Defect detection method, device, equipment, medium and product of pultruded product

Abstract

This application provides a method, apparatus, device, medium, and product for defect detection of pultruded products, belonging to the field of composite material testing technology. The method includes: controlling at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives; stitching the multiple inspection images together to obtain a panoramic image of the pultruded product; performing grayscale processing on the panoramic image to obtain a corresponding grayscale image; extracting texture features and brightness features from the grayscale image; and determining the defect detection result corresponding to the pultruded product under inspection based on the texture features, brightness features, and preset defect feature conditions; wherein different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects, and wetting defects. According to the solution of this application embodiment, multiple types of defects can be accurately distinguished, improving detection efficiency and reliability.

Description

Technical Field

[0001] This application belongs to the field of composite material testing technology, and in particular relates to a defect detection method, device, equipment, medium and product for pultruded products. Background Technology

[0002] With the development of pultrusion technology, the requirements for the quality of pultruded products are becoming increasingly higher.

[0003] In the pultrusion production process, the molding quality of the product is crucial. Pultrusion production involves processes such as fiber layup and curing. However, as important structural components, pultruded products may exhibit various defects during production, impacting their reliability. Existing technologies for defect detection in pultruded products often rely on manual submission and observation, which suffers from complex testing procedures and an inability to detect multiple defect types.

[0004] Therefore, a solution is needed to improve the detection of multiple defect types in pultruded products. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, medium, and product for detecting defects in pultruded products, which can detect multiple types of defects and improve the efficiency and reliability of defect detection.

[0006] In a first aspect, embodiments of this application provide a defect detection method for pultruded products, the method comprising: Control at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives; The multiple detection images are stitched together to obtain a panoramic image of the pultruded product to be detected; The panoramic image is processed to obtain a grayscale image corresponding to the panoramic image; Extract texture features and brightness features from the grayscale image; Based on the texture features, the brightness features, and the preset defect feature conditions, the defect detection results corresponding to the pultruded product to be tested are determined; wherein, different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects, and wetting defects.

[0007] In one feasible implementation, the texture feature includes the distribution state of multiple pixel grayscale values ​​in the grayscale image, and the brightness feature includes the first local grayscale average value corresponding to multiple local regions in the grayscale image, wherein the local regions are determined based on the distribution range of the pixel grayscale values; wherein, determining the defect detection result corresponding to the pultruded product to be inspected based on the texture feature, the brightness feature, and preset defect feature conditions includes: Based on the distribution of the multiple pixel gray values, calculate the pixel gray value difference between any two adjacent pixels in each local region of the image; The pixel grayscale value difference is compared with a preset difference threshold to obtain a first number of pixel grayscale value differences that are greater than the preset difference threshold; The first quantity is compared with a first preset quantity threshold, and the first local grayscale average value is compared with a first preset grayscale threshold; If the first quantity is greater than the first preset quantity threshold and the first local grayscale average value is greater than the first preset grayscale threshold, the defect detection result is determined to include the white silk defect.

[0008] In one feasible implementation, the texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, determining the defect detection result corresponding to the pultruded product to be inspected based on the texture feature, the brightness feature, and preset defect feature conditions includes: Based on the distribution of the gray values ​​of the multiple pixels, the second local gray average value of all pixels in each local region of the gray image is calculated. The local region where the second local grayscale average value is less than the global pixel grayscale average value is determined as the first defect region to be identified; Calculate the first grayscale difference between adjacent pixels within a plurality of the first defect regions to be identified; The first grayscale difference is compared with a preset adjacent difference threshold; If the first grayscale difference is greater than the preset adjacent difference threshold, the defect detection result is determined to include the internal void defect.

[0009] In one feasible implementation, the texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, determining the defect detection result corresponding to the pultruded product to be inspected based on the texture feature, the brightness feature, and preset defect feature conditions includes: Based on the distribution of the gray values ​​of the multiple pixels, continuous, regular, and alternating light and dark normal stripes and the stripe direction corresponding to the normal stripes are determined. The area where the stripe direction is inconsistent with the preset normal fiber direction is identified as the second defect area to be identified. Calculate the average local grayscale value corresponding to each of the second defect regions to be identified; If the second local grayscale average value is less than the global pixel grayscale average value, the defect detection result is determined to include the stripe layering defect.

[0010] In one feasible implementation, the texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, determining the defect detection result corresponding to the pultruded product to be inspected based on the texture feature, the brightness feature, and preset defect feature conditions includes: Based on the distribution of the grayscale values ​​of the multiple pixels, the texture direction of the grayscale image is extracted; The texture directions are statistically analyzed to obtain the normal texture directions with the highest proportion and the most even distribution, as well as the abnormal texture directions. Regions whose average pixel grayscale value is less than the global average pixel grayscale value, greater than the second preset grayscale threshold, and belong to the abnormal texture direction are identified as abnormal regions. When the shape of the abnormal area is irregular, the defect detection result is determined to include the wetting defect.

