Method for predicting cable extrusion length based on image recognition filter screen clogging characterization

Abstract

The application provides a method for predicting cable extrusion length based on image recognition filter screen blockage characterization, which comprises collecting an original image of a cable extrusion filter screen after stripping, pre-processing and feature extraction to obtain a feature image; calculating the blockage degree of the cable extrusion filter screen based on the feature image; constructing a corresponding cable material prediction extrusion length model based on the blockage degree of the cable extrusion filter screen, and calculating to obtain the prediction extrusion length. Based on the image recognition technology, the characterization result of the cable extrusion filter screen blockage degree can be quickly obtained in the test scene, so that the fast and accurate evaluation of the filter screen blockage degree can be realized, and the filter screen blockage degree is closely related to the insulation material characteristic parameters, extruder process parameters and other factors in the cable extrusion process, and the precise prediction of the cable extrusion length under different material conditions of the same type of filter screen is provided, which has good guidance for production, reduces the production cost and improves the production efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a method for predicting cable extrusion length based on image recognition of filter blockage representation. Background Art

[0002] Cables typically consist of a conductor, insulation layer, shielding layer, and sheath. During cable manufacturing, these layers are typically produced using an extruder. This process melts and plasticizes pre-treated cable insulation material, then extrudes it through a specialized die to form a continuous insulation layer that wraps around the conductor, providing electrical insulation and preventing current leakage, ensuring safe and stable power transmission.

[0003] Extruders are widely used in cable material production, primarily due to their ability to achieve continuous and efficient production. The rotating screw applies pressure to the material, causing it to undergo plasticization and melting within the barrel before being extruded into a finished product. This process ensures uniform mixing and plasticization of the material, ensuring the consistency of the cable material's physical and chemical properties, thus meeting the stringent requirements of cable production. The use of filters within extruders is a crucial measure to ensure cable quality. Raw materials and the production environment inevitably contain impurities such as metal particles, dust, and partially melted plastic particles. If these impurities enter the cable material, they can cause defects within the cable. Within the insulation layer, impurities can degrade insulation performance, triggering partial discharge and increasing the risk of cable breakdown. Around the conductor, impurities can disrupt current distribution, leading to localized overheating and reducing the cable's current-carrying capacity.

[0004] In actual work, not only the cross-sectional quality of the cable layer must be considered, but also the production length of the cable layer that can be processed by continuous preparation of cable materials under fixed processing conditions must be estimated. Currently, the production length prediction has always been estimated in the form of empirical summary, but this prediction method has the following problems:

[0005] 1. Inaccurate length prediction makes it difficult to determine the material required. Insufficient material input will result in substandard cable layer length in a single process, thus resulting in defective products. Excessive material input will result in excessively long cable layer length in a single process, which not only increases the number of subsequent processing steps but also leads to waste of cable materials.

[0006] 2. Poor versatility means that after changing the processing parameters, it is necessary to accumulate experience again in order to predict the processing length, which leads to increased costs and reduced quality of finished products.

[0007] The above information demonstrates that the proper functioning of the cable extrusion filter in the extruder is crucial for cable extrusion production. When the filter becomes severely clogged, the resistance to material passing through the filter increases dramatically. This forces the extruder screw to exert greater pressure to propel the material forward, increasing the motor load. Furthermore, the extrusion speed of the material can vary significantly, impacting not only production efficiency but also potentially leading to fluctuations in cable quality due to process instability, increasing scrap rates, and causing financial losses and quality risks for manufacturers.

[0008] In summary, since the state of the filter has a significant impact on the adjustment of cable extrusion process parameters, cable extrusion efficiency and cable surface quality, it is possible to consider establishing a correlation model between the degree of filter blockage and related factors such as the characteristic parameters of cable extrusion insulation materials and the process parameters of the cable extruder. On the one hand, it helps to optimize the production process and improve product quality, and it can also make scientific and accurate predictions on the cable extrusion length. Summary of the Invention

[0009] In view of the problems existing in the prior art, the present invention provides a method for predicting cable extrusion length by characterizing filter blockage, which is simple, low-cost and universal.

