Injection molding machine control method, system, medium and program product based on visual algorithm

Abstract

A method, system, medium, and program product for controlling an injection molding machine based on a visual algorithm relate to the general field of image data processing. In this method, the amount of pixel change in a target area is determined; the short-term image distortion characteristics are determined based on the amount of pixel change in the target area; the image distortion portion caused by thermal disturbances and the image change portion caused by the flow of molten plastic are separated; the thermal disturbance intensity at the injection molding machine nozzle is determined based on the distribution pattern of the image distortion portion caused by thermal disturbances; the actual flow state of the molten plastic is calculated based on the image change portion caused by the flow of molten plastic and the thermal disturbance intensity; and the process parameters of the injection molding machine are controlled based on the actual flow state. This application achieves accurate separation of the image distortion portion caused by thermal disturbances and the image change portion caused by the actual flow of molten plastic, thereby reducing the interference of thermal disturbances on visual monitoring during the injection molding process of transparent plastic optical lenses and improving the adjustment accuracy of injection molding parameters.

Description

Technical Field

[0001] The present application belongs to the general field of image data processing, and in particular relates to an injection molding machine control method, system, medium and program product based on a visual algorithm. Background Art

[0002] With the development of intelligent manufacturing, injection molding machine control technology is also constantly improving. Traditional injection molding machine control mainly relies on manual experience and fixed parameter settings. This control method has problems such as low control accuracy, large fluctuations in product quality, and low production efficiency. It is difficult to meet the requirements of modern manufacturing for product quality and production efficiency.

[0003] In the related art, a machine vision-based control method for injection molding machines is commonly used. This method uses a camera installed at the nozzle of the injection molding machine to capture images of the molten plastic during the injection molding process. Combined with an image processing algorithm, the flow state of the molten plastic is monitored in real time, thereby automatically adjusting the injection molding parameters. This method can, to a certain extent, improve the control accuracy of the injection molding machine and the stability of product quality.

[0004] However, during the injection molding process for producing high-precision transparent plastic optical lenses, due to the high transparency of the molten plastic and the high temperature required for injection molding, the heat at the nozzle causes air convection, resulting in thermal disturbances. This thermal disturbance causes irregular refraction of light in the air, resulting in localized distortion and blurring of the molten plastic image captured by the camera. This thermal disturbance is particularly pronounced during continuous high-speed injection molding, reducing the accuracy of the visual algorithm in identifying the flow state of the molten plastic, thereby reducing the precision of the injection molding parameter adjustments. Summary of the Invention

[0005] The present application provides an injection molding machine control method, system, medium and program product based on a visual algorithm, which realizes the accurate separation of the image distortion part caused by thermal disturbance and the image change part caused by the actual flow of molten plastic, thereby reducing the interference of thermal disturbance on visual monitoring during the injection molding process of transparent plastic optical lenses and improving the adjustment accuracy of injection molding parameters.

[0006] In a first aspect, the present application provides a method for controlling an injection molding machine based on a visual algorithm, which obtains an image of molten plastic at a nozzle of the injection molding machine at a first moment and an image of molten plastic at a second moment, wherein the time interval between the first moment and the second moment is less than a preset threshold;

[0007] Determine a pixel change amount of a target area in the molten plastic image at a first moment and the molten plastic image at a second moment;

[0008] Determine the short-term image distortion characteristics caused by thermal disturbance based on the pixel change in the target area;

[0009] Based on the short-term image distortion characteristics, the image distortion part caused by thermal disturbance and the image change part caused by the flow of molten plastic are separated from the real-time molten plastic image;

[0010] Determine the thermal disturbance intensity at the injection molding machine nozzle based on the distribution pattern of the image distortion caused by thermal disturbance;

[0011] Calculate the actual flow state of the molten plastic according to the image change portion caused by the flow of the molten plastic and the thermal disturbance intensity;

[0012] Control the process parameters of the injection molding machine according to the actual flow state.