[0011] In one feasible implementation, before determining the defect detection result corresponding to the pultruded product to be inspected based on the texture features, the brightness features, and preset defect feature conditions, the method further includes: The defect characteristic conditions are adjusted based on the production batch parameters of the pultruded product to be tested.

[0012] Secondly, embodiments of this application provide a defect detection device for pultruded products, the device comprising: The acquisition module is used to control at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives. The stitching module is used to stitch together the multiple detection images to obtain a panoramic image of the pultruded product to be detected. A grayscale processing module is used to perform grayscale processing on the panoramic image to obtain a grayscale image corresponding to the panoramic image; The feature extraction module is used to extract texture features and brightness features from the grayscale image; The defect detection module is used to determine the defect detection result of the pultruded product to be inspected based on the texture feature, the brightness feature and the preset defect feature conditions; wherein, different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects and wetting defects.

[0013] Thirdly, embodiments of this application provide a defect detection device for pultruded products, the device comprising: A processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method as described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the method described in the first aspect.

[0015] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0016] The defect detection method, apparatus, device, medium, and product of pultruded products in this application embodiment control at least one camera device to acquire multiple detection images of the pultruded product under test from different perspectives, and stitch the multiple detection images to obtain a panoramic image of the pultruded product under test. This can integrate multi-view images into a complete panoramic image, restore the overall shape of the product, and avoid incomplete product detection. The panoramic image is then subjected to grayscale processing to obtain a corresponding grayscale image. This can convert color information into single-channel grayscale values, reduce the computational complexity of subsequent feature extraction, and highlight the brightness differences and texture distribution between pixels, so as to capture the defect features of the pultruded product. Texture features and brightness features are extracted from the grayscale image. Based on the texture features, brightness features, and preset defect feature conditions, the defect detection result corresponding to the pultruded product under test is determined. This can accurately distinguish multiple types of defects without the need for subjective human judgment, thereby improving detection efficiency and reliability. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a defect detection method for pultruded products according to an embodiment of this application is shown; Figure 2A flowchart illustrating a defect detection method for pultruded products according to another embodiment of this application is shown; Figure 3 A flowchart illustrating a defect detection method for pultruded products according to another embodiment of this application is shown; Figure 4 A flowchart illustrating a defect detection method for pultruded products according to yet another embodiment of this application is shown; Figure 5 A flowchart illustrating a defect detection method for pultruded products according to another embodiment of this application is shown; Figure 6 This illustration shows a structural schematic diagram of a defect detection device for pultruded products provided in an embodiment of this application; Figure 7 A schematic diagram of the hardware structure of a defect detection device for pultruded products provided in an embodiment of this application is shown. Detailed Implementation

[0019] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0020] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0021] Pultruded products are fiber-reinforced composite profiles produced by the pultrusion molding process. Current methods for defect detection of pultruded samples rely heavily on manual submission and on-site testing. Defects are mainly determined by manual visual observation or destructive cutting, which is not only inefficient and prone to damaging the samples, but also results in unstandardized and unreliable data that is difficult to use for subsequent accurate analysis and traceability.

[0022] To address the problems in the prior art, embodiments of this application provide a method, apparatus, equipment, medium, and product for detecting defects in pultruded products.

[0023] The defect detection method for pultruded products provided in the embodiments of this application will be introduced first below.

[0024] Figure 1 A schematic flowchart of a defect detection method for pultruded products according to an embodiment of this application is shown. Figure 1 As shown, the method may include the following steps: S110 to S150.

[0025] S110. Control at least one camera device to acquire multiple inspection images of the pultruded product to be inspected from different perspectives.

[0026] In this embodiment, the pultruded product to be inspected refers to a pultruded product that still requires but has not yet completed quality inspection. Using one or more camera devices, the pultruded product to be inspected can be photographed from multiple different angles and positions, acquiring multiple images of different surfaces of the pultruded product to be inspected, i.e., multiple inspection images of the pultruded product to be inspected from different perspectives.

[0027] In some embodiments, four different camera devices or cameras can be controlled to acquire multiple detection images. The four camera devices or cameras can be respectively arranged at the four vertices or the middle of the four sides of the square frame. During the transportation process, the pultruded product to be inspected can pass through the center of the positive frame from a direction perpendicular to the plane where the positive frame is located, and the lenses of the four camera devices or cameras can be aimed at various sides of the cuboid pultruded product to be inspected. In this way, detection images of the pultruded product to be inspected from various angles and from various sides can be acquired by the camera devices.

