A plastic product molding control method based on graphic recognition

Abstract

The present invention discloses a plastic product molding control method based on graphic recognition, which relates to the field of plastic product molding control technology, including matching the feature descriptors of the actual molding image with the standard design image, calculating the initial deformation abnormality index; after receiving the initial deformation warning, calculating the total deformation direction coefficient of the plastic product; calculating the difference degree coefficient of the plastic product; adjusting the injection molding temperature according to the total deformation direction coefficient and the difference degree coefficient of the plastic product, if the initial deformation abnormality index of the newly generated plastic product is still abnormal after the adjustment, issuing an automatic adjustment abnormality warning, and outputting the maximum deformation position of the plastic product. It is helpful to directly locate the problem area, facilitate subsequent analysis and improvement, greatly improve the efficiency of problem solving, and reduce unnecessary troubleshooting work.

Description

Technical Field

[0001] The present invention relates to the technical field of plastic product molding control, and in particular to a plastic product molding control method based on graphic recognition. Background Art

[0002] In the field of plastic product processing, molding control has always been a technical problem. With the widespread application of plastic products in various industries and the continuous improvement of product quality requirements, the requirements for molding control technology are becoming increasingly stringent. Traditional molding control methods mostly rely on manual experience and fixed process parameters, which makes it difficult to achieve accurate and efficient control, resulting in unstable product quality, high production costs, and difficulty in adapting to complex and changing market demands. In recent years, with the rapid development of computer vision and artificial intelligence technology, the application of graphic recognition technology in the field of plastic product molding control has gradually attracted attention. Graphic recognition technology can accurately process and analyze the surface images of plastic products, extract key information related to the target molding quality, and provide reliable data support for molding control.

[0003] In a Chinese invention application with application publication number CN115294410A, a plastic product molding control method based on graphic recognition is disclosed, including using a camera to obtain a surface image of a plastic product, performing data processing on the target image to obtain target defect pixel points; further performing data processing on the target defect pixel points to obtain a weld mark defect connectivity domain and a corresponding degree of damage, and adjusting the injection pressure according to the current degree of damage on the surface of the plastic product.

[0004] In the above invention application, the degree of damage of the weld mark to the mechanical properties of the plastic product is calculated based on the graphic characteristics of the weld mark, and then the injection pressure is adjusted to achieve high-quality molding control of the plastic product. However, in the molding process of plastic products, not only weld marks are generated, but also deformation, bubbles, flash and other defects may occur. If only the influence of the weld mark is considered, the overall quality of the plastic product will not be fully guaranteed. Deformation, bubbles, flash and other defects will also have adverse effects on the mechanical properties and appearance quality of the product, thereby affecting the service life and safety of the product. In particular, deformation of plastic products may lead to inaccurate product dimensions, failure to meet assembly requirements and thus unusable.

[0005] To this end, the present invention provides a plastic product molding control method based on graphic recognition. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a plastic product molding control method based on graphic recognition. The present invention matches the actual molding image with the standard design image through feature descriptors and calculates the initial deformation anomaly index Cb, so as to quickly screen out products with problems. When problems occur in the molding of plastic products, the injection molding temperature can be automatically adjusted according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic products. If the newly produced plastic products are still abnormal after the adjustment, an automatic adjustment abnormality warning is issued to the outside, and the maximum deformation position of the plastic products is output, which helps to directly locate the problem area, facilitates subsequent analysis and improvement, greatly improves the efficiency of problem solving, and reduces unnecessary troubleshooting work, thereby solving the technical problems recorded in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A plastic product molding control method based on graphic recognition, comprising the following steps:

[0010] Perform feature descriptor matching between the actual molding image and the standard design image, and obtain the number of matching feature point pairs Pv of the plastic product image to be tested. d , combined with the standard matching total points Pv b , calculate and obtain the initial deformation anomaly index Cb;

[0011] After receiving the initial deformation warning, the feature point data set to be tested is obtained after sorting and standard feature point dataset And calculate and obtain the total deformation direction coefficient Fx of the plastic product;

[0012] Get the feature point dataset to be tested and standard feature point dataset Calculate the difference coefficient Cs of plastic products;

[0013] The injection molding temperature is adjusted according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the adjustment, an automatic adjustment abnormality warning is issued and the maximum deformation position of the plastic product is output.

[0014] Furthermore, an image acquisition device is used to capture an actual molded image of the plastic product, the actual molded image is adjusted to the same size and resolution as the standard design image, a feature point detection algorithm (such as SIFT, SURF, etc.) is used to detect common feature points in the two images, and the actual molded image is aligned with the standard design image based on the detected feature points.

