple areas, the methods include:

[0150] S90, determine the glue image based on the grayscale difference between the glue and the background in the detected image;

[0151] The detection image is initially observed and analyzed to understand the approximate distribution and differences in grayscale values ​​between the glue and the background. A grayscale histogram of the detection image is generated to more intuitively observe the distribution of grayscale values. Based on the grayscale histogram and the characteristics of the grayscale differences between the glue and the background, a suitable grayscale threshold is determined through experimentation or experience to distinguish between the glue and the background, thereby identifying the glue image.

[0152] S100, calculates the glue width of the glue image;

[0153] Determine the direction for measuring the glue width in the glue image, such as horizontal, vertical, or along a specific curve. Select the middle, edge, or multiple evenly distributed locations in the glue image for measurement. Extract the corresponding pixel columns or rows along the selected measurement direction. Select an appropriate grayscale threshold to divide the extracted pixels into glue and non-glue parts. Perform edge detection on the segmented pixels to find the left and right or top and bottom edges of the glue. Measure and calculate the number of pixels between the left and right or top and bottom edges. Based on the ratio between the image resolution and the actual size, convert the number of pixels into the actual physical size to obtain the glue width. For example, if the image resolution is 100 pixels per millimeter and the measured pixel spacing is 200 pixels, then the actual glue width is 2 millimeters. If the shape of the glue image is irregular, measurements need to be taken at multiple locations, and the average value is taken as the final glue width.

[0154] S200 compares the glue width with a preset glue width range to determine whether the glue width falls within the preset range.

[0155] If the adhesive width is within the preset adhesive width range, then the adhesive width meets the requirements.

[0156] If the adhesive width is not within the preset adhesive width range, then the adhesive width does not meet the requirements.

[0157] The calculated glue width value is compared with the lower limit and upper limit of the preset glue width range. If the glue width value is greater than or equal to the lower limit and less than or equal to the upper limit, the glue width is determined to be within the preset glue width range. If the glue width value is less than the lower limit or greater than the upper limit, the glue width is determined to be outside the preset glue width range. When the glue width is within the preset glue width range, the glue width is determined to meet the requirements. When the glue width is outside the preset glue width range, the glue width is determined to not meet the requirements.

[0158] In one possible embodiment, the product surface adhesive coating detection method includes:

[0159] S300: If the surface to be tested has a flexible board adhesive reinforcement structure, test the adhesive coating quality of the flexible board adhesive reinforcement structure.

[0160] Among them, the flexible circuit board adhesive reinforcement structure is an important design in the manufacturing and application of flexible circuit boards (flexible boards). Typically, flexible circuit boards are relatively thin and flexible. In certain specific parts or application scenarios, higher mechanical strength, stability or flexural strength may be required, which requires the use of adhesive for reinforcement. The adhesive is applied or pasted to specific areas to increase the rigidity and strength of the flexible board locally or as a whole.

[0161] In one possible embodiment, see Figure 8 Inspect the adhesive coating quality of flexible board reinforcement structures, including:

[0162] S3001, Select the area where the flexible board adhesive reinforcement structure is located among multiple regions;

[0163] Based on the image data of the flexible board adhesive reinforcement structure, the regions where flexible board adhesive reinforcement structures may exist are identified and marked.

[0164] S3002, Locate the image of the flexible board adhesive reinforcement structure in the area where the flexible board adhesive reinforcement structure is located;

[0165] Analyze the characteristics of the flexible board adhesive reinforcement structure, such as shape, color, and texture, and extract these features from the image. Prepare a standard template image of the flexible board adhesive reinforcement structure, match the extracted features with the template, determine the specific location of the flexible board adhesive reinforcement structure, and mark it to complete the positioning.

[0166] S3003, Measure the dimensional parameters of the image of the flexible board adhesive reinforcement structure, compare the dimensional parameters with the preset dimensional parameter range, and determine whether the dimensional parameters are within the preset dimensional parameter range.

[0167] If the dimensional parameters are within the preset dimensional parameter range, then the adhesive coating quality of the flexible board adhesive reinforcement structure is deemed qualified.

[0168] If the dimensional parameters are not within the preset dimensional parameter range, the adhesive coating quality of the flexible board reinforcement structure is determined to be unqualified.

