Methods and apparatus for detecting defects in cast film

Abstract

This application relates to a method and apparatus for detecting defects in cast film. The method includes: acquiring a transmitted light scan image and a reflected light scan image of the cast film to be tested; identifying the location of defects in the cast film to be tested based on the transmitted light scan image and / or the reflected light scan image; acquiring a transmitted light image corresponding to the target defect location based on the transmitted light scan image; acquiring a reflected light image corresponding to the target defect location based on the reflected light scan image; and determining the defect type at the target defect location based on the transmitted light image and the reflected light image. This method can accurately identify the type of defects in the cast film to be tested by processing the transmitted light scan image and the reflected light scan image.

Description

Technical Field

[0001] This application relates to the field of materials testing technology, and in particular to a method and apparatus for detecting defects in cast film. Background Technology

[0002] Polymer insulation materials, due to their excellent properties, have become the core basic material for the insulation layer of power cables and various power equipment. However, as power grids develop towards higher voltage, larger capacity, and longer distances, the demand for the purity and processing quality of polymer insulation materials is also increasing.

[0003] In related technologies, defect detection of polymer insulating materials mainly adopts manual visual inspection and ultrasonic testing methods, which have low detection sensitivity and low accuracy, making it difficult to accurately identify defect types. Summary of the Invention

[0004] Therefore, it is necessary to provide a method and apparatus for detecting defects in cast film that can accurately identify defect types, in order to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for detecting defects in cast film, including:

[0006] Acquire transmitted light scanning images and reflected light scanning images of the cast film under test;

[0007] Based on the transmitted light scanning image and / or reflected light scanning image, identify the location of defects in the cast film under test;

[0008] Based on the transmitted light scanning image, obtain the transmitted light image corresponding to the location of the target defect;

[0009] Based on the reflected light scanning image, obtain the reflected light image corresponding to the location of the target defect;

[0010] Based on the transmitted light image and the reflected light image, determine the defect type at the location of the target defect.

[0011] In one embodiment, determining the defect type of the target defect location based on the transmitted light image and the reflected light image includes: determining the first defect fractal dimension and the first grayscale ratio corresponding to the target defect location based on the transmitted light image; determining the second defect fractal dimension and the second grayscale ratio corresponding to the target defect location based on the reflected light image; determining the defect type as a dense impurity defect in response to at least one of the first defect fractal dimension, the second defect fractal dimension, and the grayscale ratio difference between the first and second grayscale ratios satisfying a preset condition; and determining the defect type as a gel defect in response to at least one of the first defect fractal dimension, the second defect fractal dimension, and the grayscale ratio difference between the first and second grayscale ratios not satisfying a preset condition.

[0012] In one embodiment, determining the defect type as a dense impurity defect in response to the fact that the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio all satisfy preset conditions includes: determining the defect type as a dense impurity defect in response to the fact that the first defect fractal dimension is greater than a preset value for fractal dimension, the second defect fractal dimension is greater than a preset value for fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio is less than a preset value for difference; determining the defect type as a gel defect in response to the fact that at least one of the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio does not satisfy preset conditions includes: determining the defect type as a gel defect in response to the fact that either the first defect fractal dimension or the second defect fractal dimension is less than or equal to a preset value for fractal dimension, or the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio is greater than or equal to a preset value for difference.

[0013] In one embodiment, the preset value for the fractal dimension is 1.95; the preset value for the difference is 0.025.

[0014] In one embodiment, before determining the defect type at the target defect location based on the transmitted light image and the reflected light image, the method further includes: identifying whether the defect at the target defect location is a fibrous impurity defect; wherein determining the defect type at the target defect location based on the transmitted light image and the reflected light image is performed after identifying the defect at the target defect location as a non-fibrous impurity defect.

[0015] In one embodiment, identifying whether a defect at a target defect location is a fibrous impurity defect includes: determining a first defect morphology parameter corresponding to the target defect location based on a transmitted light image; determining a second defect morphology parameter corresponding to the target defect location based on a reflected light image; and identifying whether the defect type is a fibrous impurity defect based on the morphology difference between the first and second defect morphology parameters.

[0016] In one embodiment, both the first defect morphology parameter and the second defect morphology parameter include: roundness and longest axis; wherein, identifying whether the defect type is a fibrous impurity defect based on the morphological difference between the first defect morphology parameter and the second defect morphology parameter includes: determining the defect type as a fibrous impurity defect in response to the first defect morphology parameter having a roundness < 0.3 and a longest axis > 20, and the second defect morphology parameter having a roundness < 0.2 and a longest axis > 50.

[0017] In one embodiment, the method further includes: dynamically calibrating a first acquisition time for the transmitted light image corresponding to the defect location in the target sample in the transmitted light scanning image, and a second acquisition time for the reflected light image corresponding to the defect location in the target sample in the reflected light scanning image; determining an image acquisition delay based on the time difference between the second acquisition time and the first acquisition time and the transport speed in the target sample; wherein, acquiring the transmitted light image corresponding to the target defect location based on the transmitted light scanning image and acquiring the reflected light image corresponding to the target defect location based on the reflected light scanning image further includes: acquiring the transport speed of the cast film to be tested; and acquiring the transmitted light image and the reflected light image corresponding to the target defect location from the transmitted light scanning image and the reflected light scanning image based on the image acquisition delay and the transport speed of the cast film to be tested.

[0018] Secondly, this application also provides a casting film defect detection device, comprising:

[0019] The acquisition module includes a transmitted light imaging unit and a reflected light imaging unit; the transmitted light imaging unit is used to acquire the transmitted light scanning image of the cast film under test, and the reflected light imaging unit is used to acquire the reflected light scanning image of the cast film under test.

[0020] The defect identification module connects the transmitted light imaging unit and the reflected light imaging unit, and is used to identify the location of defects in the cast film under test based on the transmitted light scanning image and / or the reflected light scanning image.

[0021] The defect determination module, connected to the defect identification module, is used to: obtain a transmitted light image corresponding to the target defect location based on the transmitted light scanning image, obtain a reflected light image corresponding to the target defect location based on the reflected light scanning image, and determine the defect type at the target defect location based on the transmitted light image and the reflected light image.