[0013] By adopting the above technical solution, by acquiring two images of molten plastic within a short time interval and analyzing the pixel changes in the target area, the short-term image distortion characteristics caused by thermal disturbances can be identified. Based on this characteristic, the image distortion caused by thermal disturbances and the image changes caused by the flow of molten plastic in real-time molten plastic images can be effectively separated. By analyzing the distribution pattern of the image distortion caused by thermal disturbances, the intensity of thermal disturbances at the injection molding machine nozzle can be accurately determined. Combining the thermal disturbance intensity with the image changes caused by the flow of molten plastic, the actual flow state of the molten plastic can be accurately calculated, thereby reducing the interference of thermal disturbances on the judgment of the plastic flow state. Controlling the process parameters of the injection molding machine based on the accurate actual flow state improves the control accuracy of the injection molding process and reduces the process parameter adjustment deviation caused by misjudgment due to thermal disturbances.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, determining the pixel change amount of the target area in the molten plastic image at the first moment and the molten plastic image at the second moment specifically includes:

[0015] Dividing the molten plastic image at the first moment and the molten plastic image at the second moment into corresponding multiple grid units respectively;

[0016] Calculating a pixel grayscale mean value for each grid cell in the molten plastic image at the first moment to obtain a first grayscale distribution matrix;

[0017] calculating a pixel grayscale mean value for each grid cell in the molten plastic image at the second moment to obtain a second grayscale distribution matrix;

[0018] Calculate the grayscale difference between the grid cells corresponding to the first grayscale distribution matrix and the second grayscale distribution matrix to obtain a grayscale difference matrix;

[0019] Performing feature point matching on corresponding grid cells in the molten plastic image at the first moment and the molten plastic image at the second moment to obtain a feature point matching result;

[0020] Calculate the pixel position offset matrix of the corresponding grid unit based on the feature point matching results;

[0021] Determining an area where the value of the grayscale difference matrix is ​​greater than a first preset threshold as a thermal disturbance affected area;

[0022] Calculate the gradient value of the pixel position offset matrix within the thermal disturbance influence area;

[0023] The pixel change in the target area is calculated based on the gradient value.

[0024] By adopting the above technical solution, the molten plastic image is divided into multiple grid cells and the mean grayscale value of the pixels in each grid cell is calculated to construct a grayscale distribution matrix. Combined with the pixel position offset matrix obtained by feature point matching, image changes caused by thermal disturbances can be detected on a fine scale. By setting a grayscale difference threshold to determine the area affected by the thermal disturbance and calculating the gradient value of the pixel position offset matrix within this area, the degree of impact of the thermal disturbance on the image can be quantified. This grid cell-based analysis method provides higher spatial resolution, making the detection of thermal disturbances more accurate. The pixel change in the target area obtained by gradient value calculation can accurately reflect the spatial distribution characteristics of the thermal disturbance, improving the accuracy of subsequent thermal disturbance analysis and compensation.

[0025] In conjunction with some embodiments of the first aspect, in some embodiments, determining the short-term image distortion characteristics caused by thermal disturbance based on the pixel change amount in the target area specifically includes:

[0026] Construct a time series data sequence based on the pixel change in the target area;

[0027] Perform Fourier transform on the time series data to obtain the frequency domain characteristic spectrum;

[0028] Determining a characteristic spectrum having a frequency greater than a preset threshold as a high-frequency characteristic spectrum;

[0029] Perform spatial distribution analysis on the high-frequency characteristic spectrum to obtain the distortion direction and distortion amplitude;

[0030] Construct a distortion feature vector based on the distortion direction and distortion amplitude;

[0031] The distortion feature vector is mapped to the image space to obtain the short-term image distortion features caused by thermal disturbance.

[0032] By employing this technical solution, the frequency domain characteristic spectrum is obtained by Fourier transforming the time series data constructed from the pixel changes in the target area, allowing the characteristics of thermal disturbances to be separated in the frequency domain. By filtering the high-frequency characteristic spectrum using a preset frequency threshold and analyzing its spatial distribution, the resulting distortion direction and amplitude fully describe the dynamic characteristics of thermal disturbances. The short-term image distortion characteristics obtained by mapping the distortion feature vector to image space accurately reflect the mode of action and intensity distribution of thermal disturbances in image space.

[0033] In conjunction with some embodiments of the first aspect, in some embodiments, calculating the actual flow state of the molten plastic based on the image change portion caused by the flow of the molten plastic and the thermal disturbance intensity specifically includes:

[0034] Establishing a spatial distortion compensation coefficient based on the thermal disturbance intensity;

[0035] Construct a three-dimensional distortion correction matrix according to the spatial distortion compensation coefficient;

[0036] Applying the three-dimensional distortion correction matrix to the image change caused by the flow of molten plastic to obtain a corrected flow image;

[0037] Extracting the flow boundary contour of the molten plastic in the corrected flow image;

[0038] Calculate the displacement vectors of the flow boundary contours at adjacent moments;

[0039] Calculate the flow velocity field and flow direction field of the molten plastic based on the displacement vector;

[0040] The actual flow state of the molten plastic is obtained based on the flow velocity field and flow direction field.

