Cooling uniformity detection method for injection molded parts based on image processing

Abstract

The present invention relates to the field of injection molded parts detection, and more specifically, the present invention relates to an injection molded parts cooling uniformity detection method based on image processing. The method comprises: obtaining a surface grayscale image of the injection molded parts during cooling based on a preset sampling interval; constructing a grayscale exclusion matrix at a target moment, and calculating the static stability of the injection molded parts; subtracting the surface grayscale image at the target moment from the surface grayscale image at a previous moment adjacent to the target moment to obtain a differential image at the target moment, and calculating the dynamic stability of the injection molded parts based on the differential image; calculating the comprehensive stability of the injection molded parts at the target moment based on the static stability and dynamic stability, and completing the uniformity detection. Through the technical solution of the present invention, the accuracy of the uniformity detection of the injection molded parts during cooling can be improved, and the quality of the injection molded parts after cooling can be improved.

Description

Technical Field

[0001] The present invention relates to the field of injection molded parts detection, and more specifically, to an injection molded parts cooling uniformity detection method based on image processing. Background Art

[0002] Injection molding is a commonly used process in modern manufacturing and is widely used in the production of plastic products, including automotive parts, electronic products, and daily necessities. The quality of injection molded parts directly affects the performance and appearance of the product. During the injection molding process, cooling is an important link to ensure the dimensional accuracy, shape stability, and physical properties of the finished product. Uneven cooling can cause warping, deformation, and internal stress concentration of injection molded parts, thereby affecting the quality of the product. At present, the detection of cooling uniformity mainly relies on manual inspection or simple physical measurement methods, which are usually not accurate enough to meet the high requirements of modern manufacturing for precision and efficiency. With the development of image processing technology, image-based detection methods have gradually become a hot topic of research. These methods automatically identify the unevenness in the cooling process by analyzing the image of the surface of the injection molded part, and have the advantages of high efficiency, non-destructiveness, and easy operation.

[0003] The existing Chinese patent application document with publication number CN117688458A discloses a method and system for automatically optimizing the process parameters of injection molds, in which a process parameter combination calculation formula is used to perform parameter combination screening on feature sorting data to obtain a parameter combination data set. The process parameter combination calculation formula includes the injection temperature and mold temperature of each dimension.

[0004] However, during the cooling process of the injection molded parts, the temperature change of the injection molded parts has a great impact on the quality of the final injection molded parts, which is particularly important in the production of plastic products. The cooling stage directly affects the dimensional stability and surface quality of the plastic parts. The above application documents only consider the overall injection temperature and mold temperature in each dimension, which will lead to abnormalities after the injection molded parts are cooled, resulting in low quality of the final injection molded parts. Summary of the invention

[0005] In order to solve the problem of abnormality after cooling of injection molded parts, the present invention proposes a method for detecting cooling uniformity of injection molded parts based on image processing.

[0006] The present invention discloses an injection molded part cooling uniformity detection method based on image processing, comprising: obtaining a surface grayscale image of the injection molded part during cooling based on a preset sampling interval, and taking any sampling moment as a target moment; constructing a grayscale exclusion matrix at the target moment, and calculating the static stability of the injection molded part; performing a difference between the surface grayscale image at the target moment and the surface grayscale image at a previous moment adjacent to the target moment to obtain a differential image at the target moment, and calculating the dynamic stability of the injection molded part according to the differential image; calculating the comprehensive stability of the injection molded part at the target moment according to the static stability and the dynamic stability, and completing the uniformity detection; the comprehensive stability satisfies the relationship: , Indicates the sampling time The overall stability of Indicates the sampling time The static stability of Indicates the sampling time The dynamic stability of represents the adjustment factor, sampling time The target moment.

[0007] The surface grayscale image of the injection molded part during cooling is obtained by presetting the sampling interval. By analyzing the changes in the grayscale image, the uniformity of the cooling of the injection molded part can be quantified. By calculating the static stability and dynamic stability, the comprehensive stability is finally obtained. The temperature distribution during the injection molding process can be monitored and evaluated in real time, and uneven cooling can be discovered and adjusted in time to reduce defects and improve production efficiency.