[0028] In one embodiment, if only a single camera device or camera is used, it can move along a track around the pultruded product to be inspected. During its movement, the camera lens remains focused on the pultruded product, allowing it to capture inspection images at appropriate positions. These images are then stitched together for comprehensive inspection. The track can be a circular track centered on the pultruded profile. Alternatively, the track can be spiral-shaped, with the camera's movement speed and rotational angular velocity adjusted according to the pultruded profile's output rate. This ensures that the camera captures images of the same cross-section of the product at different angles, achieving synchronous matching between the shooting position and the profile's output.

[0029] In some embodiments, prior to S110, the method may further include: after the pultruded products are continuously produced, determining the samples to be inspected, i.e. the pultruded products to be inspected, according to pre-set inspection rules (e.g., by batch, by length, by time), and then automatically transporting the pultruded products to be inspected to an inspection station or inspection station for defect detection using automated inspection equipment.

[0030] In some embodiments, the pre-defined inspection rules refer to information related to various defects, requiring the selected samples to fully reflect all possible defects in the products, ensuring the representativeness of the pultruded products to be inspected, and thus achieving the accuracy of automated online inspection.

[0031] S120. Stitch together multiple inspection images to obtain a panoramic image of the pultruded product to be inspected.

[0032] In this embodiment, multiple detection images are combined into a complete, blind-spot-free panoramic image, i.e., a panoramic image of the pultruded product to be detected, using an image stitching algorithm.

[0033] S130. Perform grayscale processing on the panoramic image to obtain the corresponding grayscale image.

[0034] In this embodiment, the colored panoramic image is converted into a grayscale image that only shows brightness levels and no color information. This removes irrelevant color interference, highlights texture and brightness differences, and makes it easier to identify defects such as white lines, dark holes, and layering. It also simplifies image data, reduces the amount of subsequent calculations, and makes subsequent feature extraction faster and more stable.

[0035] S140. Extract texture features and brightness features from the grayscale image.

[0036] In this embodiment, texture features refer to the distribution pattern and arrangement of grayscale values ​​of each pixel in a grayscale image, while brightness features refer to the brightness and grayscale value of a grayscale image. Texture features can characterize the regularity, direction, and thickness of patterns and textures in a grayscale image. Brightness features can reflect the light transmittance, density, and resin impregnation of pultruded products.

[0037] In this embodiment, texture direction, uniformity, and distribution status can be analyzed based on the distribution and arrangement of pixel gray values ​​in a grayscale image to form texture features; and pixel gray values ​​in a grayscale image can be statistically analyzed to form brightness features.

[0038] S150. Based on texture features, brightness features, and preset defect feature conditions, determine the defect detection results corresponding to the pultruded product to be inspected.

[0039] In this embodiment, defect feature conditions refer to pre-set various defect judgment rules or conditions. The extracted texture features and brightness features can be compared and matched with the pre-set defect judgment rules or conditions to automatically determine whether the pultruded product to be inspected has defects and what type of defect it is. The defect detection result can include no defects or a specific defect type.

[0040] In this embodiment, different defect feature conditions correspond to different defect types. The defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects, and wetting defects. It should be noted that each defect exhibits different textures and brightness in the image, so different defect feature conditions correspond to different defect types. This embodiment is not limited to detecting only one type of defect; it can simultaneously identify two, three, or all four types.

[0041] Through S110 to S150, pultruded products can be inspected without blind spots by multi-view image acquisition and stitching. The data is simplified and the defect features are highlighted after grayscale processing. Combined with texture and brightness features, various defects are automatically identified according to preset conditions, realizing automated, non-destructive, efficient and objective quality inspection, which greatly improves the accuracy and efficiency of inspection, while solving the problems of missed detection, false detection and cumbersome process in traditional manual inspection.

[0042] Figure 2 A flowchart illustrating a defect detection method for pultruded products according to another embodiment of this application is shown. In this embodiment, texture features include the distribution state of grayscale values ​​of multiple pixels in a grayscale image, and brightness features include the first local grayscale average value corresponding to multiple local regions in the grayscale image. The local regions are determined based on the distribution range of pixel grayscale values. As shown in the figure, determining the defect detection result corresponding to the pultruded product to be detected based on texture features, brightness features, and preset defect feature conditions may include the following steps: S210 to S240.

[0043] S210. Based on the distribution of multiple pixel gray values, calculate the difference in pixel gray values ​​between any two adjacent pixels in a local region of each image.