[0022] The acquisition module also includes a velocity measurement unit, used to acquire the transfer velocity in the target sample and the transfer velocity of the cast film to be tested;

[0023] The defect determination module is also connected to a speed measurement unit, which is used to obtain the transmitted light image and reflected light image corresponding to the target defect location from the transmitted light scanning image and the reflected light scanning image based on the image acquisition delay and the transmission speed of the cast film under test.

[0024] The aforementioned method and apparatus for detecting defects in cast films identify the locations of defects in the cast film by acquiring transmitted light scanning images and reflected light scanning images of the film under test. For each target defect location, a transmitted light image corresponding to the target defect location is acquired based on the transmitted light scanning image, and a reflected light image corresponding to the target defect location is acquired based on the reflected light scanning image. Then, the defect type of the target defect location is determined based on the acquired transmitted light and reflected light images. By locating the defect location based on the transmitted light scanning images and / or reflected light scanning images, and matching the corresponding transmitted light and reflected light images for each target defect location, the defect type is identified by utilizing the difference in the defect's response under different optical modes. This solves the technical problem of related technologies that can only determine the existence of defects but cannot clearly define the defect type. On the other hand, by combining scanning images under different optical modes for detection, the missed detection of defects is reduced, and the accuracy and sensitivity of defect type detection are improved. Thus, the source of defects can be determined based on the defect type, and targeted improvements can be made to address the defect source, thereby improving the quality of the cast film. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is an application environment diagram of a defect detection method for cast film in one embodiment;

[0027] Figure 2 This is a flowchart illustrating a defect detection method for cast film in one embodiment;

[0028] Figure 3 This is a flowchart illustrating the process of determining the type of defect at the location of a target defect in one embodiment.

[0029] Figure 4A This is a schematic diagram illustrating the numerical distribution of the fractal dimension of the first defect in one embodiment;

[0030] Figure 4B This is a schematic diagram illustrating the numerical distribution of the second defect fractal dimension in one embodiment;

[0031] Figure 5A This is a schematic diagram of the distribution of the first grayscale percentage in one embodiment;

[0032] Figure 5B This is a schematic diagram of the distribution of the second grayscale ratio in one embodiment;

[0033] Figure 6A A transmitted light image of a dense impurity defect in one embodiment;

[0034] Figure 6B A reflected light image of a dense impurity defect in one embodiment;

[0035] Figure 7A A transmitted light image of a gel defect in one embodiment;

[0036] Figure 7B A transmitted light image of a gel defect in one embodiment;

[0037] Figure 8 This is a flowchart illustrating the process of determining the type of defect at the location of a target defect in one embodiment.

[0038] Figure 9A A transmitted light image of a fibrous defect in one embodiment;

[0039] Figure 9B An image of reflected light from a fibrous defect in one embodiment;

[0040] Figure 10 A schematic diagram of the target sample;

[0041] Figure 11 This is a flowchart illustrating a method for detecting defects in cast film in another embodiment;

[0042] Figure 12 This is a schematic diagram of the structure of a cast film defect detection device in one embodiment;

[0043] Figure 13 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0046] For high-voltage and ultra-high-voltage power cables, under the action of high electric field strength, impurities in the insulation material (such as metal fragments, inorganic dust, fiber residues, etc.) are prone to cause local electric field distortion, accelerate insulation aging, and long-term cumulative effects can also lead to early breakdown of the insulation layer, affecting the safe and stable operation of the power grid and the service life of the cable.

[0047] Therefore, the extrusion processing performance of power cable insulation materials is crucial. To form a uniform, dense, and defect-free insulation layer, it is necessary to prevent the polymer from scorching due to overheating or excessive shearing, which would produce infusible gels (i.e., scorch residues) and thus infiltrate the insulation material, creating inherent defects. Scorch residues not only reduce the mechanical properties of the insulation material but also form conductive channels during cable operation, triggering partial discharges. Extrusion processes in related technologies can suppress scorching to some extent by controlling temperature and shear rate; however, the scorching phenomenon is essentially more dependent on the intrinsic properties of the polymer material. Therefore, material improvement can be achieved by identifying the scorch residues.

[0048] Currently, defect detection methods for polymer insulating materials mainly include manual visual inspection, automated optical inspection based on ordinary cameras, and ultrasonic testing. However, in terms of detection accuracy and recognition capability, traditional optical methods are limited by resolution and contrast, and are insensitive to gels with high transparency and small size, while ultrasonic testing has limited ability to distinguish charred material with an acoustic impedance close to that of the matrix. Secondly, in terms of detection speed, high-precision offline inspection is difficult to match the high-speed continuous production rhythm of extrusion production lines, resulting in delayed quality feedback. Most importantly, related technologies generally lack intelligent defect type identification functions; they can only detect the existence of abnormal points, but cannot specifically distinguish defects from different sources or types. Therefore, it is difficult to trace the cause of defects, and consequently, it is difficult to improve the material.

[0049] The defect detection method for cast film provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located on the cloud or other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0050] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting defects in cast film is provided, which is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S201 to S205. Among them:

[0051] S201, acquire the transmitted light scanning image and the reflected light scanning image of the cast film to be tested.

[0052] In one embodiment, the cast film to be tested can be a film obtained by heating an insulating material in an extrusion apparatus and then extruding it from an extrusion die, which can be used to prepare a cable insulation layer.

[0053] Optionally, the extrusion device includes a transmitted light imaging unit and a reflected light imaging unit. The transmitted light imaging unit includes a transmitted light source and a transmitted light acquisition camera, and the reflected light imaging unit includes a reflected light source and a reflected light acquisition camera. The reflected light acquisition camera and the transmitted light acquisition camera can be CCD (Charge Coupled Device) high-speed cameras.

[0054] For example, the transmitted light scan image is a digital image obtained by converting the transmitted light signal generated by the transmitted light source penetrating the cast film under test during the continuous extrusion process. The reflected light scan image is a digital image obtained by converting the reflected light signal generated by the reflected light source reflecting off the surface of the cast film under test during the continuous extrusion process. Furthermore, the transmitted light and reflected light cameras have high shooting frequencies, thus enabling the acquisition of transmitted light and reflected light scan images at various locations during the extrusion process of the cast film under test.