[0008] Preferably, the element value in the grayscale exclusion matrix is ​​the number of differences between the pixel values ​​of the pixel point and the neighboring pixel points in the surface grayscale image.

[0009] The construction of a grayscale rejection matrix helps to identify grayscale changes and enhances the sensitivity to surface inhomogeneities of injection molded parts. By capturing local grayscale changes, the inhomogeneity of temperature distribution can be more accurately located and analyzed, thereby optimizing the cooling process.

[0010] Preferably, the static stability satisfies the relationship:

[0011] , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, Represents an exponential function.

[0012] The response to the difference in element values ​​in the grayscale rejection matrix is ​​strengthened through an exponential function, which enhances the sensitivity to small changes. This can highlight significant changes in temperature distribution, help to quickly identify and respond to temperature anomalies, and reduce product defects caused by temperature fluctuations.

[0013] Preferably, the static stability also satisfies the relationship:

[0014] , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, Represents the sum of all element values ​​in the grayscale rejection matrix, represents the exponential function, Represents a logarithmic function.

[0015] Preferably, calculating the dynamic stability includes: taking any row of gray value sequence in the differential image as a feature sequence, and taking any row of gray value sequence other than the feature sequence as a reference sequence; calculating the correlation between the feature sequence and the reference sequence, and calculating the dynamic stability according to the correlation, and the dynamic stability satisfies the relationship: , represents dynamic stability, Represents a feature sequence and reference sequence The relevance of represents the total number of rows of the difference image, Represents an exponential function.

[0016] By analyzing the correlation between the feature sequence and the reference sequence in the differential image to calculate the dynamic stability, it is possible to capture the temperature differences that change over time during the cooling process of the injection molded part. By evaluating the temperature change pattern in the time series, it is helpful to monitor and predict the temperature change trend and adjust the cooling conditions in time.

[0017] Preferably, the correlation includes: calculating the standard deviation of the characteristic sequence and the standard deviation of the reference sequence respectively, and calculating the product of the standard deviation of the characteristic sequence and the standard deviation of the reference sequence; calculating the covariance of the characteristic sequence and the reference sequence, and taking the ratio of the covariance and the product as the correlation.

[0018] Preferably, the correlation further includes: taking the Euclidean distance between the feature sequence and the reference sequence as the correlation.

[0019] Preferably, calculating the dynamic stability further comprises: dividing the differential image into gray levels to construct a gray level co-occurrence matrix, and calculating the dynamic stability according to the gray level co-occurrence matrix, wherein the dynamic stability satisfies the relationship: , represents dynamic stability, represents the gray-level co-occurrence matrix OK, represents the gray-level co-occurrence matrix List, express Line The value of the column, Represents the gray-level co-occurrence matrix The first moment along the axis, Represents the gray-level co-occurrence matrix The first moment along the axis, Represents the gray-level co-occurrence matrix The second moment about the axis, Represents the gray-level co-occurrence matrix The second moment about the axis, Represents the normalization function.

[0020] By combining the grayscale division of the differential image and the construction of the grayscale co-occurrence matrix, it is possible to capture the subtle texture differences of the temperature changes on the image, and characterize these changes through the statistical characteristics of the grayscale co-occurrence matrix. This method can provide real-time monitoring and feedback for the quality control of injection molded parts. By analyzing the temperature stability changes at multiple consecutive sampling moments, abnormalities in the cooling process can be discovered in time, so that measures can be taken to adjust the cooling conditions to ensure the uniformity and quality of the injection molded parts.

[0021] Preferably, the grayscale classification includes: dividing the product of any grayscale value and the total number of preset grayscale levels by 255 and rounding down the result as the grayscale level of any grayscale value.

[0022] Beneficial effects of the present invention:

[0023] 1. The present invention can not only monitor the temperature distribution of injection molded parts in real time during the cooling process, but also comprehensively evaluate the cooling effect of injection molded parts by calculating static and dynamic stability. By constructing a grayscale exclusion matrix, the temperature difference at each moment on the surface of the injection molded part can be captured, which is crucial for understanding the non-uniformity that may occur during the cooling process. In addition, by analyzing the differential image, the temperature change trend can be tracked, thereby providing continuous temperature change information during the cooling process.