[0041] By adopting the above technical solution, by establishing a spatial distortion compensation coefficient based on the intensity of thermal disturbances and constructing a three-dimensional distortion correction matrix, the molten plastic flow image can be accurately corrected for distortion. By extracting the flow boundary contours from the corrected flow image and calculating the displacement vectors at adjacent moments, the flow velocity field and flow direction field of the molten plastic can be accurately obtained. This correction method, which takes into account the three-dimensional spatial distortion effect, reduces the measurement error caused by ignoring spatial distortion in traditional two-dimensional image analysis and provides more accurate flow state calculation results. The flow state calculation method achieved through spatial distortion compensation and three-dimensional correction improves the measurement accuracy of molten plastic flow parameters.

[0042] In conjunction with some embodiments of the first aspect, in some embodiments, controlling the process parameters of the injection molding machine according to the actual flow state specifically includes:

[0043] Calculate the shear rate distribution of the molten plastic based on the flow velocity field in the actual flow state;

[0044] determining the shear stress distribution of the molten plastic based on the shear rate distribution;

[0045] Calculate the apparent viscosity of the molten plastic based on the shear stress distribution and shear rate distribution, and determine the adjustment parameters of the injection pressure and injection speed based on the comparison of the change trend of the apparent viscosity with the preset viscosity curve;

[0046] The injection pressure and injection speed of the injection molding machine are synchronously adjusted according to the adjustment parameters to obtain the adjusted process parameters;

[0047] The injection molding process of the injection molding machine is controlled according to the adjusted process parameters.

[0048] By adopting the above technical solution, by calculating the shear rate distribution of the molten plastic based on the flow velocity field in the actual flow state, and determining the shear stress distribution of the molten plastic based on the shear rate distribution, the stress conditions of the molten plastic during the flow process can be accurately understood. The apparent viscosity calculated based on the shear stress distribution and the shear rate distribution can truly reflect the rheological properties of the molten plastic. By comparing the changing trend of the apparent viscosity with the preset viscosity curve, the adjustment parameters of the injection pressure and injection speed can be scientifically determined. Synchronously adjusting the injection molding machine's injection pressure and injection speed based on these adjustment parameters can match the injection molding process parameters with the actual rheological properties of the molten plastic, achieve precise adjustment of the injection molding parameters, improve the quality stability of the molded products, and reduce the scrap rate caused by unreasonable process parameters.

[0049] In conjunction with some embodiments of the first aspect, in some embodiments, after controlling the process parameters of the injection molding machine according to the actual flow state, the method further includes:

[0050] Between the first moment and the second moment, a molten plastic image is acquired at every preset sampling time interval to obtain a time sequence image group;

[0051] Calculate the pixel difference of the target area of ​​two adjacent molten plastic images in the time sequence image group;

[0052] When it is determined that the pixel difference value of the target area is greater than a second preset threshold, the molten plastic image at the corresponding moment is used as a key image;

[0053] The short-term image distortion features are updated according to the key image.

[0054] By adopting the above technical solution, by acquiring images of molten plastic in a continuous time series to form a time-series image group, calculating the pixel differences of the target area of ​​adjacent images, and setting a threshold to filter key images, the moment when the molten plastic state changes significantly can be quickly located within a large amount of image data. This pixel difference-based key image recognition method can promptly capture abnormal changes in the molten plastic flow process. By analyzing key images and updating short-term image distortion features, the system can dynamically track the impact of thermal disturbances on image quality, ensuring the accuracy of image processing results. This adaptive image feature update mechanism enables the system to continuously learn and optimize, adapting to various dynamic changes in the injection molding process, better responding to various interference factors in actual production, and ensuring the long-term stable operation of the system.

[0055] In conjunction with some embodiments of the first aspect, in some embodiments, updating the short-term image distortion feature according to the key image specifically includes:

[0056] Compare the key image with the molten plastic images at two adjacent moments to obtain a grayscale difference matrix;

[0057] Determine the continuous pixel area with the largest difference according to the grayscale difference matrix, and obtain the coordinates of the distortion focus area;

[0058] Calculate the grayscale change vectors of adjacent pixels within the coordinate range of the distortion focus area to obtain the distortion direction matrix;

[0059] Calculate the main distortion direction and distortion amplitude of the distortion focus area according to the distortion direction matrix to obtain updated distortion feature parameters;

[0060] The updated distortion feature parameters are used to replace the parameters of the corresponding area in the short-time image distortion feature to obtain the updated short-time image distortion feature.