[0055] S202, Identify the location of defects in the cast film under test based on the transmitted light scanning image and / or reflected light scanning image.

[0056] In one embodiment, defects may include contaminant impurities (such as metallic impurities, dust, etc.) and gels generated by scorching. That is, the materials of the defects in the cast film under test are different from those of the cast film itself. Therefore, the density, refractive index, and structural compactness of the cast film under test and the defects are different. The presence of defects in the transmitted light scanning images and / or reflected light scanning images can be determined by preprocessing the transmitted light scanning images and / or reflected light scanning images, thereby identifying the location of defects in the cast film under test.

[0057] For example, the preprocessing of transmitted light scan images and / or reflected light scan images can first involve filtering the grayscale transmitted light scan images and / or reflected light scan images to eliminate local noise interference. Then, based on the mean and variance of grayscale values ​​in each local region of the transmitted light scan images and / or reflected light scan images, a binarization threshold is dynamically set to adjust pixels with grayscale values ​​below the local threshold to black and pixels with grayscale values ​​above the local threshold to white. This transforms the transmitted light scan images and / or reflected light scan images corresponding to the defect location from grayscale images into binarized images where the defect area is pure black and the background is pure white, thus breaking the grayscale correlation between the defect and the background. For locations that do not contain defects, the grayscale values ​​of each pixel in the corresponding transmitted light scan images and / or reflected light scan images may not have significant differences globally. Therefore, the binarized image corresponding to non-defect locations may be a completely white image.

[0058] S203, Based on the transmitted light scanning image, obtain the transmitted light image corresponding to the location of the target defect.

[0059] In some embodiments, the target defect location is any one of the defect locations in the cast film to be tested.

[0060] Optionally, based on the binarized image obtained above, residual noise in the binarized image can be filtered out. Specifically, morphological opening and morphological closing operations can be used. Morphological opening first shrinks the defective region in the image, eliminates isolated dark spots, and then restores the defective region to its original size, preserving the outline of the defective region. Similarly, morphological closing first enlarges the defective region, fills in the bright spots within the defective region, and then shrinks the defective region to its original size, making the defective region a complete and continuous black pixel block.

[0061] For example, to clearly define the specific location of the defect region in the binarized image and avoid introducing irrelevant pixels in subsequent analysis, an edge detection algorithm can be used to identify the boundary pixels between the black defect region and the white background region in the denoised binarized image, and then connect each boundary pixel one by one to obtain the defect contour. However, the size of the defect region varies in different binarized images. To unify the subsequent analysis scale, the binarized image with the largest defect contour size can be selected, and the analysis region size of each binarized image can be determined based on the defect contour in the selected binarized image. Subsequently, the corresponding analysis region size is cropped from the binarized image to obtain the transmitted light image.

[0062] S204. Based on the reflected light scanning image, obtain the reflected light image corresponding to the location of the target defect.

[0063] In one embodiment, the process of obtaining the reflected light image corresponding to the target defect location can refer to the process of obtaining the transmitted light image corresponding to the target defect location in S203 above, and will not be repeated here.

[0064] S205. Based on the transmitted light image and the reflected light image, determine the defect type at the target defect location.

[0065] In some embodiments, transmitted light images can reflect the effect of defects on transmitted light sources. Specifically, high-density impurities strongly block transmitted light sources, while translucent gels scatter them. Therefore, internal property information of defects (such as density, transparency, and internal compactness) can be obtained from transmitted light images.

[0066] In some embodiments, reflected light images can reflect the effect of defects on the reflected light source. Specifically, smooth-surfaced metallic impurities produce specular reflection, while rough-surfaced gels produce diffuse reflection. Therefore, surface morphology information of defects (e.g., surface roughness, refractive index, surface shape) can be obtained from reflected light images.

[0067] For example, different types of defects may have the same or similar internal attribute information, and they may also have the same or similar surface morphology information. However, the internal attribute information and surface morphology information of defects of the same type have certain similarities. Therefore, the reflected light image and transmitted light image of the target defect location can be analyzed simultaneously to determine the defect type of the target defect location based on the obtained internal attribute information and surface morphology information, thus overcoming the technical problem that it is difficult to accurately identify the defect type with a single illumination mode.

[0068] Furthermore, after determining the defect type at the target defect location, the defect source can be analyzed based on the defect type, thereby adjusting the insulation material or extrusion device to prepare a cast film with higher purity to meet the higher quality requirements of cable insulation layers.

[0069] For example, the types of gel defects include fiber impurity defects, dense impurity defects, and gel impurity defects. Among them, the main source of fiber impurity defects is the low cleanliness of the extrusion environment, the main source of dense impurity defects is the low purity of the raw materials, and the main source of gel impurity defects is improper setting of the kinetic parameters or process parameters of the extrusion device.

[0070] The aforementioned method for detecting defects in cast film identifies the location of defects by acquiring transmitted light and reflected light scanning images of the cast film under test. For each target defect location, a transmitted light image corresponding to the target defect location is acquired based on the transmitted light scanning image, and a reflected light image corresponding to the target defect location is acquired based on the reflected light scanning image. Then, the defect type of the target defect location is determined based on the acquired transmitted light and reflected light images. By locating the defect location based on the transmitted light and / or reflected light scanning images, and matching the corresponding transmitted light and reflected light images for each target defect location, the method identifies the defect type by utilizing the difference in the defect's response under different optical modes. This solves the technical problem of related technologies that can only determine the existence of defects but cannot determine the defect type. On the other hand, combining scanning images under different optical modes for detection reduces missed defects and improves the accuracy and sensitivity of defect type detection. Thus, the source of defects can be determined based on the defect type, allowing for targeted improvements to the defect source and thus improving the quality of the cast film.

[0071] In one exemplary embodiment, such as Figure 3 As shown, S205 above includes:

[0072] S301, Based on the transmitted light image, determine the first defect fractal dimension and the first grayscale ratio corresponding to the target defect location.