[0010] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0011] A method for predicting cable extrusion length based on image recognition of filter blockage characterization mainly includes the following steps:

[0012] Establishing a correspondence between the cable extrusion filter and the cable material, and peeling off the cable extrusion filter after the cable is extruded to a preset length;

[0013] Collecting an original image of the cable extrusion filter after stripping;

[0014] Preprocessing the original image to obtain a processed image;

[0015] Extracting features from the processed image using a feature extraction algorithm to obtain a feature image;

[0016] Calculating the blockage degree of the cable extrusion filter based on the characteristic image;

[0017] Based on the blockage degree of the cable extrusion filter, a corresponding cable material extrusion length prediction model is constructed. The formula of the predicted extrusion length model is:

[0018] Where, is the blockage threshold of the cable extrusion filter, is the predicted extrusion length of the corresponding cable material, The degree of clogging of the cable extrusion filter when the cable is extruded to a preset length. Extruding a preset length of said cable, The insulation thickness of the cable when it is extruded to a preset length, The conductor diameter of the cable when it is extruded to a preset length, is the insulation thickness of the cable extruded from the corresponding cable material, The conductor diameter of the cable is extruded from the corresponding cable material.

[0019] Optionally, the original image acquisition operation includes:

[0020] shaping the cable extrusion filter, and laying the cable extrusion filter flat on a solid color background;

[0021] Continuously photographing the working surface of the cable extrusion filter screen vertically with a camera to obtain a photographic atlas;

[0022] The photographic atlas is screened based on imaging effects, and at least one photo is selected as the original image.

[0023] Optionally, the original image acquisition operation further includes:

[0024] A shooting light source is placed on the side of the cable extrusion filter, and the shooting light source is equipped with a soft light tool;

[0025] The camera is equipped with a polarizing filter, and during continuous shooting, the angle of the polarizing filter is periodically adjusted.

[0026] Optionally, screening the captured atlas based on imaging effects includes:

[0027] The amount of light reflected from each photo in the photo collection is calculated based on a pixel brightness value analysis method, and the photo with the smallest amount of light reflected is taken as the original image.

[0028] Optionally, the preset extrusion length of the cable is 20%-35% of the theoretical extrusion length of the cable material.

[0029] Optionally, the step of preprocessing the original image includes at least:

[0030] Denoising, contrast enhancement and color correction are performed on the original image.

[0031] Optionally, the feature extraction algorithm includes edge detection and texture analysis;

[0032] The characteristic information of the characteristic image includes porosity, texture and color distribution.

[0033] Optionally, calculating the blockage degree of the cable extrusion filter based on the characteristic image includes:

[0034] Gray-scale the feature image to obtain a grayscale image;

[0035] Extracting local texture information from the grayscale image using a local binary pattern algorithm, and performing binary segmentation to obtain a binary image;

[0036] Performing median filtering on the binary image to obtain an optimized image;

[0037] The contour of the optimized image is extracted by a center diffusion method, and a contour map is obtained.

[0038] Optionally, calculating the blockage degree of the cable extrusion filter based on the characteristic image further includes:

[0039] The proportion of blocked grids in the cable extrusion filter is calculated based on the contour map and used as the blockage degree of the cable extrusion filter.

[0040] Optionally, the blockage threshold of the cable extrusion filter is 25%.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention is based on image recognition technology and can quickly obtain the characterization results of the cable extrusion filter blockage degree in the test scenario, thereby realizing a rapid and accurate evaluation of the filter blockage degree, and closely linking the filter blockage degree with factors such as the insulation material characteristic parameters and extruder process parameters in the cable extrusion process, and providing accurate prediction of the cable extrusion length under different material dosage conditions of the same type of filter, which has good guidance for production, reduces production costs and improves production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A flow chart of a method in a specific embodiment of the present invention;

[0045] Figure 2 is the original image in Example 1 of the present invention;

[0046] Figure 3 The enhanced image in Example 1 of the present invention;

[0047] Figure 4 is the binarized image in Example 1 of the present invention;

[0048] Figure 5 This is the outline diagram in Example 1 of the present invention. DETAILED DESCRIPTION

[0049] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0050] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0051] In the description of the present invention, “plurality” means two or more, unless otherwise clearly defined.