[0044] In this embodiment, the distribution state of multiple pixel grayscale values ​​refers to the arrangement of multiple pixel grayscale values ​​in a grayscale image, such as whether it is uniform and smooth or chaotic and abrupt. Specifically, the distribution state of multiple pixel grayscale values ​​may include the magnitude and concentration range of the pixel grayscale values, as well as the changes in pixel grayscale values ​​between adjacent pixels. The changes in pixel grayscale values ​​between adjacent pixels are used to characterize whether the pixel grayscale values ​​between adjacent pixels are smooth transitions or abrupt changes (sudden large contrast between light and dark). It should be noted that pixel grayscale values ​​can reflect light and dark features or brightness features; the larger the pixel grayscale value, the brighter the color, and vice versa.

[0045] In this embodiment, pixels with similar grayscale values ​​and similar distribution ranges can be grouped into the same local image region based on the distribution of pixel grayscale values. Specifically, adjacent pixels whose grayscale value difference is less than or equal to a preset grayscale threshold can be classified into the same local image region. The first local grayscale average value refers to the average of the grayscale values ​​of all pixels within the same local image region. The difference in grayscale values ​​between any two adjacent pixels within each local image region can be calculated.

[0046] S220. Compare the pixel grayscale value difference with a preset difference threshold to obtain a first number of pixel grayscale value differences that are greater than the preset difference threshold.

[0047] In this embodiment, the difference in grayscale value of each pixel is compared with a preset difference threshold, and the number of pixels whose grayscale value difference is greater than the preset difference threshold is the first number.

[0048] S230. Compare the first quantity with the first preset quantity threshold, and compare the first local grayscale average value with the first preset grayscale threshold.

[0049] In this embodiment, the first preset quantity threshold is a pre-set quantity threshold used to determine whether the texture or stripes in a local area of ​​the image are messy or whether there are too many jumps. The first preset grayscale threshold is a pre-set grayscale threshold used to determine whether a local area of ​​the image is bright enough. S230 may include: for each local area of ​​the image, comparing the first quantity with the first preset quantity threshold, and comparing the first local grayscale average value of the same local area of ​​the image with the first preset grayscale threshold.

[0050] S240. If the first quantity is greater than the first preset quantity threshold and the first local grayscale average value is greater than the first preset grayscale threshold, determine that the defect detection result includes white silk defect.

[0051] In this embodiment, if the first quantity is greater than the first preset quantity threshold and the first local grayscale average value is greater than the first preset grayscale threshold, it means that the texture of the local area of ​​the image is very messy and the overall brightness is obviously too bright. At this time, the grayscale feature corresponding to the white silk defect or the preset defect feature condition corresponding to the white silk defect is satisfied. Therefore, it can be determined that the defect detection result may include the white silk defect.

[0052] Through steps S210 to S240, the statistical analysis of the grayscale difference between adjacent pixels within a local area of ​​the image can determine whether the texture of that area is messy or has drastic abrupt changes; at the same time, the average local grayscale value can be used to determine whether the overall brightness of that area is too high; when both the degree of texture messiness and the brightness of the area exceed the corresponding preset threshold, it can be accurately determined as a white silk defect.

[0053] In some embodiments, stitching together multiple detection images to obtain a panoramic image of the pultruded product to be inspected may include the following steps: S121, removing lens distortion from each detection image to obtain a flat detection image; stitching together the flat detection images based on the spatial position corresponding to the flat detection image to obtain a panoramic image.

[0054] Figure 3 This illustration shows a flowchart of a defect detection method for pultruded products according to another embodiment of this application. In this embodiment, the texture feature includes the distribution state of grayscale values ​​of multiple pixels in a grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image. As shown in the figure, determining the defect detection result corresponding to the pultruded product to be detected based on the texture feature, brightness feature, and preset defect feature conditions may include the following steps: S310 to S350.

[0055] S310. Based on the distribution of multiple pixel gray values, calculate the second local gray average value of all pixels in each local region of the gray image.

[0056] In this embodiment, a local region is a regular area into which a grayscale image is divided according to a fixed size, based on the distribution of grayscale values ​​of multiple pixels. The average grayscale value of all pixels in a local region is the second local grayscale average value. The average grayscale value of each local region can be calculated to obtain the second local grayscale average value.

[0057] S320. The local area where the average local gray level is less than the average global pixel gray level is determined as the first defect area to be identified.

[0058] In this embodiment, if the second local grayscale average value of a local area is less than the global pixel grayscale average value, it indicates that the local area is significantly darker than the whole. The local area can be identified as the first defect area to be identified, which refers to the area where there is a suspected or possible internal void.

[0059] S330. Calculate the first grayscale difference between adjacent pixels within multiple first defect regions to be identified.

[0060] In this embodiment, the difference in pixel grayscale values ​​between adjacent pixels within the first defect region to be identified is calculated to obtain the first grayscale difference.

[0061] S340. Compare the first grayscale difference with the preset adjacent difference threshold.