[0073] To determine the quantification parameters that can reflect the differences in the response of defects under different optical modes, sample cast films were selected, and various quantification parameters in the transmitted light images and reflected light images of different defect locations in the sample cast films were analyzed.

[0074] In one embodiment, the first defect fractal dimension can reflect the geometric complexity and space-filling ability of the defect under the action of a transmitted light source.

[0075] For example, the transmitted light image can be converted into a gridded image according to a preset grid scale. Then, square pixel blocks with gradually increasing side lengths are used to cover the defect area in sequence, and the number of square pixel blocks with different side lengths required to cover the defect area is recorded, so as to determine the first defect fractal dimension based on the recorded number.

[0076] Optionally, the process for determining the fractal dimension of the first defect can be as follows: (1).

[0077] lnN(ε) = D1×ln(1 / ε) + C (1)

[0078] Where ε is the side length of the square pixel block, N(ε) represents the number of square pixel blocks with a side length of ε required to cover the defect area, D1 is the fractal dimension of the first defect, and C is a constant term.

[0079] In one embodiment, the first grayscale ratio can reflect the grayscale distribution of each pixel under the action of a transmitted light source.

[0080] For example, the first grayscale ratio R1 can be calculated using the following formula (2).

[0081] R1=N l / N a (2)

[0082] Where, N l N represents the number of pixels in the transmitted light image whose gray level is lower than a preset gray level threshold. a This represents the total number of pixels in the transmitted light image.

[0083] S302, Based on the reflected light image, determine the second defect fractal dimension and the second grayscale ratio corresponding to the target defect location.

[0084] In one embodiment, the second defect fractal dimension can reflect the geometric complexity and space-filling ability of the defect under the influence of a reflected light source.

[0085] It should be noted that the specific process for determining the fractal dimension of the second defect is similar to that for the first defect, and will not be elaborated here.

[0086] In one embodiment, the second grayscale ratio can reflect the grayscale distribution of each pixel under the influence of a reflected light source.

[0087] It should be noted that the specific process for determining the second grayscale ratio is similar to that for the first grayscale ratio, and will not be elaborated here.

[0088] For example, the first defect fractal dimension and the first grayscale ratio are calculated for transmitted light images at different defect locations, and the second defect fractal dimension and the second grayscale ratio are calculated for reflected light images at different defect locations, resulting in the following: Figure 4A , Figure 4B , Figure 5A and Figure 5B .

[0089] The sample cast film contained a total of 1757 defects, including 1285 gel defects, 450 dense impurity defects, and the remaining 25 fibrous impurity defects.

[0090] Figure 4A A schematic diagram of the numerical distribution of the fractal dimension of the first defect is shown.

[0091] Figure 4B A schematic diagram of the numerical distribution of the fractal dimension of the second defect is shown.

[0092] like Figure 4A and Figure 4B As shown, the horizontal axis represents the numerical value of the fractal dimension of the first defect or the fractal dimension of the second defect, and the vertical axis represents the number of defects n in the sample cast film.

[0093] Therefore, it can be seen that the numerical distribution of the first defect fractal dimension of most defects is relatively concentrated, while the distribution of the second defect fractal dimension of each defect has a relatively clear boundary, and the numerical value of the boundary is... Figure 4B The dividing line corresponds to 1.95.

[0094] like Figure 5A and 5B The horizontal axis represents the first grayscale ratio or the second grayscale ratio, and the vertical axis represents the number of defects n in the sample cast film.

[0095] It can be seen that the first grayscale ratio distribution of each transmitted light scan image in the sample cast film is relatively concentrated and there is no clear boundary. At the same time, the second grayscale ratio distribution of each reflected light scan image in the sample cast film is also relatively concentrated and there is no clear boundary.

[0096] S303, in response to the fact that the fractal dimension of the first defect, the fractal dimension of the second defect, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio all meet the preset conditions, the defect type is determined to be a dense impurity defect.

[0097] In one embodiment, in conjunction with the above Figure 4A , Figure 4B , Figure 5A and Figure 5B Dense impurity defects include inorganic materials such as metal particles, carbon black, and dust. Therefore, the outlines of these defects are sharp and compact in transmitted light and / or reflected light images, meaning that the fractal dimensions of the first and second defects are relatively high. On the other hand, dense impurity defects have strong light-blocking properties, meaning that the values ​​of the first and second grayscale proportions are similar.

[0098] Therefore, based on the patterns of the first defect fractal dimension, the second defect fractal dimension, the first grayscale ratio, and the second grayscale ratio of the dense impurity defect, preset conditions can be set. Thus, when the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio all meet the preset conditions, the defect type can be determined to be a dense impurity defect.

[0099] S304, in response to at least one of the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio not meeting the preset conditions, the defect type is determined to be a gel defect.

[0100] In one embodiment, in conjunction with the above Figure 4A , Figure 4B , Figure 5A and Figure 5B Gel defects are loose clusters formed in the cast film under test due to uneven cross-linking of polymers in the raw materials. These defects contain numerous micropores, resulting in blurred outlines of the defect areas in transmitted and / or reflected light images; that is, the fractal dimensions of the first and second defects are relatively low. Furthermore, gel defects are essentially semi-transparent media, causing scattering and refraction of transmitted light and reflection of reflected light on their surface. Therefore, gel defects have different effects on the two light sources, resulting in significant differences in the grayscale percentages of the first and second defects.

[0101] Therefore, if at least one of the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first and second grayscale ratios does not meet the preset conditions, the defect type can be determined to be a gel defect.

[0102] To better understand the differences between the defect regions in the transmitted light and reflected light images of dense impurity defects and gel defects, the following will be conducted through... Figure 6A , Figure 6B , Figure 7A and Figure 7B To illustrate.

[0103] like Figure 6A The image shows a transmitted light image of a dense impurity defect, where the defect area appears as a well-defined dark spot.

[0104] like Figure 6B The image shows a reflected light image of a dense impurity defect, in which the boundary between the defect area and the background area is relatively clear.