[0052] In the present invention, unless otherwise expressly specified or limited, terms such as "mounted" and "connected" should be interpreted broadly. For example, they may refer to fixed connection, detachable connection, or integration; they may refer to direct connection or indirect connection through an intermediate medium; they may refer to internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0053] It is worth noting that the methods used in the present invention are all conventional methods unless otherwise specified; the raw materials and devices used are all conventional commercially available products, and their sources are not specifically limited unless otherwise specified.

[0054] like Figure 1 As shown, this embodiment provides a method for predicting cable extrusion length based on image recognition of filter blockage characterization, which mainly includes the following steps:

[0055] First, it's important to confirm that in order to accurately predict the length of a cable layer that can be processed using a given material type, given the material quantity and the equipment's operating conditions, appropriate experiments are necessary. Therefore, we define short-term extrusion and long-term extrusion. Short-term extrusion occurs when the machine stops after a preset length of cable is extruded, while long-term extrusion occurs when a certain amount of cable material is completely processed and extruded into a cable, and then the machine stops.

[0056] Among them, filter clogging is a known working condition that affects the continuous extrusion of cable materials. During the actual extrusion process, when the filter clogging reaches the clogging threshold, the melt pressure will increase sharply, causing subsequent cable production to stop. Therefore, the goal of this example is to study the relationship between cable extrusion filter clogging and extrusion length.

[0057] The cable extrusion preset length is designed to allow the filter to display a certain working state (clearly identify blockages) while minimizing the use of raw materials. Therefore, the cable extrusion preset length is designed to be 20%-35% of the theoretical extrusion length of the cable material. The theoretical extrusion length of the cable material is calculated by dividing the cable material used in the task by the cross-sectional area of ​​the cable layer after processing.

[0058] Therefore, the corresponding relationship between the cable extrusion filter and the cable material is first established, that is, the model of the cable extrusion filter used and the model of the cable material to be used (components, ratios or trade names, etc.) are determined according to the processing task. In each group of experiments, ensure that the model of the cable extrusion filter and the model of the cable material are consistent with the processing task information.

[0059] In each group of experiments, the cable extrusion filter is peeled off after the cable is extruded to a preset length, that is, the machine is stopped after a short extrusion, and the operator uses tooling and tools to remove the cable extrusion filter used in this group without damage, such as Figure 2 shown.

[0060] Collecting an original image of the stripped cable extrusion filter; optionally, the operation of collecting the original image includes the following steps:

[0061] (1) Shape the cable extrusion filter and lay it flat on a solid color background. Optionally, since the cable extrusion filter is usually made of metal and the force direction during operation is fixed, it may cause edge warping, which results in poor imaging effect, especially when there are few blockages, making it difficult to obtain accurate results. Therefore, it is necessary to reshape the cable extrusion filter to a certain extent to restore its lateral straight state. Currently, there are three common shaping methods, specifically:

[0062] The splint shaping method uses a splint to clamp the cable and extrude it through the filter, and performs high and low temperature cycles or high temperature-natural cooling to reshape it. Its advantages are simple operation and controllable costs. However, since the splint will contact the working surface of the filter, it will cause problems such as blockage falling off and deformation, resulting in inaccurate image recognition results, especially in the case of mild blockage.

[0063] The fixture extension method uses a ring fixture to fix the outer ring of the cable extrusion filter and applies a certain tension to reshape it. Its advantage is that it avoids contact with the working area and avoids affecting the shape and total amount of the blockage. However, due to the complexity of the process, it leads to high operating requirements and complex tooling structure. In addition, unreasonable application of tension will cause the filter to deform, affecting the calculation accuracy. Moreover, due to the existence of the tooling, additional tooling image removal work is required in the subsequent image recognition process, which not only increases the workload, but also causes the blockage degree to increase abnormally due to pixel residue, or excessive removal reduces the blockage area, resulting in an abnormal decrease in the blockage degree.

[0064] The direct scanning method is similar to the plywood shaping method, that is, the filter is directly placed in the scanning device for cover scanning. It not only has accuracy problems, but also due to the light control problem of the scanning device, there are a lot of reflections, which distorts the image.