[0015] Furthermore, an image processing library or framework (such as OpenCV, scikit-image, etc.) is used to calculate the attributes of each feature point, including position, direction, and gradient strength, and the feature descriptor in the image of the plastic product to be inspected is matched with the feature descriptor in the standard design image to obtain multiple matching feature point pairs.

[0016] Furthermore, the matching feature point pairs between the normal molded plastic product image and the standard design image are counted and recorded as the total number of standard matching points Pv b Similarly, the number of matching feature point pairs Pv of the plastic product image to be detected is obtained by counting d , after linear normalization, the initial deformation anomaly index Cb is calculated:

[0017]

[0018] When the initial deformation abnormality index Cb is less than 0.9, an initial deformation warning is issued.

[0019] Furthermore, after receiving the initial deformation warning, the attributes of the feature point pairs of the image of the plastic product to be tested are obtained, and the feature point data set to be tested is obtained after sorting. and standard feature point dataset

[0020] Among them, x and y represent the horizontal and vertical coordinates of the feature point in the standard coordinate system, respectively, θ represents the direction angle of the feature point, and ρ represents the gradient strength of the feature point.

[0021] Furthermore, obtain the feature point data set to be tested and standard feature point dataset After linear normalization, the total deformation directional coefficient Fx of the plastic product is calculated:

[0022]

[0023] Wherein, the superscript d represents the data to be tested, the superscript b represents the standard data, and i represents the sequential number of each feature point pair, i=1, 2, ..., m, where m is a positive integer.

[0024] Furthermore, obtain the feature point data set to be tested and standard feature point dataset After linear normalization, the difference degree weights k1 and k2 are calculated:

[0025]

[0026] The calculation formulas for the corresponding difference degree weights k1 and k2 are as above.

[0027] Furthermore, obtain the feature point data set to be tested and standard feature point dataset After linear normalization, the difference coefficient Cs of plastic products is calculated:

[0028]

[0029] The calculation formula of the corresponding plastic product difference coefficient Cs is as above.

[0030] Furthermore, the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product are obtained. When the total deformation direction coefficient Fx is greater than 0, it indicates that the plastic product is tensile deformed. The injection molding temperature is lowered until the difference degree coefficient Cs is minimized. When the total deformation direction coefficient Fx is not greater than 0, it indicates that the plastic product is compressively deformed. The injection molding temperature is increased until the difference degree coefficient Cs is minimized.

[0031] It should be noted that when the injection molding temperature is adjusted, the difference coefficient Cs of the newly produced plastic products will change accordingly. During the adjustment process, the difference coefficient Cs will have a minimum point, which is a dynamic threshold. The adjustment results are not consistent each time.

[0032] Furthermore, if the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the difference degree coefficient Cs is adjusted to the minimum, an automatic adjustment abnormality warning is issued to the outside, and the maximum deformation position of the plastic product is output, indicating that the production process and mold related to this position need to be checked.

[0033] Among them, the maximum deformation position of the plastic product is The position corresponding to the feature point number.

[0034] (III) Beneficial effects

[0035] The present invention provides a plastic product molding control method based on graphic recognition, which has the following beneficial effects:

[0036] 1. By matching the feature descriptors of the actual molding image with the standard design image, the number of matching feature point pairs Pv of the plastic product image to be tested is obtained statistically d , combined with the standard matching total points Pv b , the calculation of the initial deformation anomaly index Cb can greatly improve the detection efficiency, especially on large-scale production lines, and can quickly screen out products with problems and reduce the need for manual intervention.

[0037] 2. After receiving the initial deformation warning, the feature point data set to be tested is obtained after sorting and standard feature point dataset The total deformation directional coefficient Fx of the plastic product is calculated, which can clearly reveal the overall trend and direction of the deformation of the plastic product. This helps to analyze the cause of the deformation and put forward targeted improvement measures.

[0038] 3. Obtain the feature point data set to be tested and standard feature point dataset Calculating the difference coefficient Cs of plastic products provides an accurate quantitative indicator for measuring the difference between the plastic product to be tested and the standard design. This quantitative method is more objective and accurate than traditional visual inspection and can more accurately reflect the actual deformation or deviation of plastic products.

[0039] 4. Adjust the injection molding temperature according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the adjustment, issue an automatic adjustment abnormality warning and output the maximum deformation position of the plastic product. Automatic adjustment and warning can significantly improve the quality control capability of plastic product production. At the same time, outputting the maximum deformation position of the plastic product helps to directly locate the problem area, facilitate subsequent analysis and improvement, greatly improve the efficiency of problem solving, and reduce unnecessary troubleshooting work. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The figure is a flow chart of a plastic product molding control method based on graphic recognition according to the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 The present invention provides a plastic product molding control method based on graphic recognition, comprising the following steps:

[0043] Step 1: Match the feature descriptors of the actual molding image with the standard design image, and obtain the number of matching feature point pairs Pv of the plastic product image to be tested. d , combined with the standard matching total points Pv b , calculate and obtain the initial deformation anomaly index Cb.