[0105] like Figure 7A The image shows a transmitted light image of a gel defect, in which the defect area appears as a blurry shadow, and the boundary between the defect area and the background area is blurry and irregular.

[0106] like Figure 7B The image shows a reflected light image of a gel defect, in which the defect area appears as a blurry shadow, and the boundary between the defect area and the background area is blurry and irregular.

[0107] The above calculation of the first defect fractal dimension and first grayscale ratio of the transmitted light image, and the second defect fractal dimension and second grayscale ratio of the reflected light image, quantifies the effect of defects on the reflected and transmitted light sources. Thus, based on the different characteristics of the effects of different types of defects on the reflected and transmitted light sources, preset conditions are set, and the defect type is determined by the relationship between the values ​​of the first defect fractal dimension, first grayscale ratio, second defect fractal dimension and second grayscale ratio and the preset conditions.

[0108] For example, the preset conditions may include a first preset condition and a second preset condition, wherein the first preset condition is that the fractal dimension of the first defect is greater than a preset value of the fractal dimension and the fractal dimension of the second defect is greater than a preset value of the fractal dimension, and the second preset condition may be that the difference between the grayscale ratios of the first grayscale ratio and the second grayscale ratio is less than a preset value of the difference.

[0109] Accordingly, S303 may include determining the defect type as a dense impurity defect in response to the fact that the fractal dimension of the first defect is greater than a preset value of the fractal dimension, the fractal dimension of the second defect is greater than a preset value of the fractal dimension, and the difference between the grayscale proportions of the first grayscale proportion and the second grayscale proportion is less than a preset value of the difference.

[0110] Accordingly, S304 may include determining the defect type as a gel defect in response to either the first defect fractal dimension or the second defect fractal dimension being less than or equal to a preset value for the fractal dimension, or the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio being greater than or equal to a preset value for the difference.

[0111] It should be noted that the preset values ​​of fractal dimension and difference can be flexibly adjusted according to the proportion of each material in the polymer raw material and the preparation requirements, and this application does not impose any restrictions.

[0112] For example, the preset value for fractal dimension can be 1.95, and the preset value for difference can be 0.02.

[0113] In some embodiments, if the proportion of gel defects is high among the defect types corresponding to each defect location, it indicates that the current crosslinking reaction rate is too fast, and the kinetic parameters or process parameters of the extrusion device can be adjusted; if the proportion of dense impurity defects is high among the defect types corresponding to each defect location, the raw material batch can be changed; if the proportion of fibrous impurity defects is high among the defect types corresponding to each defect location, the extrusion environment can be cleaned.

[0114] The above-mentioned method establishes an objective and quantitative standard for determining defect types by setting specific predetermined conditions and the values ​​of each preset value in the predetermined conditions. This breaks through the technical barrier that related technologies cannot distinguish between foreign impurities and gels caused by scorching. It realizes the automatic identification of different types of defects in the cast film to be tested. Then, the raw materials, extrusion device or extrusion environment can be adjusted in a targeted manner according to the proportion of different types of defects in the cast film to be tested, so as to improve the quality of the cast film obtained in the subsequent preparation.

[0115] Based on the above embodiments, before performing S205, the cast film defect detection method further includes: identifying whether the defect at the target defect location is a fibrous impurity defect.

[0116] For example, fibrous impurity defects can include fabric fibers, hair, etc., meaning that fibrous impurity defects have an irregular and non-compact morphology with a high aspect ratio. Therefore, the shape of the defect region in the transmitted light and reflected light images corresponding to fibrous impurity defects differs significantly from the shape of the defect region in the transmitted light and reflected light images corresponding to approximately circular defects such as dense impurity defects and gel defects.

[0117] In other words, after acquiring the transmitted light image and the reflected light image, first determine whether the defect at the target defect location is a fibrous impurity defect, and after determining that the defect at the target defect location is a non-fibrous impurity defect, execute the above S301~S304 to identify whether the defect type at the target defect location is a gel defect or a dense impurity defect.

[0118] In one exemplary embodiment, such as Figure 8 As shown, the process for identifying fibrous impurity defects includes:

[0119] S801, based on the transmitted light image, determine the first defect morphology parameters corresponding to the target defect location.

[0120] Among them, the first defect morphology parameter is used to quantify the shape, size, and contour of the defect region in the transmitted light image of the target defect location.

[0121] For example, features can be extracted from the transmitted light image of the target defect location, and the extracted features can be analyzed to determine the first defect morphology parameters.

[0122] S802, based on the reflected light image, determine the second defect morphology parameters corresponding to the target defect location.

[0123] The second defect morphology parameter is used to quantify the shape, size, and contour of the defect region in the reflected light image of the target defect location.

[0124] For example, features can be extracted from the reflected light image of the target defect location, and the extracted features can be analyzed to determine the morphological parameters of the second defect.

[0125] S803, based on the morphological difference between the first defect morphology parameter and the second defect morphology parameter, identify whether the defect type is a fibrous impurity defect.

[0126] Optionally, the first defect morphology parameter reflects the effect of the defect on the transmitted light source, and the second defect morphology parameter reflects the effect of the defect on the reflected light source. Therefore, by analyzing the morphological differences between the first and second defect morphology parameters, the mechanism of the defect's effect on different light sources can be analyzed, and it can be determined whether it is a fibrous defect.

[0127] In one embodiment, both the first defect morphology parameter and the second defect morphology parameter include: roundness and longest axis; wherein, identifying whether the defect type is a fibrous impurity defect based on the morphological difference between the first defect morphology parameter and the second defect morphology parameter includes: determining the defect type as a fibrous impurity defect in response to the first defect morphology parameter having a roundness < 0.3 and a longest axis > 20, and the second defect morphology parameter having a roundness < 0.2 and a longest axis > 50.

[0128] Optionally, roundness is used to characterize the degree to which the shape of the defect area approximates a circle, and can be calculated by the following formula (3).

[0129] C = (4 × π × A) / p 2 (3)

[0130] Where C represents the roundness, A represents the area of ​​the defect region, and p represents the perimeter of the defect region.