[0065] Therefore, to enhance the image quality, adhesive bonding can be used. If the warping is not severe, use a suitable adhesive to secure the warped area to a flat substrate. Choose an adhesive that adheres well to stainless steel, such as some epoxy resins. Before bonding, ensure that the cable extrusion filter and the substrate (solid background) are clean and dry. Apply the adhesive evenly to the contact area between the warped area and the substrate. Press the edge of the cable extrusion filter against the substrate to maintain a flat surface. After the adhesive has fully cured, remove the filter from the substrate. This method is relatively simple to use, but may cause some contamination to the cable extrusion filter surface, and the adhesive must be carefully selected and used to ensure that it meets the required performance. Alternatively, if the cable extrusion filter is made of martensitic or ferritic stainless steel, prepare a non-metallic platform and install an electromagnet behind the imaging area. Place the cable extrusion filter on the imaging area and level it. Then, energize the electromagnet to attach the cable extrusion filter to the area.

[0066] Regardless of which of the above fixing methods is adopted, in order to obtain better shooting effects, a solid color background can be a white background, such as a flat white background board, a white table top, a white tablecloth, etc.

[0067] (2) Continuously shoot vertically the working surface of the cable extrusion filter with a camera and obtain a set of photos; optionally, this step also includes:

[0068] The camera is fixed in position, directly above the cable extrusion filter. Therefore, there are no adjustment options other than shooting distance. Furthermore, the cable extrusion filter is generally made of metal, such as stainless steel. Therefore, its reflection needs to be considered, as it will significantly affect the judgment of blockages and thus directly affect the blockage degree calculation results.

[0069] Therefore, the design placed the shooting light source to the side of the cable extrusion filter, and installed a softening tool such as a diffuser or diffuser to prevent strong light from directly hitting the metal mesh and causing strong reflections. At the same time, the position and direction of the light source can be changed to illuminate the metal mesh from the side or above at a certain angle, rather than vertically, to reduce specular reflections. This ensures that the light is evenly distributed on the metal mesh, reducing localized strong reflections and improving the overall shooting effect.

[0070] Furthermore, a polarizing filter is installed on the camera. This filter can effectively reduce reflections from non-metallic surfaces and is also effective for metal meshes. By installing a polarizing filter in front of the lens and rotating it to adjust its angle, it can selectively block polarized light in a specific direction, thereby reducing the intensity of reflections from the metal mesh and enhancing image clarity and detail. This principle utilizes the optical properties of the polarizing filter to allow only light in a specific direction to pass through, filtering out most of the polarized light generated by reflections, making the texture and structure of the metal mesh easier to capture. Furthermore, it is necessary to periodically adjust the angle of the polarizing filter during continuous shooting to produce a collection of shots with different imaging effects.

[0071] (3) Screening the photo collection based on imaging effects and selecting at least one photo as the original image; optionally, screening the photo collection based on imaging effects includes:

[0072] Based on pixel brightness analysis, the amount of reflectivity in each photo in the collection is calculated, and the photo with the lowest reflectivity is selected as the original image. Specifically, in a digital image, each pixel has a corresponding brightness value. The amount of reflectivity can be quantified by extracting the brightness values ​​of pixels in the reflective area. Using image processing software (such as Adobe Photoshop or the Python OpenCV library), select pixels in the reflective area and calculate their average brightness or distribution. For example, in OpenCV, you can read an image file, convert it to a grayscale image, and then use a function to obtain the brightness information of pixels in the reflective area. Higher brightness values ​​generally indicate stronger reflectivity. By recording and analyzing these values, the amount of reflectivity can be measured to a certain extent. This method is relatively simple and direct, and can quickly obtain the basic brightness characteristics of the reflective area in the image. Then, sort the photos in the collection from lowest to highest in terms of reflectivity, and select the photo with the lowest reflectivity as the original image.

[0073] Preprocess the original image and obtain the processed image; optionally, to effectively retain edge information, the collected original image is first subjected to Gaussian denoising, then bilateral filtering denoising, and the image signal-to-noise ratio is calculated. If the signal-to-noise ratio is greater than the standard value, the subsequent steps are performed; if it is lower than the standard value, the denoising operation is continued; then the Sobel operator is applied to calculate the gradient amplitude of the image to further identify the edge, and finally, to avoid the influence of noise, the contrast-limited adaptive histogram equalization method is applied to the local area of ​​the image for equalization preprocessing to improve the accuracy of subsequent analysis. Figure 3 As shown, it is the processed image after HSV color enhancement method.