[0061] By employing this technical solution, pixel-by-pixel comparisons between the key image and images at adjacent moments and analyzing the grayscale difference matrix can precisely locate areas of interest for image distortion. By calculating the grayscale change vectors of adjacent pixels within the defined distortion area of ​​interest, the direction of the distortion can be accurately determined. The principal distortion direction and magnitude calculated from the distortion direction matrix can quantitatively characterize image distortion caused by thermal disturbances. This distortion feature update method, based on local area analysis, improves system efficiency.

[0062] In the second aspect, an embodiment of the present application provides an injection molding machine control system based on a visual algorithm, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions, and one or more processors call the computer instructions to enable the system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0063] In a third aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions, which, when executed on a system, enables the system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0064] In a fourth aspect, an embodiment of the present application provides a computer program product, which, when executed on a system, enables the system to execute the method described in any possible implementation manner in the first aspect.

[0065] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0066] 1. The present application provides an injection molding machine control method based on a visual algorithm. By acquiring two molten plastic images within a short time interval and analyzing the pixel changes in the target area, the short-term image distortion characteristics caused by thermal disturbances can be identified. Based on this feature, the image distortion part caused by thermal disturbances and the image change part caused by the flow of molten plastic in the real-time molten plastic image can be effectively separated. By analyzing the distribution pattern of the image distortion part caused by thermal disturbances, the thermal disturbance intensity at the injection molding machine nozzle can be accurately determined. Combining the thermal disturbance intensity with the image change part caused by the flow of molten plastic for analysis, the actual flow state of the molten plastic can be accurately calculated, thereby reducing the interference of thermal disturbances on the judgment of the plastic flow state. By controlling the process parameters of the injection molding machine according to the accurate actual flow state, the control accuracy of the injection molding process is improved, and the process parameter adjustment deviation caused by misjudgment due to thermal disturbances is reduced.

[0067] 2. The present application provides an injection molding machine control method based on a visual algorithm. By acquiring molten plastic images in a continuous time series to form a time-series image group, calculating the pixel difference of the target area of ​​adjacent images and setting a threshold to filter the key image, the moment when the state of the molten plastic changes significantly can be quickly located in a large amount of image data. This key image recognition method based on pixel difference can timely capture abnormal changes in the flow process of molten plastic. By analyzing the key images and updating the short-term image distortion features, the system can dynamically track the impact of thermal disturbances on image quality and ensure the accuracy of the image processing results. This adaptive image feature update mechanism enables the system to have the ability to continuously learn and optimize, can adapt to various dynamic changes in the injection molding process, can better deal with various interference factors in actual production, and ensure the long-term stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a flow chart of an injection molding machine control method based on a visual algorithm in an embodiment of the present application.

[0069] Figure 2 This is a flow chart of a method for dynamically updating thermal disturbance features based on a key image in an embodiment of the present application.

[0070] Figure 3 This is a schematic diagram of the physical device structure of an injection molding machine control system based on a visual algorithm provided in an embodiment of the present application.

[0071] Explanation of reference numerals: 301 -CPU; 302 -ROM; 303 -RAM; 304 -bus; 305 -I / O interface; 306 -input portion; 307 -output portion; 308 -storage portion; 309 -communication portion; 310 -drive; 311 -removable medium. DETAILED DESCRIPTION

[0072] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0073] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0074] The following uses an embodiment and combines Figure 1 , a control method for an injection molding machine based on a visual algorithm in an embodiment of the present application is described:

[0075] See also Figure 1 , which is a flow chart of an injection molding machine control method based on a visual algorithm in an embodiment of the present application.

[0076] S101, acquiring an image of molten plastic at a nozzle of an injection molding machine at a first moment and an image of molten plastic at a second moment;

[0077] The system obtains an image of molten plastic at a first moment and an image of molten plastic at a second moment at a nozzle of an injection molding machine, wherein the time interval between the first moment and the second moment is less than a preset threshold.

[0078] The system can capture images of molten plastic using various imaging devices, such as visible light cameras and infrared cameras. The acquisition interval can be flexibly set according to actual needs. Furthermore, the system can perform pre-processing on the captured images, such as denoising and enhancement, to improve image quality.

[0079] Specifically, the system can install a high-speed camera at the injection molding machine nozzle to continuously capture images of the molten plastic at a specific frame rate. Furthermore, the system can set a timer. When the time interval between the acquisition of two frames is less than a preset threshold, these two frames are transmitted to the subsequent processing module as the first and second images of the molten plastic at the time of capture. Furthermore, to improve image contrast and clarity, the system can perform image enhancement operations such as histogram equalization and sharpening on the captured raw images.