[0131] For example, the longest axis is used to characterize the elongation of the shape of the defect region. Specifically, the longest axis is the longest line segment that passes through the geometric center of the defect region and whose two ends fall on the boundary of the defect region. It can be calculated by the following formula (4).

[0132] L = 2 × max(a, b) (4)

[0133] Where L represents the longest axis, a and b represent the lengths of the major and minor axes of the ellipse, respectively, and a and b satisfy the following formula (5) for the equation of the ellipse.

[0134] (5)

[0135] Where h represents the x-axis coordinate of the center of the ellipse, k represents the y-axis coordinate of the center of the ellipse, and θ represents the angle between the major axis of the ellipse and the horizontal direction.

[0136] For example, if the roundness of the first defect morphology parameter is <0.3 and the longest axis is >20, and the roundness of the second defect morphology parameter is <0.2 and the longest axis is >50, the defect type can be determined to be a fibrous impurity defect. Conversely, if at least one of the first and second defect morphology parameters does not meet the above conditions of roundness and longest axis, the defect type is determined not to be a fibrous impurity defect, and the defect type is further identified as a dense impurity defect or a gel defect.

[0137] like Figure 9A The image shows a transmitted light image of a fibrous defect, in which the defect area is shaped as a thin strip of shadow, and the boundary between the defect area and the background area is clear because the fibrous impurities have a certain blocking effect on the transmitted light source.

[0138] like Figure 9B The image shows a reflected light image of a fibrous defect, in which the defect area is shaped as a thin strip of shadow, and a slight halo appears at the boundary between the defect area and the background area due to the scattering effect of the fibrous impurities on the reflected light source.

[0139] The above embodiments extract two core morphological parameters of defects—circularity and longest axis—from transmitted and reflected light images, and construct judgment rules for fibrous impurity defects by selecting appropriate quantization thresholds. This achieves automatic and accurate identification of fibrous impurity defects, effectively solving the problems in related technologies where detection methods struggle to distinguish fibrous impurities from other defects and rely on subjective experience, leading to misjudgments and omissions. It improves the objectivity and accuracy of defect classification, providing a reliable basis for tracing the root cause of defects and guiding the optimization of production processes.

[0140] Furthermore, before starting to capture the transmitted light scanning image and the reflected light scanning image, the above-mentioned cast film defect detection method also includes: dynamically calibrating the first acquisition time of the transmitted light image corresponding to the defect position in the target sample in the transmitted light scanning image, and the second acquisition time of the reflected light image corresponding to the defect position in the target sample in the reflected light scanning image; and determining the image acquisition delay based on the time difference between the second acquisition time and the first acquisition time and the transmission speed of the target sample.

[0141] In one embodiment, there is a gap between the transmitted light imaging unit and the reflected light imaging unit. Therefore, even if the transmitted light imaging unit and the reflected light imaging unit acquire transmitted light scanning images and reflected light scanning images at the same acquisition frequency, the transmitted light scanning images and reflected light scanning images acquired at the same time do not correspond to the same defect location. Therefore, calibration can be performed before the transmission light scanning images and reflected light scanning images are captured to determine the image acquisition delay between the transmission light scanning images and reflected light scanning images.

[0142] like Figure 10 As shown, the target sample is a transparent thin plate 100, which is cast from colorless transparent epoxy resin or colorless transparent plexiglass. The transparent thin plate 100 contains multiple simulated defects, including simulated fibrous impurity defects 1001, simulated gel defects 1002, and simulated dense impurity defects 1003. The simulated fibrous impurity defect 1001 can be a metal wire, the simulated gel defect 1002 can be a frosted glass ball, and the simulated dense impurity defect 1003 can be a steel ball.

[0143] Theoretically, with a distance of d between the transmitted light imaging unit and the reflected light imaging unit and a transport speed of v1 for the target sample, the image acquisition delay for the same defect location in the target sample by the transmitted light imaging unit and the reflected light imaging unit is d / v1. However, during the actual transport of the target sample, the image acquisition delay may have some deviation. Therefore, the image acquisition delay can be obtained by calibrating the imaging experiment using the first acquisition time and the second acquisition time. In other words, the actual imaging time delay of the transmitted light imaging unit and the reflected light imaging unit at the same defect location in the target sample is d / v1+. .

[0144] For example, based on the determined image acquisition delay, a transmitted light image corresponding to the target defect location is obtained based on the transmitted light scanning image; a reflected light image corresponding to the target defect location is obtained based on the reflected light scanning image; the method further includes: obtaining the transport speed of the cast film under test; and obtaining the transmitted light image and reflected light image corresponding to the target defect location from the transmitted light scanning image and the reflected light scanning image according to the image acquisition delay and the transport speed of the cast film under test.

[0145] Given that the conveying speed of the cast film under test is v2 and the distance between the transmitted light imaging unit and the reflected light imaging unit is d, the image acquired at time t0 in the transmitted light scanning image and the image acquired at time t0 + d / v2 + d in the reflected light scanning image are compared. The images correspond to the same defect location.

[0146] The above embodiments, through dynamic calibration of the target sample, obtain the image acquisition delay of the transmitted light imaging unit and the reflected light imaging unit for the same defect. Thus, based on the determined image acquisition delay and the transmission speed of the cast film under test, the transmitted light image and the reflected light image corresponding to the same defect in the cast film under test can be accurately matched. This solves the image acquisition deviation caused by the physical distance between the transmitted light imaging unit and the reflected light imaging unit, mechanical transmission error, and circuit delay. It provides a precise data association basis for subsequent defect type identification, improving the accuracy and reliability of defect classification.

[0147] In one exemplary embodiment, such as Figure 11 As shown, the above-mentioned method for detecting defects in cast film includes:

[0148] S1101, acquire the transmitted light scanning image and the reflected light scanning image of the cast film to be tested.

[0149] S1102, Identify the location of defects in the cast film under test based on the transmitted light scanning image and / or reflected light scanning image.

[0150] S1103, obtain the transfer speed of the cast film to be tested.

[0151] S1104, based on the image acquisition delay and the conveying speed of the cast film under test, acquire the transmitted light image and reflected light image corresponding to the target defect location from the transmitted light scanning image and the reflected light scanning image.