[0024] 2. The present invention can improve the accuracy and reliability of injection molded parts cooling uniformity detection. By accurately calculating static stability and dynamic stability, and combining the adjustment factor to balance the relationship between the two, a more accurate comprehensive stability evaluation can be provided. This comprehensive evaluation helps to timely discover and correct the uneven temperature distribution during the cooling process of the injection molded parts, thereby reducing product quality problems caused by uneven cooling, such as warping, deformation and internal stress concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0026] Figure 1 It is a flow chart of a method for detecting cooling uniformity of injection molded parts based on image processing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0028] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0029] The present invention provides a method for detecting the cooling uniformity of injection molded parts based on image processing. Figure 1 As shown, the method for detecting cooling uniformity of injection molded parts based on image processing includes steps S1 to S4, which are described in detail below.

[0030] S1, obtaining a surface grayscale image of the injection molded part during cooling based on a preset sampling interval.

[0031] In one embodiment, a thermal imaging camera is used to collect RGB (Red Green Blue) images of automobile injection molding parts during cooling based on a preset sampling interval, and the collected RGB images are denoised and grayed to obtain a surface grayscale image, and the denoising method is Gaussian filtering. The purpose of obtaining the surface grayscale image of the injection molding part during cooling is to determine whether the injection molding part is cooled uniformly based on the changes of the injection molding part at each sampling moment during the entire cooling process, which is more accurate than judging whether the cooling is uniform only based on the final surface grayscale image after cooling.

[0032] It should be noted that using thermal imaging cameras to collect images of the cooling process of automotive injection molded parts and performing Gaussian filtering to remove noise is crucial for monitoring the uniformity of injection molded parts cooling. Thermal imaging technology can capture the temperature distribution on the surface of injection molded parts. By removing noise from these images, subtle differences in temperature changes can be more clearly identified, allowing for a more accurate assessment of whether cooling is uniform.

[0033] As a linear smoothing filter, Gaussian filtering is particularly suitable for removing Gaussian noise. It reduces random noise in the image, making the pattern of temperature change more obvious, which helps to improve the accuracy of subsequent analysis.

[0034] Collecting images of the entire cooling process of the injection molded part and continuously monitoring it can provide more dynamic information than judging whether the cooling is uniform based on the final cooling image alone. This method can capture the uneven cooling phenomenon that may occur during the cooling process of the injection molded part in real time, such as local overheating or insufficient cooling, which is crucial to ensuring product quality and performance.

[0035] S2, construct the grayscale rejection matrix at the target time and calculate the static stability of the injection molded parts.

[0036] It should be noted that static stability refers to the temperature stability of the injection molded parts at each sampling moment. During the cooling process of the injection molded parts, calculating the temperature stability or temperature uniformity at each sampling moment is crucial to ensuring the quality of the final product. The cooling stage in injection molding has a direct impact on the final surface quality, residual stress and crystallinity of the product. Effective cooling can ensure that the time required for plastic solidification is shortened, improve production capacity, and achieve uniform temperature distribution, reduce internal stress, warping and shrinkage, and improve the dimensional stability and surface finish of molded parts. The study of temperature uniformity helps to optimize process conditions, control mold temperature, prevent residual stress caused by uneven temperature and warping caused by uneven shrinkage, thereby shortening the molding cycle. In addition, by continuously monitoring the temperature changes during the cooling process, possible uneven cooling phenomena can be discovered and adjusted in a timely manner, which is crucial to ensuring product quality and performance.

[0037] In one embodiment, any sampling moment is taken as the target moment, and a grayscale exclusion matrix of the target moment is constructed. The element value in the grayscale exclusion matrix is ​​the number of differences between the pixel value of the pixel point and the neighboring pixel points in the surface grayscale image.