[0062] In this embodiment, the preset adjacent difference threshold is a pre-set grayscale difference reference value used to determine whether the brightness change between adjacent pixels is obvious or drastic. By comparing the first grayscale difference value with the preset adjacent difference threshold, it can be determined whether the brightness change of the first defect area to be identified is obvious or drastic.

[0063] S350. If the first grayscale difference is greater than the preset adjacent difference threshold, the defect detection result is determined to include internal void defects.

[0064] In this embodiment, if the first grayscale difference is greater than the preset adjacent difference threshold, it means that the internal gap is not a gradient shadow, but a dark area with obvious changes or boundaries. This satisfies the preset defect feature condition of being dark and having sharp boundaries, that is, it satisfies the preset defect feature condition corresponding to the internal gap defect. Therefore, it can be determined that the defect detection result includes the internal gap defect.

[0065] Figure 4 This illustration shows a flowchart of a defect detection method for pultruded products according to another embodiment of this application. In this embodiment, texture features include the distribution of grayscale values ​​of multiple pixels in a grayscale image, and brightness features include the global pixel grayscale average value corresponding to all pixels in the grayscale image. As shown in the figure, determining the defect detection result corresponding to the pultruded product to be detected based on texture features, brightness features, and preset defect feature conditions may include the following steps: S410 to S440.

[0066] S410. Based on the distribution of multiple pixel gray values, determine continuous, regular, and alternating bright and dark normal stripes and the stripe direction corresponding to the normal stripes.

[0067] In this embodiment, continuous stripes refer to stripes formed by extended, non-scattered dots; regular stripes refer to stripes with basically the same width, spacing, and arrangement direction; and alternating light and dark stripes refer to one or more groups of stripes where one light stripe is followed by one dark stripe, repeating in sequence. Normal stripes refer to stripes that simultaneously belong to continuous stripes, regular stripes, and alternating light and dark stripes.

[0068] In this embodiment, each stripe can be identified based on the distribution of multiple pixel grayscale values, and statistics can be performed on each stripe to determine normal stripes and their corresponding stripe directions. The stripe direction corresponding to a normal stripe refers to the direction in which the normal stripe extends and is arranged. For example, if the normal stripe extends vertically, then the stripe direction corresponding to the normal stripe is vertical.

[0069] In this embodiment, based on the distribution of multiple pixel gray values, the arrangement pattern of pixel brightness in a region can be analyzed to find one or more continuous and neat bright and dark stripes in the image, and determine the direction in which these stripes extend, that is, to obtain the stripe direction corresponding to the alternating bright and dark stripes.

[0070] S420. The area where the stripe direction is inconsistent with the preset normal fiber direction is identified as the second defect area to be identified.

[0071] In this embodiment, the preset normal fiber refers to the pre-defined extension direction of the fiber stripes when the product is normal, such as uniformly vertical, uniformly horizontal, or uniformly at a certain tilt angle. This can be set using different types of normal products. Areas where the stripe direction is inconsistent with the preset normal fiber direction can be identified as the second defect area to be identified. This second defect area to be identified refers to areas where stripe delamination defects may occur or are suspected to exist.

[0072] S430. Calculate the average local grayscale value corresponding to each second defect region to be identified.

[0073] In this embodiment, the second local grayscale average value refers to the average grayscale value of all pixels within the second defect area to be identified.

[0074] S440. If the average local grayscale value is less than the average global pixel grayscale value, determine that the defect detection result includes stripe layering defects.

[0075] In this embodiment, if the second local grayscale average value is less than the global pixel grayscale average value, it means that the average brightness of this area (the second local grayscale average value) is darker than the overall brightness. Thus, it can be determined that there is a region with disordered and dark stripe direction, which meets the preset defect feature conditions corresponding to stripe layering defects. Therefore, it can be determined that the defect detection result includes stripe layering defects.

[0076] Figure 5 A flowchart illustrating a defect detection method for pultruded products according to another embodiment of this application is shown. In this embodiment, texture features include the distribution of grayscale values ​​of multiple pixels in a grayscale image, and brightness features include the global pixel grayscale average value corresponding to all pixels in the grayscale image. As shown, determining the defect detection result corresponding to the pultruded product to be detected based on texture features, brightness features, and preset defect feature conditions may include the following steps: S510 to S540.

[0077] S510. Based on the distribution of multiple pixel grayscale values, extract the texture direction of the grayscale image.

[0078] In this embodiment, the direction in which the fiber texture extends can be determined based on the distribution of multiple pixel grayscale values, i.e., the arrangement pattern of pixel brightness in the image. The surface of a pultruded product consists of multiple rows of fibers, forming distinct textures. By analyzing the distribution of pixel grayscale values, the direction of the fiber texture can be determined.

[0079] S520. Statistical analysis of texture directions is performed to obtain the normal texture directions with the highest proportion and the most even distribution, as well as the abnormal texture directions.