[0152] S1105, Based on the transmitted light image, determine the first defect morphology parameters corresponding to the target defect location.

[0153] S1106, Based on the reflected light image, determine the second defect morphology parameters corresponding to the target defect location.

[0154] S1107, responding to the first defect morphology parameter where the roundness is <0.3 and the longest axis is >20, and the second defect morphology parameter where the roundness is <0.2 and the longest axis is >50, the defect type is determined to be a fibrous impurity defect.

[0155] After identifying the defect at the target defect location as a non-fibrous impurity defect, execute the following S1108.

[0156] S1108, Based on the transmitted light image, determine the first defect fractal dimension and the first grayscale ratio corresponding to the target defect location.

[0157] S1109, Based on the reflected light image, determine the second defect fractal dimension and the second grayscale ratio corresponding to the target defect location.

[0158] S1110, responding to the fact that the fractal dimension of the first defect is greater than the preset value of the fractal dimension, the fractal dimension of the second defect is greater than the preset value of the fractal dimension, and the difference between the grayscale proportions of the first grayscale proportion and the second grayscale proportion is less than the preset value of the difference, the defect type is determined to be a dense impurity defect.

[0159] S1111, in response to either the first defect fractal dimension or the second defect fractal dimension being less than or equal to a preset value for the fractal dimension, or the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio being greater than or equal to a preset value for the difference, the defect type is determined to be a gel defect.

[0160] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0161] Based on the same inventive concept, this application also provides a casting film defect detection device for implementing the casting film defect detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more casting film defect detection device embodiments provided below can be found in the limitations of the casting film defect detection method described above, and will not be repeated here.

[0162] In one exemplary embodiment, such as Figure 12 As shown, a defect detection device for cast film is provided, comprising: a data acquisition module, a defect identification module, a defect determination module, and a velocity measurement unit, wherein:

[0163] The acquisition module includes a transmitted light imaging unit and a reflected light imaging unit. The transmitted light imaging unit is used to acquire the transmitted light scanning image of the cast film under test, and the reflected light imaging unit is used to acquire the reflected light scanning image of the cast film under test. The transmitted light imaging unit includes a transmitted light source 1214 and a transmitted light acquisition camera 1215, and the reflected light imaging unit includes a reflected light source 1216 and a reflected light acquisition camera 1217.

[0164] In this system, the central axes of the lenses of both the transmitted light acquisition camera 1215 and the reflected light acquisition camera 1217 are always perpendicular to the horizontal line. The center-to-center distance between the lenses of each camera is less than or equal to 600 mm and greater than or equal to 200 mm. Too large a distance would result in a large device size, while too small a distance would cause interference between the transmitted light source 1214 and the reflected light source 1216. The transmitted light source 1214, located below the cast film, can be a strip-shaped LED (Light-Emitting Diode) light source, allowing its light path to penetrate the cast film. The transmitted light acquisition camera 1215, located above the cast film, converts the light signal from the transmitted light source 1214 after passing through the cast film into a digital image. The reflected light source 1216, also located above the cast film, can be a ring-shaped LED light source. The reflected light acquisition camera 1217 converts the light signal from the reflected light source 1216 after passing through the cast film into a digital image.

[0165] The acquisition module also includes a speed measurement unit, which includes a speed measurement rotor 1221, used to acquire the transmission speed of the target sample position and the transmission speed of the cast film to be tested.

[0166] The acquisition module is also used to dynamically calibrate the first acquisition time of the transmitted light image corresponding to the defect location in the target sample in the transmitted light scanning image, and the second acquisition time of the reflected light image corresponding to the defect location in the target sample in the reflected light scanning image; and to determine the image acquisition delay based on the time difference between the second acquisition time and the first acquisition time and the transmission speed in the target sample.

[0167] The defect identification module, connected to the transmitted light imaging unit and the reflected light imaging unit in the acquisition module, is used to identify the location of defects in the cast film under test based on the transmitted light scanning image and / or the reflected light scanning image.

[0168] The defect processing module, connected to the defect identification module, is used to: obtain the transmitted light image corresponding to the target defect location based on the transmitted light scanning image, obtain the reflected light image corresponding to the target defect location based on the reflected light scanning image, and determine the defect type at the target defect location based on the transmitted light image and the reflected light image.

[0169] The defect processing module is also connected to the speed measurement unit, which is used to: obtain the transmitted light image and reflected light image corresponding to the target defect location from the transmitted light scanning image and the reflected light scanning image based on the image acquisition delay and the conveying speed of the cast film under test.

[0170] Specifically, the cast film defect detection device includes an extruder platform 1201, an extruder display 1202, a drive shaft 1203, a feed port 1204, an extruder screw and barrel 1205, an extrusion die 1206, a detection instrument platform 1207, a detection instrument control console 1208, a detection instrument display 1209, a cast film 1210, a cast roller 1211, an automatic edge trimming device 1212, a positioning rotor 1213, a transmitted light source 1214, a transmitted light acquisition camera 1215, a reflected light source 1216, a reflected light acquisition camera 1217, an insulating isolation cover 1218, a high-voltage electrode 1219, a grounding electrode 1220, a speed measuring rotor 1221, an edge waste 1222, a cast film collection device 1223, and a waste collection device 1224.

[0171] The extruder platform 1201 provides mechanical support for the extruder display 1202, drive shaft 1203, feed port 1204, extruder screw, and barrel 1205. Cable insulation material is fed into the extrusion equipment through the feed port 1204, heated and melted by multiple temperature zones in the extruder screw and barrel 1205, and then extruded through the extrusion die 1206 to obtain a cast film 1210. The extruder display 1202 displays the heating temperature and other process parameters of each temperature zone in the extruder screw and barrel 1205. The casting roller 1211 drives the cast film 1210, and the automatic edge trimming device 1212 removes uneven or irregular portions of the film during transmission to ensure a consistent width. The positioning rotor 1213 positions the transmission path of the cast film.