[0038] It should be noted that the size of the grayscale rejection matrix is ​​consistent with the size of the surface grayscale image. The grayscale rejection matrix is ​​constructed based on the surface grayscale image. Line The gray value of the column Line The number of differences between the grayscale value of a column and the grayscale values ​​of pixels in its neighborhood is taken as the grayscale exclusion matrix. Line The element value of the column, wherein the neighborhood is an 8-neighborhood or a 16-neighborhood, preferably, the neighborhood is an 8-neighborhood.

[0039] The grayscale exclusion matrix is ​​obtained by traversal, and the static stability is calculated according to the element values ​​of the grayscale exclusion matrix. The static stability satisfies the relationship:

[0040] , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, Represents an exponential function.

[0041] In another embodiment, the static stability further satisfies the relationship:

[0042] , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, Represents the sum of all element values ​​in the grayscale rejection matrix, represents the exponential function, Represents a logarithmic function.

[0043] S3, subtracting the surface grayscale image at the target moment from the surface grayscale image at the previous moment adjacent to the target moment to obtain a differential image at the target moment, and calculating the dynamic stability of the injection molded part according to the differential image.

[0044] It should be noted that continuous detection of temperature changes during the cooling process of injection molded parts can timely detect and adjust possible uneven cooling phenomena, which is crucial to ensuring the quality and performance of injection molded parts.

[0045] This inspection helps maintain the temperature uniformity of injection molded parts during the molding process, thereby reducing problems such as internal stress, warping and shrinkage. Temperature uniformity directly affects the dimensional stability and appearance quality of injection molded parts. Uneven cooling may cause problems such as uneven shrinkage and excessive internal stress in the product. Through continuous inspection, mold design can be optimized and the layout of the cooling system can be improved to ensure that the coolant flows evenly in the mold, effectively improving cooling uniformity. In addition, precise control of mold temperature and coolant temperature can ensure that the plastic is cooled and solidified within the optimal temperature range, improving product quality and production efficiency.

[0046] In one embodiment, a surface grayscale image at a target moment and a surface grayscale image at a previous moment adjacent to the target moment are subtracted to obtain a differential image at the target moment, any row of grayscale value sequence in the differential image is used as a feature sequence, and any row of grayscale value sequence other than the feature sequence is used as a reference sequence, and the correlation between the feature sequence and the reference sequence is calculated, and the correlation includes: calculating the feature sequence standard deviation and the reference sequence standard deviation respectively, and calculating the product of the feature sequence standard deviation and the reference sequence standard deviation; calculating the covariance of the feature sequence and the reference sequence, and taking the ratio of the covariance and the product as the correlation.

[0047] The correlation also includes: taking the Euclidean distance between the feature sequence and the reference sequence as the correlation.

[0048] The dynamic stability is calculated based on the correlation, and the dynamic stability satisfies the relationship:

[0049] , represents dynamic stability, Represents a feature sequence and reference sequence The relevance of represents the total number of rows of the difference image, Represents an exponential function.

[0050] In another embodiment, dynamic stability further comprises:

[0051] The grayscale level division is performed on the differential image, and the level division includes: dividing the product of any grayscale value and the total number of preset grayscale levels by 255 and rounding down the result as the grayscale level of any grayscale value. For example, if the grayscale value of a certain pixel in the differential image is 208, the differential image needs to be divided into 16 grayscale levels, that is, the total number of preset grayscale levels is 16, firstly the product of 208 and 16 is calculated, and then the ratio of the product to 255 is calculated to determine that the pixel with a grayscale value of 208 is divided into level 13.

[0052] A gray level co-occurrence matrix is ​​constructed according to the divided gray levels. Constructing a gray level co-occurrence matrix is ​​a well-known technology for those skilled in the art and will not be described in detail here.

[0053] The dynamic stability is calculated based on the gray level co-occurrence matrix, and the dynamic stability satisfies the relationship:

[0054] , represents dynamic stability, represents the gray-level co-occurrence matrix OK, represents the gray-level co-occurrence matrix List, express Line The value of the column, Represents the gray-level co-occurrence matrix The first moment along the axis, Represents the gray-level co-occurrence matrix The first moment along the axis, Represents the gray-level co-occurrence matrix The second moment about the axis, Represents the gray-level co-occurrence matrix The second moment about the axis, Represents the normalization function.