[0080] In this embodiment, the texture directions of all regions in the grayscale image are statistically analyzed. The texture direction with the highest proportion and the most uniform distribution is determined as the normal texture direction, while the texture direction that differs excessively from the normal texture direction is determined as the abnormal texture direction. Specifically, if the texture in the same direction appears continuously over a large area in the image, covers a wide range, and has small directional deviations at each position, then the texture direction can be determined to be uniformly distributed.

[0081] S530, Identify regions whose average pixel grayscale value is less than the global average pixel grayscale value, greater than the second preset grayscale threshold, and belong to abnormal texture directions as abnormal regions.

[0082] In this embodiment, if the average pixel grayscale value is less than the global average pixel grayscale value, it means that the overall brightness of the area is relatively dark. If the average pixel grayscale value is greater than the second preset grayscale threshold, it means that the area is not particularly dark, and areas with extremely low brightness, such as gaps or layers, can be excluded. If the texture direction of the area belongs to an abnormal texture direction, it indicates that the fiber texture direction of the area is disordered. If there is an area that simultaneously satisfies the conditions of an average pixel grayscale value less than the global average pixel grayscale value, greater than the second preset grayscale threshold, and belonging to an abnormal texture direction, then the area is dark but not particularly black, and the texture is disordered, so the area can be determined as an abnormal area.

[0083] S540. When the shape of the abnormal area is irregular, the defect detection results shall include wetting defects.

[0084] In this embodiment, it is also necessary to examine the outline of the abnormal area, that is, the shape of the abnormal area. If the shape of the abnormal area does not have a regular geometric pattern (such as not being a circle, square, straight strip, long strip, etc.), it indicates that it is a scattered, irregular, and blurry area that meets the preset defect feature conditions corresponding to the wetting defect. At this time, it can be determined that the defect detection result includes the wetting defect.

[0085] In some embodiments, prior to S150, the method further includes: adjusting defect feature conditions based on production batch parameters of the pultruded product to be inspected.

[0086] Specifically, the criteria (defect characteristic conditions) used to judge defects can be dynamically adjusted based on the production batch of the pultruded product to be inspected and the corresponding process parameters of that batch, instead of using a fixed set of parameters throughout the entire process. Because different batches of products have slight differences in brightness and texture, using the same set of standards can easily lead to misjudgments. Therefore, the defect judgment criteria are first fine-tuned to suit the current production batch before inspection begins. For example, at least one of the following can be adjusted based on the production batch parameters of the pultruded product to be inspected: a preset adjacent difference threshold, a second preset grayscale threshold, a preset difference threshold, a first preset quantity threshold, and a first preset grayscale threshold.

[0087] Figure 6 A schematic diagram of a defect detection device for pultruded products according to an embodiment of this application is shown. As shown, the defect detection device 600 for pultruded products may include a data acquisition module 610, a splicing module 620, a grayscale processing module 630, a feature extraction module 640, and a defect detection module 650.

[0088] The acquisition module 610 is used to control at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives.

[0089] The stitching module 620 is used to stitch together the multiple detection images to obtain a panoramic image of the pultruded product to be detected.

[0090] The grayscale processing module 630 is used to perform grayscale processing on the panoramic image to obtain a grayscale image corresponding to the panoramic image.

[0091] The feature extraction module 640 is used to extract texture features and brightness features from the grayscale image.

[0092] The defect detection module 650 is used to determine the defect detection result of the pultruded product to be inspected based on the texture feature, the brightness feature and the preset defect feature conditions; wherein, different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defect, internal void defect, stripe delamination defect and wetting defect.

[0093] In some embodiments, the texture feature includes the distribution state of gray values ​​of multiple pixels in the grayscale image, and the brightness feature includes the first local gray average value corresponding to multiple local regions in the grayscale image, wherein the local regions are determined based on the distribution range of the pixel gray values; the defect detection device 600 for pultruded products may further include a gray value difference calculation module, a difference comparison module, a quantity comparison module, and a white silk defect detection module.

[0094] The grayscale difference calculation module is used to calculate the pixel grayscale difference between any two adjacent pixels in each local region of the image based on the distribution of the grayscale values ​​of the multiple pixels.

[0095] The difference comparison module is used to compare the pixel grayscale value difference with a preset difference threshold to obtain a first number of pixel grayscale value differences that are greater than the preset difference threshold.

[0096] The quantity comparison module is used to compare the first quantity with a first preset quantity threshold, and to compare the first local grayscale average value with the first preset grayscale threshold.

[0097] The white silk defect detection module is used to determine that the defect detection result includes the white silk defect when the first quantity is greater than the first preset quantity threshold and the first local grayscale average value is greater than the first preset grayscale threshold.