[0169] The dimensional parameters include, but are not limited to, the glue width, glue length, number of glue breaks, and glue break length of the glue at the flexible board glue reinforcement structure. The preset dimensional parameters are the dimensional parameters of the glue when the glue application quality is qualified.

[0170] In one possible embodiment, detecting the adhesive coating quality of the flexible board adhesive reinforcement structure includes:

[0171] S3004, detect whether there are broken glue or holes in the image of the flexible board adhesive reinforcement structure.

[0172] In one possible embodiment, see Figure 9 After photographing the surfaces of the product to be tested, so that multiple surfaces are imaged onto a single detection image, and before dividing the detection image into multiple regions that correspond one-to-one with multiple surfaces to be tested, the above method includes:

[0173] S101 performs distortion correction on the detection image obtained by taking a picture.

[0174] Specifically, the camera's intrinsic and extrinsic parameters, including focal length, principal point coordinates, radial distortion coefficient, and tangential distortion coefficient, are obtained from the camera's configuration file or through a dedicated calibration procedure. Based on the camera type and distortion characteristics, a suitable distortion model is selected; common models include barrel distortion and pincushion distortion. Using the selected distortion model and the obtained camera parameters, the corresponding position of each pixel in the distorted image in the ideal distortion-free image is calculated, establishing a pixel mapping relationship. For positions in the ideal distortion-free image that do not precisely correspond to pixels in the original distorted image, interpolation algorithms (such as bilinear interpolation, bicubic interpolation, etc.) are used to estimate pixel values. Based on the calculated pixel mapping relationship and interpolation results, a corrected distortion-free detection image is generated. The corrected image is then optimized, such as through smoothing and edge enhancement, to improve image quality.

[0175] In one possible embodiment, see Figure 10 The image is divided into multiple regions, including:

[0176] S201, Define the size of the selection box for each area;

[0177] A detailed analysis is conducted on each region to be selected, including its shape, size, distribution, and potential key elements. Based on factors such as the detection objective, image resolution, and subsequent processing requirements, a reference standard for determining the size of the selection box is established. To ensure complete coverage of the target region while avoiding excessive inclusion of irrelevant parts, a certain margin is reserved when calculating the size. For irregularly shaped regions, the size of their minimum bounding rectangle is calculated. If the region has certain proportional characteristics, the length and width of the selection box are set according to this proportion. Considering factors such as the relative positional relationships between regions and the overall image layout, the initially proposed size is adjusted. Finally, a simulation is performed on a sample image to check whether the proposed selection box size meets the requirements, whether important information is missed, or whether it contains too much irrelevant content. Based on the verification results, the final size of the selection box is determined and recorded.

[0178] S202, using each selection box to select each region on the detection image according to pixel coordinates;

[0179] Based on the starting pixel coordinates of the selection box, accurately locate the corresponding position on the detection image as the starting point for the selection. Starting from the starting point, draw the border pixel by pixel on the image according to the length and width of the selection box. Different colors or line styles can be used to highlight the selection boundary. During the process of drawing the selection boundary, continuously check whether the border completely matches the size of the selection box and whether it accurately covers the target area to ensure that the border of the selection box is completely drawn and accurately selects the corresponding area.

[0180] S203, acquire the images of each region within each selected box to complete the region division.

[0181] See Figures 11 to 13 This application also provides a detection device 1, applied to the above-described product surface adhesive detection method. The detection device 1 includes:

[0182] The system comprises a mounting bracket 100, an image acquisition module 200, and a prism module 300. The image acquisition module 200 includes a camera 210 and a lens 220, which are mounted on the mounting bracket 100. The image acquisition module 200 is used to take pictures of multiple test surfaces 21 of the product 20. The prism module 300 includes a prism mounting base 310 and at least three reflectors 320 mounted on the prism mounting base 310. The prism mounting base 310 is mounted on the mounting bracket 100 along the optical axis of the lens 220. The at least three reflectors 320 are arranged around the optical axis of the lens 220 and have a preset angle with the optical axis of the lens 220, so that the at least three reflectors 320 can reflect images of multiple test surfaces 21 to the image acquisition module 200 respectively.