[0044] The step 1 includes the following contents:

[0045] Step 101: Use an image acquisition device to acquire an actual molding image of the plastic product, adjust the actual molding image to the same size and resolution as the standard design image, use a feature point detection algorithm (such as SIFT, SURF, etc.) to detect common feature points in the two images, and align the actual molding image with the standard design image based on the detected feature points.

[0046] Step 102: Use an image processing library or framework (such as OpenCV, scikit-image, etc.) to calculate the attributes of each feature point, including position, direction, and gradient strength, and match the feature descriptor in the image of the plastic product to be inspected with the feature descriptor in the standard design image to obtain multiple matching feature point pairs.

[0047] Step 103: Count the matching feature point pairs between the normal molded plastic product image and the standard design image, and record them as the total number of standard matching points Pv b Similarly, the number of matching feature point pairs Pv of the plastic product image to be detected is obtained by counting d , after linear normalization, the initial deformation anomaly index Cb is calculated:

[0048]

[0049] When the initial deformation abnormality index Cb is less than 0.9, an initial deformation warning is issued.

[0050] When using, combine the contents in steps 101 to 103:

[0051] By matching the feature descriptors of the actual molding image with the standard design image, the number of matching feature point pairs Pv of the plastic product image to be inspected is obtained statistically. d , combined with the standard matching total points Pv b , the calculation of the initial deformation anomaly index Cb can greatly improve the detection efficiency, especially on large-scale production lines, and can quickly screen out products with problems and reduce the need for manual intervention.

[0052] Step 2: After receiving the initial deformation warning, the feature point data set to be tested is obtained after sorting and standard feature point dataset And calculate and obtain the total deformation directional coefficient Fx of the plastic product.

[0053] The step 2 includes the following contents:

[0054] Step 201: After receiving the initial deformation warning, obtain the attributes of the feature point pairs of the image of the plastic product to be tested, and obtain the feature point data set to be tested after sorting. and standard feature point dataset

[0055] Among them, x and y represent the horizontal and vertical coordinates of the feature point in the standard coordinate system, respectively, θ represents the direction angle of the feature point, and ρ represents the gradient strength of the feature point.

[0056] Gradient strength is the rate of change of a physical quantity in space or time. The magnitude of the gradient strength indicates the rate of change of the physical quantity.

[0057] Step 202: Obtain the feature point data set to be measured and standard feature point dataset After linear normalization, the total deformation directional coefficient Fx of the plastic product is calculated:

[0058]

[0059] Wherein, the superscript d represents the data to be tested, the superscript b represents the standard data, and i represents the sequential number of each feature point pair, i=1, 2, ..., m, where m is a positive integer.

[0060] When using, combine the contents in steps 201 and 202:

[0061] After receiving the initial deformation warning, the feature point data set to be tested is obtained after sorting and standard feature point dataset The total deformation directional coefficient Fx of the plastic product is calculated, which can clearly reveal the overall trend and direction of the deformation of the plastic product. This helps to analyze the cause of the deformation and put forward targeted improvement measures.

[0062] Step 3: Get the feature point dataset to be tested and standard feature point dataset Calculate and obtain the coefficient of difference Cs of plastic products.

[0063] The step three includes the following contents:

[0064] Step 301: Obtain the feature point data set to be measured and standard feature point dataset After linear normalization, the difference degree weights k1 and k2 are calculated:

[0065]

[0066] The calculation formulas for the corresponding difference degree weights k1 and k2 are as above.

[0067] Step 302: Obtain the feature point data set to be measured and standard feature point dataset After linear normalization, the difference coefficient Cs of plastic products is calculated:

[0068]

[0069] The calculation formula of the corresponding plastic product difference coefficient Cs is as above.

[0070] When used, combine the contents in steps 301 and 302:

[0071] By obtaining the feature point data set to be tested and standard feature point dataset Calculating the difference coefficient Cs of plastic products provides an accurate quantitative indicator for measuring the difference between the plastic product to be tested and the standard design. This quantitative method is more objective and accurate than traditional visual inspection and can more accurately reflect the actual deformation or deviation of plastic products.

[0072] Step 4: Adjust the injection temperature according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the adjustment, issue an automatic adjustment abnormality warning and output the maximum deformation position of the plastic product.