[0080] S102, determining a pixel change amount of a target area in the molten plastic image at the first moment and the molten plastic image at the second moment;

[0081] The system determines the pixel change amount of the target area in the molten plastic image at the first moment and the molten plastic image at the second moment, specifically including: dividing the molten plastic image at the first moment and the molten plastic image at the second moment into corresponding multiple grid units respectively; calculating the pixel grayscale mean of each grid unit in the molten plastic image at the first moment to obtain a first grayscale distribution matrix; calculating the pixel grayscale mean of each grid unit in the molten plastic image at the second moment to obtain a second grayscale distribution matrix; calculating the grayscale difference between the grid units corresponding to the first grayscale distribution matrix and the second grayscale distribution matrix to obtain a grayscale difference matrix; performing feature point matching on the corresponding grid units in the molten plastic image at the first moment and the molten plastic image at the second moment to obtain a feature point matching result; calculating the pixel position offset matrix of the corresponding grid unit based on the feature point matching result; determining the area where the value of the grayscale difference matrix is ​​greater than a first preset threshold as a thermal disturbance influence area; calculating the gradient value of the pixel position offset matrix in the thermal disturbance influence area; and calculating the pixel change amount of the target area based on the gradient value.

[0082] Specifically, the system can use a uniform gridding method to divide the image into rectangular grid cells of equal size. For each grid cell, the system calculates the arithmetic mean of the grayscale values ​​of all pixels within it as the grayscale mean of the cell. By performing gridding and grayscale mean calculation on the images at two time points, the system generates two grayscale distribution matrices. Next, the system subtracts the elements in the two grayscale distribution matrices at the same position to generate a grayscale difference matrix, which reflects the grayscale change of pixels between the two images at the two time points. Regions with large grayscale differences typically correspond to areas where the molten plastic has deformed due to thermal disturbances. To further analyze the pixel motion within these regions, the system uses a feature point matching algorithm, such as SIFT or ORB, to find matching feature point pairs within corresponding grid cells in the two images at the two time points. The system then calculates the positional offset of these feature point pairs to generate a pixel position offset matrix. Finally, the system calculates the gradient of the position offset matrix. A larger gradient indicates more intense pixel motion, indicating a stronger thermal disturbance.

[0083] When calculating the grayscale difference matrix, due to the presence of noise and slight brightness changes in the image, a large number of non-zero values ​​with small amplitudes may appear in the grayscale difference matrix, affecting the determination of the thermal disturbance area. To this end, the system can binarize the grayscale difference matrix, setting the elements with differences greater than the threshold to 1 and the remaining elements to 0, to obtain a binary difference matrix. In this way, the system can quickly determine the location and range of the area affected by the thermal disturbance based on the distribution of 1 in the binary difference matrix. When performing feature point matching, due to the relatively simple texture of the molten plastic image, directly using traditional feature point extraction algorithms may result in fewer matching point pairs, affecting the calculation accuracy of the position offset matrix. To solve this problem, the system can first perform texture enhancement on the image, such as adding local noise, gradient, etc., and then perform feature point extraction and matching to obtain more and more reliable matching point pairs, thereby improving the calculation accuracy of the position offset matrix.

[0084] S103, determining the short-term image distortion characteristics caused by thermal disturbance according to the pixel change amount in the target area;

[0085] The system determines the short-term image distortion characteristics caused by thermal disturbances based on the pixel changes in the target area, specifically including: constructing a time series data sequence based on the pixel changes in the target area; performing Fourier transform on the time series data sequence to obtain a frequency domain characteristic spectrum; determining the characteristic spectrum greater than a preset frequency threshold as a high-frequency characteristic spectrum; performing spatial distribution analysis on the high-frequency characteristic spectrum to obtain the distortion direction and distortion amplitude; constructing a distortion feature vector based on the distortion direction and distortion amplitude; and mapping the distortion feature vector to the image space to obtain the short-term image distortion characteristics caused by thermal disturbances.

[0086] Specifically, the system uses a sliding window approach to organize the pixel changes in the target area into a time-series data sequence. The system then performs a fast Fourier transform (FFT) on this time-series data sequence to obtain its frequency domain signature. By analyzing the frequency domain signature, the system finds that pixel changes caused by thermal disturbances typically manifest as high-frequency components, while pixel changes caused by the normal flow of molten plastic are primarily concentrated in low-frequency components. Therefore, the system sets a frequency threshold and extracts the portion of the frequency domain signature spectrum above this threshold as the high-frequency signature, reflecting the short-term image distortion characteristics caused by thermal disturbances. The system then performs a spatial distribution analysis of the high-frequency signature spectrum to determine the primary direction and relative magnitude of distortion in different regions. Distortion feature vectors are constructed using these two parameters: direction and magnitude. Finally, the system uses an inverse Fourier transform to map the distortion feature vector from the frequency domain back to the image space, resulting in a two-dimensional distortion feature matrix with the same size as the original image. Each element in the matrix represents the direction and magnitude of the distortion at the corresponding pixel due to thermal disturbances.