[0183] Since at least three reflectors 320 are mounted on the prism mounting base 310 in the prism module 300, and these reflectors 320 are arranged around the optical axis of the lens 220 and have a preset angle with the optical axis, when light shines on multiple test surfaces 21 of the product under test 20 (at least four test surfaces 21, three side surfaces and one front facing the lens 220), the light reflected from each surface will be directed at a specific angle to the corresponding reflector 320. Because there is a preset angle between the reflector 320 and the optical axis of the lens 220, these reflectors 320 will change the direction of the light from different test surfaces 21 according to a specific optical path, so that they can enter the lens 220 simultaneously. The lens 220 collects and focuses the light reflected by the reflectors 320 and transmits it to the camera 210. This allows the camera 210 to receive image information from multiple test surfaces 21 at once, enabling simultaneous imaging of multiple surfaces. Furthermore, by reflecting the images of multiple test surfaces 21 to the image acquisition module 200 simultaneously through multiple reflectors 320, images of multiple surfaces can be acquired in a single shot, reducing the number of shots and the time required for equipment adjustment, thus significantly improving detection efficiency. This compact design eliminates the need for separate image acquisition devices for each test surface, saving installation space. Additionally, since the images of multiple test surfaces are captured through the same image acquisition module 200, consistent image acquisition conditions (such as lighting and focal length) are ensured, facilitating subsequent image analysis and processing and improving detection accuracy and consistency.

[0184] In one possible embodiment, see Figure 12 The prism mount 310 is rotatably mounted on the mounting bracket 100 along a direction perpendicular to the optical axis of the lens 220.

[0185] By rotating the prism mounting base 310, the angle between the reflector 320 and the optical axis of the lens 220 can be changed, thereby flexibly adjusting the angle of the reflected light from the reflector 320 to adapt to the test surface 21 at different positions and angles, improving the system's adaptability to different products 20 or testing scenarios. For example, during actual installation and use, various factors may cause the optical path to be less than ideal. By rotating the prism mounting base 310, the optical path can be fine-tuned to obtain a clearer and more accurate image reflection effect, improving the quality of image acquisition. Alternatively, if the position of the test surface 21 of the product 20 changes slightly, or if a different model of product is used for testing, the optical path can be quickly adjusted by rotating the prism mounting base 310 without redesigning and reassembling the entire testing device 1.

[0186] In one possible embodiment, along the optical axis of the lens 220, the distance between the end face of the lens 220 facing the prism module 300 and the prism module 300 is 40mm to 80mm.

[0187] By positioning the end face of lens 220 facing prism module 300 within a distance range of 40mm to 80mm, suitable space is provided for light propagation and refraction. This allows light reflected from prism module 300 to enter lens 220 in a better state, reducing aberrations and distortion. This ensures that the image acquisition module obtains clear and accurate images. Furthermore, the distance between the test surface and prism module 300 may vary for products of different sizes. The 40mm-80mm range provides flexibility to adapt to various product sizes, ensuring that images of the test surface 21 can be effectively acquired under various conditions. In addition, it provides some space for the design and optimization of the optical path. Within this range, the angle and position of reflector 320 can be set more reasonably to achieve accurate reflection and acquisition of images of multiple test surfaces 21.

[0188] In one possible embodiment, see Figure 12 The detection device 1 also includes a light source 400 disposed on the mounting bracket 100. Along the optical axis of the lens 220, the light source 400 is disposed between the image acquisition module 200 and the prism module 300. The light source 400 is a coaxial light source or a strip light source.

[0189] In the detection device 1, a light source 400 is set on the mounting bracket 100 between the image acquisition module 200 and the prism module 300. A coaxial light source or a strip light source is used to provide uniform light to the surface to be tested 21, reducing shadows and uneven lighting. This makes the image acquired by the image acquisition module 200 more uniform in brightness and contrast, which helps to accurately detect and analyze the surface. Furthermore, the coaxial light source is suitable for detecting surfaces with reflective properties, while the strip light source can better illuminate the surface to be tested 21 with complex shapes or large areas, increasing the adaptability of the detection device 1 to the characteristics of different surfaces to be tested 21.

[0190] In one possible embodiment, any one of at least three reflectors 320 is used to reflect an image of the flexible board adhesive reinforcement structure of the product 20 to the image acquisition module 200.