[0098] In some embodiments, the texture feature includes the distribution state of gray values ​​of multiple pixels in the grayscale image, and the brightness feature also includes the global pixel gray average value corresponding to all pixels in the grayscale image; the defect detection device 600 for pultruded products may further include an average value statistics module, a first defect area determination module, a first grayscale difference module, an adjacent difference comparison module, and an internal void defect detection module.

[0099] The average value statistics module is used to calculate the second local gray value average of all pixels in each local region of the grayscale image based on the distribution status of the grayscale values ​​of the multiple pixels.

[0100] The first defect region determination module is used to determine the local region where the second local grayscale average value is less than the global pixel grayscale average value as the first defect region to be identified.

[0101] The first grayscale difference module is used to calculate the first grayscale difference between adjacent pixels within a plurality of the first defect regions to be identified.

[0102] The adjacent difference comparison module is used to compare the first grayscale difference with a preset adjacent difference threshold.

[0103] An internal void defect detection module is used to determine that the defect detection result includes the internal void defect when the first grayscale difference is greater than the preset adjacent difference threshold.

[0104] In some embodiments, the texture feature includes the distribution state of gray values ​​of multiple pixels in the grayscale image, and the brightness feature also includes the global pixel gray average value corresponding to all pixels in the grayscale image; the defect detection device 600 for pultruded products may further include a stripe direction determination module, a second defect area determination module, a second local gray average value calculation module, and a stripe layering defect detection module.

[0105] The stripe direction determination module is used to determine continuous, regular, and alternating light and dark normal stripes and the stripe direction corresponding to the normal stripes based on the distribution state of the gray values ​​of the multiple pixels. The second defect region determination module is used to determine the region where the stripe direction is inconsistent with the preset normal fiber direction as the second defect region to be identified. The second local grayscale average value calculation module is used to calculate the second local grayscale average value corresponding to each of the second defect regions to be identified. The stripe layering defect detection module is used to determine that the defect detection result includes the stripe layering defect when the second local grayscale average value is less than the global pixel grayscale average value.

[0106] In some embodiments, the texture feature includes the distribution state of gray values ​​of multiple pixels in the grayscale image, and the brightness feature also includes the global pixel gray average value corresponding to all pixels in the grayscale image; the defect detection device 600 for pultruded products may further include a texture direction determination module, a texture direction statistics module, an abnormal region determination module, and an immersion defect detection module.

[0107] The texture direction determination module is used to extract the texture direction of the grayscale image based on the distribution state of the grayscale values ​​of the multiple pixels.

[0108] The texture direction statistics module is used to perform statistics on the texture directions to obtain the normal texture directions with the highest proportion and the most uniform distribution, as well as the abnormal texture directions.

[0109] The abnormal region determination module is used to determine regions whose average pixel grayscale value is less than the global average pixel grayscale value, greater than the second preset grayscale threshold, and belong to the abnormal texture direction as abnormal regions.

[0110] The wetting defect detection module is used to determine whether the defect detection result includes the wetting defect when the abnormal area has an irregular shape.

[0111] In some embodiments, the defect detection device 600 for pultruded products may also include a defect feature condition adjustment module for adjusting defect feature conditions based on the production batch parameters of the pultruded product to be detected.

[0112] Figure 7 A schematic diagram of the hardware structure of a defect detection device for pultruded products provided in an embodiment of this application is shown.

[0113] The defect detection equipment for pultruded products may include a processor 301 and a memory 302 storing computer program instructions.

[0114] Specifically, the processor 301 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0115] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 302 may include removable or non-removable (or fixed) media, or memory 302 may be non-volatile solid-state memory. Memory 302 may be internal or external to the integrated gateway disaster recovery device.

[0116] In one instance, memory 302 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0117] Memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0118] The processor 301 reads and executes computer program instructions stored in the memory 302 to achieve... Figure 1 The defect detection method for pultruded products in the illustrated embodiment.

[0119] In one example, the defect detection device for pultruded products may also include a communication interface 303 and a bus 304. As shown in the figure, the processor 301, memory 302, and communication interface 303 are connected via the bus 304 and communicate with each other.

[0120] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0121] Bus 304 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not as a limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 304 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0122] The defect detection equipment for pultruded products can execute the online data traffic billing method described in this application embodiment based on currently blocked spam SMS messages and SMS messages reported by users, thereby achieving a combination of... Figure 1 Defect detection method for pultruded products described.

[0123] Furthermore, in conjunction with the defect detection method for pultruded products in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the defect detection methods for pultruded products in the above embodiments.

[0124] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the defect detection methods for pultruded products described in the above embodiments.