[0087] When performing frequency domain analysis, since the Fourier transform assumes that the signal is periodic, and the actual pixel change data collected may not meet the periodic conditions, directly performing an FFT on it may introduce errors such as spectrum leakage. To reduce this error, the system can add a section of zero-value data at the beginning and end of the time series data sequence to make it meet the periodic conditions before performing an FFT. In addition, when extracting high-frequency feature spectra, selecting an appropriate frequency threshold is a key issue. If the threshold is too low, the extracted high-frequency component may contain more pixel change information caused by normal flow; if the threshold is too high, the extracted high-frequency component may be too small, affecting the accuracy of subsequent spatial distribution analysis. To this end, the system can adaptively adjust the frequency threshold based on the process parameters of the injection molding machine and previous process experience to achieve the purpose of accurately extracting the high-frequency distortion characteristics caused by thermal disturbances.

[0088] S104, separating the image distortion portion caused by thermal disturbance and the image change portion caused by molten plastic flow from the real-time molten plastic image based on the short-term image distortion feature;

[0089] Based on the short-term image distortion features obtained in step S103, the system separates the image distortion caused by thermal disturbances from the image changes caused by the normal flow of the molten plastic from the real-time captured molten plastic image. This step aims to further distinguish the impact of thermal disturbances and normal flow on the molten plastic molding process, providing a basis for subsequent optimization of injection molding process parameters.

[0090] Specifically, the system uses the short-term image distortion characteristics as a filter template and performs a convolution operation on the real-time image of the molten plastic to obtain the image distortion caused by thermal disturbances. During the convolution process, areas with large distortion amplitudes in the filter template have a greater impact on the image, while areas with small distortion amplitudes have a smaller impact. By subtracting this from the original image, the system obtains the image changes after removing the influence of thermal disturbances. This portion mainly reflects the normal flow state of the molten plastic. In this way, the system effectively separates image distortion caused by thermal disturbances from image changes caused by normal flow.

[0091] During the actual separation process, the original image may contain other noise interference. Using a single distortion feature filter template is difficult to completely remove these interferences, which affects the separation effect. To this end, the system can adopt a multi-scale convolution filtering strategy. The original image is filtered by multiple distortion feature templates of different scales to obtain multiple image distortion parts. These distorted parts are then weighted and fused to effectively suppress the influence of other noise and improve the accuracy of thermal disturbance distortion separation. In addition, when applying the distortion feature filter template, the directionality of the template needs to be considered. Since the image distortion caused by thermal disturbance often has a certain directionality in space, the system can construct a set of directional filter templates based on the directional information of the distortion feature vector to enhance the extraction effect of distortion features from different directions, further improving the accuracy of separating image distortion and normal changes.

[0092] S105, determining the thermal disturbance intensity at the nozzle of the injection molding machine according to the distribution pattern of the image distortion portion caused by the thermal disturbance;

[0093] The system separates the image distortion caused by thermal disturbances in step S104, analyzes its spatial distribution, and determines the thermal disturbance intensity at the injection molding machine nozzle. Specifically, the system binarizes the distorted image to obtain a binary image of the thermal disturbance-affected area. The system then performs morphological analysis on the binary image to extract geometric parameters of the thermal disturbance-affected area, such as area, perimeter, and eccentricity. By tracking and analyzing the temporal trends of these geometric parameters, the system can determine the spatial distribution of the thermal disturbance-affected area. For example, a larger area indicates a wider range of thermal disturbances; a larger eccentricity indicates a more uneven distribution of thermal disturbances. Finally, based on the spatial distribution of the affected area and combined with injection molding machine process parameters such as nozzle temperature and ambient temperature, the system calculates the thermal disturbance intensity at the nozzle using empirical formulas or machine learning models. This value reflects the degree of impact of thermal disturbances on the stability of the injection molding process. A larger value indicates a stronger thermal disturbance and a more unstable injection molding process.