[0073] The step 4 includes the following contents:

[0074] Step 401, obtain the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. When the total deformation direction coefficient Fx is greater than 0, it indicates that the plastic product is tensile deformed. Lower the injection molding temperature until the difference degree coefficient Cs is minimized. When the total deformation direction coefficient Fx is not greater than 0, it indicates that the plastic product is compressively deformed. Increase the injection molding temperature until the difference degree coefficient Cs is minimized.

[0075] It should be noted that when the injection molding temperature is adjusted, the difference coefficient Cs of the newly produced plastic products will change accordingly. During the adjustment process, the difference coefficient Cs will have a minimum point, which is a dynamic threshold. The adjustment results are not consistent each time.

[0076] Step 402: If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the difference degree coefficient Cs is adjusted to the minimum, an automatic adjustment abnormality warning is issued to the outside, and the maximum deformation position of the plastic product is output, indicating that the production process and mold related to this position need to be checked.

[0077] Among them, the maximum deformation position of the plastic product is The position corresponding to the feature point number.

[0078] When using, combine the contents in steps 401 and 402:

[0079] The injection molding temperature is adjusted according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the adjustment, an automatic adjustment abnormality warning will be issued and the maximum deformation position of the plastic product will be output. Automatic adjustment and warning can significantly improve the quality control capability of plastic product production. At the same time, outputting the maximum deformation position of the plastic product will help to directly locate the problem area, facilitate subsequent analysis and improvement, greatly improve the efficiency of problem solving, and reduce unnecessary troubleshooting work.

[0080] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0081] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A plastic product molding control method based on pattern recognition, characterized in that: The steps include: Perform feature descriptor matching between the actual molding image and the standard design image, and obtain the number of matching feature point pairs Pv of the plastic product image to be tested. d , combined with the standard matching total points Pv b , calculate and obtain the initial deformation anomaly index Cb; After receiving the initial deformation warning, the feature point data set to be tested is obtained after sorting and standard feature point dataset And calculate and obtain the total deformation direction coefficient Fx of the plastic product; Get the feature point dataset to be tested and standard feature point dataset Calculate the difference coefficient Cs of plastic products; The injection molding temperature is adjusted according to the total deformation direction coefficient Fx and the difference degree coefficient Cs of the plastic product. If the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal after the adjustment, an automatic adjustment abnormality warning is issued and the maximum deformation position of the plastic product is output; Count the matching feature point pairs between the normal molded plastic product image and the standard design image, and record them as the total number of standard matching points Pv b Similarly, the number of matching feature point pairs Pv of the plastic product image to be detected is obtained by counting d , after linear normalization, the initial deformation anomaly index Cb is calculated: When the initial deformation abnormality index Cb is less than 0.9, an initial deformation warning is issued; After receiving the initial deformation warning, obtain the attributes of the feature point pairs of the image of the plastic product to be tested, and obtain the feature point data set to be tested after sorting and standard feature point dataset Among them, x and y represent the horizontal and vertical coordinates of the feature point in the standard coordinate system, θ represents the direction angle of the feature point, and ρ represents the gradient strength of the feature point; Get the feature point dataset to be tested and standard feature point dataset After linear normalization, the total deformation directional coefficient Fx of the plastic product is calculated: Wherein, the superscript d represents the data to be tested, the superscript b represents the standard data, and i represents the sequence number of each feature point pair, i=1, 2, ..., m, and m is a positive integer; Get the feature point dataset to be tested and standard feature point dataset After linear normalization, the difference degree weights k1 and k2 are calculated: The calculation formulas for the corresponding difference degree weights k1 and k2 are as above; Get the feature point dataset to be tested and standard feature point dataset After linear normalization, the difference coefficient Cs of plastic products is calculated: The calculation formula of the corresponding plastic product difference coefficient Cs is as above; Obtain the total deformation direction coefficient Fx and difference degree coefficient Cs of plastic products; When the total deformation direction coefficient Fx is greater than 0, it means that the plastic product is tensile deformed. The injection molding temperature should be lowered until the difference degree coefficient Cs is minimized. When the total deformation direction coefficient Fx is not greater than 0, it means that the plastic product is compressively deformed. The injection molding temperature should be increased until the difference degree coefficient Cs is minimized.

2. A plastic product molding control method based on graphic recognition according to claim 1, characterized in that: If the difference coefficient Cs is adjusted to the minimum, the initial deformation abnormality index Cb of the newly produced plastic product is still abnormal, and an automatic adjustment abnormality warning is issued to the outside, and the maximum deformation position of the plastic product is output. Among them, the maximum deformation position of the plastic product is The position corresponding to the feature point number.

Images