[0191] At least one of the three reflectors 320 is used to reflect an image of the flexible board adhesive reinforcement structure of product 20 to the image acquisition module 200, so that the flexible board adhesive reinforcement structure does not need to be inspected separately, saving inspection time.

[0192] In one possible embodiment, see Figure 13 The number of reflectors 320 is 7, so that the detection device 1 can simultaneously detect eight surfaces of the product 20.

[0193] In one possible embodiment, lens 220 is a telecentric lens.

[0194] Because of its unique optical design, the telecentric lens can minimize measurement errors caused by perspective and depth of field. For example, when photographing multiple surfaces 21 to be measured, ordinary lenses may cause deviations in measurement results due to the different distances of the product 20 from the lens. However, the telecentric lens can ensure the accuracy and consistency of measurement results within a certain working distance range, and the distortion of the captured image is very low, which can truly reflect the shape and characteristics of the surface 21 to be measured.

[0195] Of course, the lens 220 is not limited to a telecentric lens; it can also be a macro lens, an industrial fixed-focus lens, or a high-definition zoom lens, as those skilled in the art should know.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the product surface adhesive detection method and detection device of this application, and are not intended to limit it. Although the product surface adhesive detection method and detection device of this application have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for detecting adhesive coating quality on product surfaces, characterized in that, Includes the following steps: Take pictures of multiple surfaces of the product to be tested, and image the multiple surfaces to be tested onto a single test image; The detected image is divided into multiple regions, and each of the multiple regions corresponds one-to-one with the multiple surfaces to be tested; Select the locations in each of the aforementioned regions that require adhesive application on the detection image; The average grayscale value of the location requiring adhesive application in each of the aforementioned regions is detected separately; The average grayscale value is compared with a first preset grayscale range value to determine whether the average grayscale value is within the first preset grayscale range value. If the average grayscale value is within the first preset grayscale range, then it is determined that the dispensing in that area has been completed. If the average grayscale value is not within the first preset grayscale range, it is determined that the dispensing in that area has not been completed.

2. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, If multiple areas are determined to have been dispensed, the method includes: Determine whether glue has overflowed in the non-glued area at the boundary between two adjacent areas.

3. The method for detecting adhesive coating on product surfaces according to claim 2, characterized in that, The determination of whether glue has overflowed in the non-glued area at the boundary of two adjacent areas includes: On the detection image, the outer contour lines of two glue images in two adjacent regions are determined respectively; Extend the outline segments of the two glue images that are perpendicular to the boundary lines of the two adjacent regions and form an intersection point; The portion of the detection image that is close to the intersection point and located outside the outline of the glue image is identified as the non-glue area at the intersection. Detect the grayscale value of the non-coated area at the boundary of the detected image; The grayscale value of the non-coated area at the boundary is compared with the second preset grayscale range value to determine whether the grayscale value of the non-coated area at the boundary falls within the second preset grayscale range value. If the gray value of the non-glued area at the junction is not within the second preset gray value range, it is determined that glue has overflowed in that area. If the gray value of the non-adhesive area at the boundary is within the second preset gray value range, it is determined that no glue overflow has occurred in that area.

4. The method for detecting adhesive coating on product surfaces according to claim 3, characterized in that, The step of determining the outer contour lines of two glue images in two adjacent regions on the detection image includes: Based on the grayscale difference between the glue and the background in the detected image, the outer contour lines of the glue images in the two adjacent regions are extracted respectively.

5. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, If multiple areas are determined to have been dispensed, the method includes: The glue image is determined based on the grayscale difference between the glue and the background in the detected image; The glue image is checked for broken glue or holes.

6. The method for detecting adhesive coating on product surfaces according to claim 4, characterized in that, The detection of whether the glue image has breaks or holes includes: Calculate the center point set of the glue image; Based on the distribution pattern of the center point set, screen for abrupt change anomalies in the glue image. If there is an abrupt change point, it is determined that there is a break in the glue or a hole in the glue image. The abrupt change point is the location of the break in the glue or the hole. If no abrupt change points are found, it is determined that there are no broken glue or holes in the glue image.

7. The method for detecting adhesive coating on product surfaces according to claim 4, characterized in that, The detection of whether the glue image has breaks or holes includes: Identify whether there are areas of alternating light and dark gray levels in the glue image. If the alternating light and dark areas exist, it is determined that there are breaks or holes in the glue image, and the dark areas are the locations of the breaks or holes. If there are no areas of alternating light and dark, then the glue image is determined to be free of breaks or holes.

8. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, If multiple areas are determined to have been dispensed, the method includes: The glue image is determined based on the grayscale difference between the glue and the background in the detected image; Calculate the glue width of the glue image; The adhesive width is compared with a preset adhesive width range to determine whether the adhesive width falls within the preset range. If the adhesive width is within the preset adhesive width range, then the adhesive width is determined to meet the requirements; If the adhesive width is not within the preset adhesive width range, then the adhesive width is determined to not meet the requirements.

9. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, The method for detecting adhesive coating on the product surface includes: If the surface to be tested has a flexible board adhesive reinforcement structure, the adhesive coating quality of the flexible board adhesive reinforcement structure is tested.

10. The method for detecting adhesive coating on product surfaces according to claim 8, characterized in that, Inspecting the adhesive coating quality of flexible board reinforcement structures includes: Select the area where the flexible board adhesive reinforcement structure is located from among the multiple regions; The image of the flexible board adhesive reinforcement structure is located in the area where the flexible board adhesive reinforcement structure is located. The dimensional parameters of the image of the flexible board adhesive reinforcement structure are measured, and the dimensional parameters are compared with a preset dimensional parameter range to determine whether the dimensional parameters are within the preset dimensional parameter range. If the dimensional parameters are within the preset dimensional parameter range, then the adhesive coating quality of the flexible board adhesive reinforcement structure is determined to be qualified. If the dimensional parameters are not within the preset dimensional parameter range, the adhesive coating quality of the flexible board adhesive reinforcement structure is determined to be unqualified.

11. The method for detecting adhesive coating on product surfaces according to claim 10, characterized in that, The method for detecting the adhesive coating quality of the flexible board reinforcement structure includes: The image of the flexible board adhesive reinforcement structure is checked to see if there are any breaks in the adhesive or holes.

12. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, After photographing the surfaces to be tested of the product, so that multiple surfaces are imaged onto a single detection image, and before dividing the detection image into multiple regions, with each region corresponding one-to-one with one of the multiple surfaces to be tested, the above method includes: Distortion correction is performed on the detection image obtained by taking the picture.

13. The method for detecting adhesive coating on product surfaces according to claim 1, characterized in that, The process of dividing the image into multiple regions includes: Determine the size of the selection box for each of the aforementioned regions; Each of the selected regions is selected on the detection image based on pixel coordinates using the selected boxes; Images of each region within each selected box are acquired to complete the region division.

14. A testing device (1), said testing device (1) being applied to the product surface adhesive coating testing method according to any one of claims 1-13, characterized in that, include: Mounting bracket (100); An image acquisition module (200) is provided, comprising a camera (210) and a lens (220), wherein the camera (210) and the lens (220) are mounted on the mounting bracket (100), and the image acquisition module (200) is used to take pictures of multiple test surfaces (21) of the product (20); A prism module (300) includes a prism mounting base (310) and at least three reflectors (320) mounted on the prism mounting base (310). The prism mounting base (310) is mounted on the mounting frame (100) along the optical axis of the lens (220). The at least three reflectors (320) are arranged around the optical axis of the lens (220) and have a preset angle with the optical axis of the lens (220) so that the at least three reflectors (320) can reflect images of multiple surfaces (21) to the image acquisition module (200) respectively.

15. The detection device (1) according to claim 14, characterized in that, The prism mount (310) is rotatably mounted on the mounting bracket (100) along a direction perpendicular to the optical axis of the lens (220).

16. The detection device (1) according to claim 14, characterized in that, Along the optical axis of the lens (220), the distance between the end face of the lens (220) facing the prism module (300) and the prism module (300) is 40mm to 80mm.

17. The detection device (1) according to claim 14, characterized in that, The detection device (1) further includes a light source (400) disposed on the mounting bracket (100). Along the optical axis of the lens (220), the light source (400) is disposed between the image acquisition module (200) and the prism module (300). The light source (400) is a coaxial light source or a strip light source.

18. The detection device (1) according to claim 14, characterized in that, At least one of the three reflectors (320) is used to reflect an image of the flexible board adhesive reinforcement structure (22) of the product (20) to the image acquisition module (200).

Images