[0094] When calculating thermal disturbance intensity, considering only image distortion information at a single moment may not be comprehensive; the temporal variation also contains important information. To this end, the system constructs three-dimensional spatiotemporal data from the image distortion over a period of time. Using spatiotemporal data analysis methods, it extracts temporal variations in the thermal disturbance-affected area, such as velocity and acceleration, and uses these as supplementary indicators for thermal disturbance intensity assessment. Furthermore, the injection molding process is a complex, nonlinear process, and the relationship between thermal disturbance intensity and process parameters is often difficult to describe using simple empirical formulas. Therefore, the system can also employ machine learning methods, such as support vector machines and neural networks, to establish a nonlinear mapping model between thermal disturbance intensity, process parameters, and image distortion characteristics. This model can then be trained and optimized using extensive historical data to improve the accuracy of thermal disturbance intensity assessment. This system thus considers not only the spatial distribution of thermal disturbance effects but also their temporal dynamics and their inherent relationship with process parameters, enabling a more comprehensive and accurate assessment of thermal disturbance intensity during the injection molding process.

[0095] S106, calculating the actual flow state of the molten plastic based on the image change portion caused by the flow of the molten plastic and the thermal disturbance intensity;

[0096] The system calculates the actual flow state of the molten plastic based on the image changes caused by the flow of the molten plastic and the thermal disturbance intensity, specifically including: establishing a spatial distortion compensation coefficient based on the thermal disturbance intensity; constructing a three-dimensional distortion correction matrix based on the spatial distortion compensation coefficient; applying the three-dimensional distortion correction matrix to the image changes caused by the flow of the molten plastic to obtain a corrected flow image; extracting the flow boundary contour of the molten plastic in the corrected flow image; calculating the displacement vector of the flow boundary contour at adjacent moments; calculating the flow velocity field and flow direction field of the molten plastic based on the displacement vector; and obtaining the actual flow state of the molten plastic based on the flow velocity field and flow direction field.

[0097] This step aims to calculate the actual flow state of the molten plastic based on the image changes caused by the molten plastic flow and the thermal disturbance intensity obtained in the previous step. The actual flow state here can include information such as the molten plastic's flow velocity field and flow direction field. The system comprehensively considers the image distortion caused by thermal disturbances and the flow characteristics of the plastic itself to obtain a more accurate flow state estimate. In addition to the specific implementation methods mentioned in this article, the system can also use other image processing and fluid dynamics analysis methods to complete this step.

[0098] Specifically, the system first establishes a spatial distortion compensation coefficient based on the thermal disturbance intensity obtained in the previous step, which is used to quantify the degree of spatial distortion caused by thermal disturbances on the image. Then, the system constructs a three-dimensional distortion correction matrix based on the compensation coefficient and applies it to the image changes caused by the flow of molten plastic to correct the image and eliminate the distortion caused by thermal disturbances. Next, the system extracts the flow boundary contour of the molten plastic in the corrected flow image, calculates the displacement vector of the flow boundary contour at adjacent moments, and calculates the flow velocity field and flow direction field of the molten plastic based on this. Finally, the system obtains the actual flow state of the molten plastic based on the flow velocity field and flow direction field.

[0099] S107. Control the process parameters of the injection molding machine according to the actual flow state.

[0100] The system controls the process parameters of the injection molding machine according to the actual flow state, specifically including: calculating the shear rate distribution of the molten plastic according to the flow velocity field in the actual flow state; determining the shear stress distribution of the molten plastic based on the shear rate distribution; calculating the apparent viscosity of the molten plastic according to the shear stress distribution and the shear rate distribution, and determining the adjustment parameters of the injection pressure and injection speed according to the comparison results of the change trend of the apparent viscosity with the preset viscosity curve; synchronously adjusting the injection pressure and injection speed of the injection molding machine according to the adjustment parameters to obtain the adjusted process parameters; and controlling the injection molding process of the injection molding machine according to the adjusted process parameters.

[0101] The purpose of this step is to dynamically adjust and optimize the injection molding machine's process parameters based on the actual flow state of the molten plastic obtained in the previous step to ensure the quality of the molded part. These process parameters can include injection pressure, injection speed, barrel temperature, and other parameters. By monitoring the plastic's flow state in real time and providing corresponding feedback control instructions, the system can achieve closed-loop optimization of the injection molding process, improving production efficiency and product quality. Of course, in addition to the control methods mentioned in this article, the system can also adopt other control algorithms and strategies to complete this step.

[0102] As described in the paper, the system first calculates the shear rate distribution of the molten plastic based on the flow velocity field in the actual flow state. It then determines the shear stress distribution of the molten plastic based on the shear rate distribution. Next, the system calculates the apparent viscosity of the molten plastic based on the shear stress and shear rate distributions, compares its changing trend with a preset viscosity curve, and determines the adjustment parameters for injection pressure and speed. Finally, the system synchronously adjusts the injection molding machine's injection pressure and speed based on the adjustment parameters, obtaining the adjusted process parameters, and then controls the injection molding process based on the adjusted process parameters.