[0055] S4, calculates the comprehensive stability of the injection molded part at the target moment based on the static stability and dynamic stability, and completes the uniformity test.

[0056] In one embodiment, the comprehensive stability of the injection molded part at the target time is calculated based on the static stability and the dynamic stability, and the comprehensive stability satisfies the relationship:

[0057] , Indicates the sampling time The overall stability of Indicates the sampling time The static stability of Indicates the sampling time The dynamic stability of represents the adjustment factor, sampling time The target moment.

[0058] The adjustment factor is used to measure the importance of static stability and dynamic stability. When the adjustment factor is greater than 1, it means that dynamic stability is more important; when the adjustment factor is less than 1, it means that static stability is more important; when the adjustment factor is equal to 1, the importance of static stability and dynamic stability is the same.

[0059] In the present invention, the adjustment factor is preferably set to 1.2, because if the dynamic stability is large, it means that the cooling uniformity of the injection molded parts at multiple consecutive sampling moments is consistent, that is, the static stability consistency at each moment is high, so the dynamic stability is more important than the static stability. However, only the dynamic stability can be considered, because although the static stability consistency at each moment is high, there will be two situations where the static stability at each moment is very good or the static stability at each moment is very poor, so it is necessary to consider comprehensively. The specific value can be set by those skilled in the art according to the actual situation.

[0060] At this point, the cooling uniformity test of the injection molded parts is completed.

[0061] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0062] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting cooling uniformity of injection molded parts based on image processing, characterized in that: include: The surface grayscale image of the injection molded part during cooling is obtained based on a preset sampling interval, and any sampling moment is taken as the target moment; By obtaining the surface grayscale image of the injection molded part during cooling, the cooling uniformity of the injection molded part can be judged according to the temperature change of the injection molded part at each sampling moment during the entire cooling process; Construct a grayscale rejection matrix at the target time, wherein the element value in the grayscale rejection matrix is ​​the number of different pixel values ​​between the pixel point and the neighboring pixel point in the surface grayscale image; calculate the static stability of the injection molded part to satisfy the relationship: , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, represents the exponential function; The surface grayscale image at the target moment is subtracted from the surface grayscale image at the previous moment adjacent to the target moment to obtain the differential image at the target moment, and the dynamic stability of the injection molded part is calculated based on the differential image, including: Any row of gray value sequence in the differential image is used as a feature sequence, and any row of gray value sequence other than the feature sequence is used as a reference sequence; Calculate the correlation between the characteristic sequence and the reference sequence, and calculate the dynamic stability based on the correlation. The dynamic stability satisfies the relationship: , represents dynamic stability, Represents a feature sequence and reference sequence The relevance of Indicates the total number of rows of the difference image; Calculate the comprehensive stability of the injection molded parts at the target time based on static stability and dynamic stability, and complete uniformity testing; The comprehensive stability satisfies the relationship: , Indicates the sampling time The overall stability of Indicates the sampling time The static stability of Indicates the sampling time The dynamic stability of represents the adjustment factor, sampling time The target moment.

2. The method for detecting cooling uniformity of injection molded parts based on image processing according to claim 1, characterized in that: The relationship for calculating static stability can also be replaced by: , represents static stability, Represents the grayscale rejection matrix Line The element value of the column, Represents the sum of all element values ​​in the grayscale rejection matrix, represents the exponential function, Represents a logarithmic function.

3. The method for detecting cooling uniformity of injection molded parts based on image processing according to claim 1, characterized in that: The dependencies include: Calculate the characteristic sequence standard deviation and the reference sequence standard deviation respectively, and calculate the product of the characteristic sequence standard deviation and the reference sequence standard deviation; The covariance of the feature sequence and the reference sequence is calculated, and the ratio of the covariance and the product is taken as the correlation.

4. The method for detecting cooling uniformity of injection molded parts based on image processing according to claim 1, characterized in that: The correlation also includes: The Euclidean distance between the feature sequence and the reference sequence is taken as the correlation.

Images