[0074] The processed image is subjected to feature extraction by a feature extraction algorithm to obtain a feature image; optionally, this step is to use edge detection, texture analysis and other algorithms to extract features in the processed image, such as porosity, texture changes and color distribution, in order to perform subsequent image recognition work.

[0075] Calculating the blockage degree of the cable extrusion filter based on the characteristic image; optionally, calculating the blockage degree of the cable extrusion filter based on the characteristic image includes:

[0076] Grayscale the feature image to obtain a grayscale image;

[0077] The local binary pattern (LBP) algorithm is used to extract local texture information from the grayscale image and perform binary segmentation to obtain a binary image, such as Figure 4 As shown;

[0078] Perform Canny edge detection filtering on the binary image and obtain an optimized image. Considering the difference in image quality, based on the porosity and local texture information obtained above, extract the pixel counts of the five grids with obvious sharp edges on the outer edge of the filter, and finally calculate the grid area of ​​the filter.

[0079] Considering the diffusion of filter clogging, the contour of the optimized image is extracted by the center diffusion method, and the contour map is obtained, such as Figure 5 shown.

[0080] Furthermore, the proportion of pixels in the contour area is calculated based on the contour map programming, and the proportion of blocked grids is determined based on the comparison and matching with the unblocked filter image, and is output as the blockage degree of the cable extrusion filter.

[0081] Based on the blockage degree of the cable extrusion filter, a corresponding cable material prediction extrusion length model is constructed. In the process of cable extrusion, the unit volume flow rate of the cable material remains constant, and the proportion of impurities in the unit volume of the cable material is constant. As the impurity particles are enriched in the cable extrusion filter, the blockage degree of the filter increases; according to the Bernoulli equation, the flow rate and flow velocity show a power nonlinear relationship. Since the unit volume flow rate of the cable material remains constant, the flow velocity will increase nonlinearly. Therefore, the volume of cable material passing through the unblocked filter increases, resulting in a nonlinear increase in the blockage degree. Moreover, the longer the extruded cable length, the faster the blockage rate increases, indicating that the nonlinear power coefficient is related to the extruded cable length. Assuming the preset length The congestion degree is (0< <1), therefore, according to the above theory, the formula for predicting the extrusion length model is:

[0082] Where, is the clogging threshold of the cable extrusion filter, is the predicted extrusion length of the corresponding cable material, The degree of clogging of the cable extrusion filter when the cable is extruded to a preset length. Extrude a preset length for the cable, The insulation thickness of the cable when it is extruded to a preset length, The conductor diameter of the cable when it is extruded to a preset length, is the insulation thickness of the cable extruded from the corresponding cable material, The conductor diameter of the cable extruded from the corresponding cable material.

[0083] in, and The extruder processing parameters corresponding to the long-term extrusion are and Complete consistency is not required.

[0084] Therefore, in this embodiment, the blockage threshold of the cable extrusion filter is set to 25% according to the processing conditions. By inputting the blockage of the cable extrusion filter obtained by the aforementioned image recognition, the length of the cable to be extruded for a long time can be predicted, that is, the expected extrusion length when the upper limit of continuous processing allowed by the extruder is reached.

[0085] From this, the timing for shutdown can be clearly determined. That is, when the length of the continuously extruded cable layer reaches the above-mentioned predicted length, it means that the blockage degree is about 25%. This can be used as a guide for shutdown work. After that, the cable extrusion filter can be replaced or cleaned, thereby reducing extruder failure or product quality failure.

[0086] Example 1;

[0087] S1: Image acquisition: Use a high-resolution camera to capture the cable extrusion filter that has been stripped after short-term extrusion of 500 m (the preset cable extrusion length) of cable to obtain the original image of the filter surface.

[0088] S2: Image preprocessing: performing preprocessing operations such as denoising, contrast enhancement, and color correction on the collected original images to improve the accuracy of subsequent analysis.

[0089] S3: Feature extraction: using edge detection, texture analysis and other algorithms to extract features from the processed image, such as porosity, texture changes and color distribution.