[0172] The testing instrument platform 1207 provides mechanical support for the testing instrument control console 1208, testing instrument display 1209, cast film 1210, cast roller 1211, automatic edge trimming device 1212, positioning rotor 1213, transmitted light source 1214, transmitted light acquisition camera 1215, reflected light source 1216, reflected light acquisition camera 1217, insulating cover 1218, high voltage electrode 1219, grounding electrode 1220, speed measuring rotor 1221, edge waste 1222, cast film collection device 1223, and waste collection device 1224. The testing instrument control console 1208 controls the transmitted light source 1214, transmitted light acquisition camera 1215, reflected light source 1216, and reflected light acquisition camera 1217, while the testing instrument display 1209 displays the transmitted light scanning image and the reflected light scanning image in real time.

[0173] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 13As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores transmitted light scan images and reflected light scan images. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for detecting defects in cast film.

[0174] Those skilled in the art will understand that Figure 13 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0175] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0176] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0177] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0178] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0179] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0180] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting defects in cast film, characterized in that, The method includes: Acquire transmitted light scanning images and reflected light scanning images of the cast film under test; Based on the transmitted light scanning image and / or the reflected light scanning image, identify the location of defects in the cast film under test; Based on the transmitted light scanning image, obtain the transmitted light image corresponding to the target defect location; Based on the reflected light scan image, obtain the reflected light image corresponding to the location of the target defect; Based on the transmitted light image and the reflected light image, the defect type at the target defect location is determined.

2. The method according to claim 1, characterized in that, Determining the defect type at the target defect location based on the transmitted light image and the reflected light image includes: Based on the transmitted light image, determine the first defect fractal dimension and the first grayscale ratio corresponding to the target defect location; Based on the reflected light image, determine the second defect fractal dimension and the second grayscale ratio corresponding to the target defect location; In response to the fact that the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio all meet the preset conditions, the defect type is determined to be a dense impurity defect. If at least one of the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio does not meet the preset condition, the defect type is determined to be a gel defect.

3. The method according to claim 2, characterized in that, The step of determining the defect type as a dense impurity defect in response to the fact that the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio all satisfy preset conditions includes: determining the defect type as a dense impurity defect in response to the fact that the first defect fractal dimension is greater than a preset value of fractal dimension, the second defect fractal dimension is greater than the preset value of fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio is less than a preset value of difference. The step of determining the defect type as a gel defect in response to at least one of the first defect fractal dimension, the second defect fractal dimension, and the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio not satisfying the preset condition includes: determining the defect type as a gel defect in response to either the first defect fractal dimension or the second defect fractal dimension being less than or equal to the preset value of the fractal dimension, or the difference in grayscale ratio between the first grayscale ratio and the second grayscale ratio being greater than or equal to the preset value of the difference.

4. The method according to claim 3, characterized in that, The preset value for the fractal dimension is 1.95; the preset value for the difference is 0.

025.

5. The method according to claim 1, characterized in that, Before determining the defect type of the target defect location based on the transmitted light image and the reflected light image, the method further includes: Identify whether the defect at the target defect location is a fibrous impurity defect; The step of determining the defect type at the target defect location based on the transmitted light image and the reflected light image is performed after identifying the defect at the target defect location as a non-fibrous impurity defect.

6. The method according to claim 5, characterized in that, The step of identifying whether the defect at the target defect location is a fibrous impurity defect includes: Based on the transmitted light image, determine the first defect morphology parameters corresponding to the location of the target defect; Based on the reflected light image, determine the second defect morphology parameters corresponding to the location of the target defect; Based on the morphological differences between the first defect morphology parameter and the second defect morphology parameter, the defect type is identified as a fibrous impurity defect.

7. The method according to claim 6, characterized in that, Both the first defect morphology parameter and the second defect morphology parameter include: roundness and longest axis; wherein, the step of identifying whether the defect type is a fibrous impurity defect based on the morphological difference between the first defect morphology parameter and the second defect morphology parameter includes: In response to the first defect morphology parameter having a roundness of <0.3 and a longest axis of >20, and the second defect morphology parameter having a roundness of <0.2 and a longest axis of >50, the defect type is determined to be a fibrous impurity defect.

8. The method according to claim 1, characterized in that, Also includes: The first acquisition time of the transmitted light image corresponding to the defect location in the target sample in the transmitted light scanning image, and the second acquisition time of the reflected light image corresponding to the defect location in the target sample in the reflected light scanning image are dynamically calibrated. The image acquisition delay is determined based on the time difference between the second acquisition time and the first acquisition time, as well as the transmission speed in the target sample. The step of obtaining a transmitted light image corresponding to the target defect location based on the transmitted light scanning image and obtaining a reflected light image corresponding to the target defect location based on the reflected light scanning image further includes: Obtain the conveying speed of the cast film to be tested; Based on the image acquisition delay and the transmission speed of the cast film under test, the transmitted light image and the reflected light image corresponding to the target defect location are obtained from the transmitted light scan image and the reflected light scan image.

9. A defect detection device for cast film, characterized in that, The device includes: The acquisition module includes a transmitted light imaging unit and a reflected light imaging unit; the transmitted light imaging unit is used to acquire a transmitted light scanning image of the cast film under test, and the reflected light imaging unit is used to acquire a reflected light scanning image of the cast film under test. A defect identification module, connected to the transmitted light imaging unit and the reflected light imaging unit, is used to identify the location of defects in the cast film under test based on the transmitted light scanning image and / or the reflected light scanning image; The defect determination module, connected to the defect identification module, is used to: obtain a transmitted light image corresponding to the target defect location based on the transmitted light scanning image, obtain a reflected light image corresponding to the target defect location based on the reflected light scanning image, and determine the defect type of the target defect location based on the transmitted light image and the reflected light image.

10. The apparatus according to claim 9, characterized in that, The acquisition module also includes: A speed measuring unit is used to obtain the conveying speed in the target sample and the conveying speed of the cast film to be tested. The defect determination module is also connected to the speed measurement unit, and is used to obtain the transmitted light image and the reflected light image corresponding to the target defect location from the transmitted light scan image and the reflected light scan image based on the image acquisition delay and the transmission speed of the cast film to be tested.

Images