[0125] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0126] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact disc read-only memory (CD-ROM), optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0127] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0128] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0129] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A defect detection method for pultruded products, characterized in that, include: Control at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives; The multiple detection images are stitched together to obtain a panoramic image of the pultruded product to be detected; The panoramic image is processed to obtain a grayscale image corresponding to the panoramic image; Extract texture features and brightness features from the grayscale image; Based on the texture features, the brightness features, and the preset defect feature conditions, the defect detection results corresponding to the pultruded product to be inspected are determined; wherein, different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects, and wetting defects.

2. The method according to claim 1, characterized in that, The texture feature includes the distribution state of multiple pixel grayscale values ​​in the grayscale image, and the brightness feature includes the first local grayscale average value corresponding to multiple local regions in the grayscale image, wherein the local regions are determined based on the distribution range of the pixel grayscale values; wherein, based on the texture feature, the brightness feature, and preset defect feature conditions, the defect detection result corresponding to the pultruded product to be inspected is determined, including: Based on the distribution of the multiple pixel gray values, calculate the pixel gray value difference between any two adjacent pixels in each local region of the image; The pixel grayscale value difference is compared with a preset difference threshold to obtain a first number of pixel grayscale value differences that are greater than the preset difference threshold; The first quantity is compared with a first preset quantity threshold, and the first local grayscale average value is compared with a first preset grayscale threshold; If the first quantity is greater than the first preset quantity threshold and the first local grayscale average value is greater than the first preset grayscale threshold, the defect detection result is determined to include the white silk defect.

3. The method according to claim 1, characterized in that, The texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, based on the texture feature, the brightness feature, and preset defect feature conditions, the defect detection result corresponding to the pultruded product to be detected is determined, including: Based on the distribution of the gray values ​​of the multiple pixels, the second local gray average value of all pixels in each local region of the gray image is calculated. The local region where the second local grayscale average value is less than the global pixel grayscale average value is determined as the first defect region to be identified; Calculate the first grayscale difference between adjacent pixels within a plurality of the first defect regions to be identified; The first grayscale difference is compared with a preset adjacent difference threshold; If the first grayscale difference is greater than the preset adjacent difference threshold, the defect detection result is determined to include the internal void defect.

4. The method according to claim 1, characterized in that, The texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, based on the texture feature, the brightness feature, and preset defect feature conditions, the defect detection result corresponding to the pultruded product to be detected is determined, including: Based on the distribution of the gray values ​​of the multiple pixels, continuous, regular, and alternating light and dark normal stripes and the stripe direction corresponding to the normal stripes are determined. The area where the stripe direction is inconsistent with the preset normal fiber direction is identified as the second defect area to be identified. Calculate the average local grayscale value corresponding to each of the second defect regions to be identified; If the second local grayscale average value is less than the global pixel grayscale average value, the defect detection result is determined to include the stripe layering defect.

5. The method according to claim 1, characterized in that, The texture feature includes the distribution state of grayscale values ​​of multiple pixels in the grayscale image, and the brightness feature includes the global pixel grayscale average value corresponding to all pixels in the grayscale image; wherein, based on the texture feature, the brightness feature, and preset defect feature conditions, the defect detection result corresponding to the pultruded product to be detected is determined, including: Based on the distribution of the grayscale values ​​of the multiple pixels, the texture direction of the grayscale image is extracted; The texture directions are statistically analyzed to obtain the normal texture directions with the highest proportion and the most even distribution, as well as the abnormal texture directions. Regions whose average pixel grayscale value is less than the global average pixel grayscale value, greater than the second preset grayscale threshold, and belong to the abnormal texture direction are identified as abnormal regions. When the shape of the abnormal area is irregular, the defect detection result is determined to include the wetting defect.

6. The method according to claim 1, characterized in that, Before determining the defect detection result corresponding to the pultruded product to be inspected based on the texture features, the brightness features, and preset defect feature conditions, the method further includes: The defect characteristic conditions are adjusted based on the production batch parameters of the pultruded product to be tested.

7. A defect detection device for pultruded products, characterized in that, The device includes: The acquisition module is used to control at least one camera device to acquire multiple inspection images of the pultruded product under inspection from different perspectives. The stitching module is used to stitch together the multiple detection images to obtain a panoramic image of the pultruded product to be detected. A grayscale processing module is used to perform grayscale processing on the panoramic image to obtain a grayscale image corresponding to the panoramic image; The feature extraction module is used to extract texture features and brightness features from the grayscale image; The defect detection module is used to determine the defect detection result of the pultruded product to be inspected based on the texture feature, the brightness feature and the preset defect feature conditions; wherein, different defect feature conditions correspond to different defect types, and the defect types include at least two of the following: white thread defects, internal void defects, stripe delamination defects and wetting defects.

8. A defect detection device for pultruded products, characterized in that, The device includes: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer program instructions, which, when executed by a processor, implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-6.

Images