[0090] S4: Blockage assessment: Based on the extracted features, the image is grayscaled and the local texture information of the filter surface is extracted using an algorithm based on local binary patterns (LBP). Binary segmentation is achieved by setting an appropriate grayscale threshold. The image is further optimized using a median filter method. Considering the diffusivity of filter blockage, the contour is extracted using a center diffusion method.

[0091] S5: Result output and decision-making; program to calculate the pixel ratio in the contour area, and determine the ratio of blocked grids based on the comparison and matching with the unblocked filter image. The calculated blockage degree is 1.2%.

[0092] S6: Theoretical prediction: According to the above formula, when the filter blockage area is 25%, the predicted extrusion length is 20095.33m.

[0093] in, and The extruder processing parameters corresponding to the long-term extrusion are and The parameters are summarized in Table 1:

[0094] Table 1

[0095]

[0096] Further, even if and The extruder processing parameters corresponding to the long-term extrusion are and In another embodiment, the same method can obtain the following parameters, which are summarized in Table 2:

[0097] Table 2

[0098]

[0099] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A method for predicting cable extrusion length based on image recognition of filter blockage characterization, characterized by: The steps include: Establishing a correspondence between the cable extrusion filter and the cable material, and peeling off the cable extrusion filter after the cable is extruded to a preset length; Collecting an original image of the cable extrusion filter after stripping; Preprocessing the original image to obtain a processed image; Extracting features from the processed image using a feature extraction algorithm to obtain a feature image; Calculating the blockage degree of the cable extrusion filter based on the characteristic image; Based on the blockage degree of the cable extrusion filter, a corresponding cable material extrusion length prediction model is constructed. The formula of the predicted extrusion length model is: Where, is the blockage threshold of the cable extrusion filter, is the predicted extrusion length of the corresponding cable material, The degree of clogging of the cable extrusion filter when the cable is extruded to a preset length. Extruding a preset length of said cable, The insulation thickness of the cable when it is extruded to a preset length, The conductor diameter of the cable when it is extruded to a preset length, is the insulation thickness of the cable extruded from the corresponding cable material, The conductor diameter of the cable extruded from the corresponding cable material; The preset extrusion length of the cable is 20%-35% of the theoretical extrusion length of the cable material.

2. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 1 is characterized by: The acquisition operation of the original image includes: shaping the cable extrusion filter, and laying the cable extrusion filter flat on a solid color background; Continuously photographing the working surface of the cable extrusion filter screen vertically with a camera to obtain a photographic atlas; The photographic atlas is screened based on imaging effects, and at least one photo is selected as the original image.

3. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 2, characterized in that: The acquisition operation of the original image further includes: A shooting light source is placed on the side of the cable extrusion filter, and the shooting light source is equipped with a soft light tool; The camera is equipped with a polarizing filter, and during continuous shooting, the angle of the polarizing filter is periodically adjusted.

4. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 2 or 3, characterized in that: Screening the photographic atlas based on imaging effects includes: The amount of light reflected from each photo in the photo collection is calculated based on a pixel brightness value analysis method, and the photo with the smallest amount of light reflected is taken as the original image.

5. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 1 is characterized in that: The step of preprocessing the original image at least includes: Denoising, contrast enhancement and color correction are performed on the original image.

6. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 1, characterized in that: The feature extraction algorithm includes edge detection and texture analysis; The characteristic information of the characteristic image includes porosity, texture and color distribution.

7. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 1, characterized in that: Calculating the blockage degree of the cable extrusion filter based on the characteristic image includes: Gray-scale the feature image to obtain a grayscale image; Extracting local texture information from the grayscale image using a local binary pattern algorithm, and performing binary segmentation to obtain a binary image; Performing median filtering on the binary image to obtain an optimized image; The contour of the optimized image is extracted by a center diffusion method, and a contour map is obtained.

8. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 7, characterized in that: Calculating the blockage degree of the cable extrusion filter based on the characteristic image further includes: The proportion of blocked grids in the cable extrusion filter is calculated based on the contour map and used as the blockage degree of the cable extrusion filter.

9. The method for predicting cable extrusion length based on image recognition of filter blockage characterization according to claim 1, characterized in that: The cable extrusion filter has a blockage threshold of 25%.

Images