[0103] In the above embodiment, by acquiring two images of molten plastic within a short time interval and analyzing the pixel changes in the target area, the short-term image distortion characteristics caused by thermal disturbances can be identified. Based on this characteristic, the image distortion portion caused by thermal disturbances and the image change portion caused by the flow of molten plastic in the real-time molten plastic image can be effectively separated. By analyzing the distribution pattern of the image distortion portion caused by thermal disturbances, the thermal disturbance intensity at the injection molding machine nozzle can be accurately determined. Combining the thermal disturbance intensity with the image change portion caused by the flow of molten plastic for analysis, the actual flow state of the molten plastic can be accurately calculated, thereby reducing the interference of thermal disturbances on the judgment of the plastic flow state. By controlling the process parameters of the injection molding machine according to the accurate actual flow state, the control accuracy of the injection molding process is improved, and the process parameter adjustment deviation caused by misjudgment due to thermal disturbances is reduced.

[0104] The injection molding machine control method based on visual algorithms in the above embodiment achieves accurate judgment of the flow state of molten plastic by analyzing short-term image features. However, in actual application, the characteristics of thermal disturbances may change dynamically as the injection molding process proceeds. In order to enable the system to adapt to such dynamic changes and maintain an accurate description of thermal disturbance characteristics, the present application also provides a method for dynamically updating thermal disturbance characteristics based on key images. This method realizes real-time monitoring and updating of thermal disturbance characteristics by continuously acquiring images and analyzing the differences between adjacent images, further improving the adaptability and control accuracy of the system. The following is combined with Figure 2 , a method for dynamically updating thermal disturbance features based on a key image in an embodiment of the present application is described:

[0105] See also Figure 2 , which is a flow chart of a method for dynamically updating thermal disturbance features based on key images in an embodiment of the present application.

[0106] S201, between a first moment and a second moment, acquiring a molten plastic image at every preset sampling time interval to obtain a time sequence image group;

[0107] This step aims to obtain a set of time-series image data by continuously capturing images of the molten plastic, providing a data basis for the subsequent dynamic update of thermal disturbance characteristics. The first moment and the second moment here can be any two time points set according to actual needs, and their interval determines the time span of the entire dynamic update process. Between the first moment and the second moment, the system acquires images of the molten plastic at a preset sampling time interval, and the size of the sampling time interval determines the time resolution of the time-series image group. In addition to the fixed sampling time interval mentioned in the article, the system can also dynamically adjust the sampling frequency based on the real-time monitoring of the injection molding process status.

[0108] Specifically, the system can capture real-time visible light images of the molten plastic flow by installing a high-speed camera at the nozzle of the injection molding machine. To ensure image quality, the system can optimize the camera's optical parameters (such as exposure time, gain, etc.) and mechanical parameters (such as focal length, aperture, etc.) according to on-site environmental conditions. At the same time, the system can also improve the image's signal-to-noise ratio and contrast through image preprocessing algorithms (such as denoising and enhancement) to facilitate subsequent feature analysis. After obtaining a set of time-series images, the system arranges them in chronological order to form a time-series image group.

[0109] S202, calculating the pixel difference of the target area of ​​two adjacent molten plastic images in the time sequence image group;

[0110] This step quantifies the degree of change in image content by calculating the pixel difference between two adjacent images in the time-series image group, providing a basis for identifying key images. The target region here refers to the region of interest selected for analysis within the image, and its location and size can be flexibly adjusted based on actual needs. The system compares the grayscale value differences of corresponding pixels within the target region to generate a pixel difference matrix that reflects the degree of regional change. Based on this pixel difference matrix, the system can determine whether the change in image content is caused by normal molten plastic flow or abnormal thermal disturbances.

[0111] In specific implementations, the system can use a variety of metrics to calculate pixel differences, such as the absolute value of grayscale differences or the square of grayscale differences. To mitigate the effects of image noise, the system can also perform smoothing filtering on the pixel differences, improving the robustness of the difference calculation. Furthermore, given that thermal disturbances can be unevenly distributed within an image, the system can further divide the target area into multiple sub-regions and calculate the pixel differences for each sub-region separately, yielding a more detailed description of the changes.

[0112] S203: When it is determined that the pixel difference value of the target area is greater than a second preset threshold, the molten plastic image at the corresponding moment is used as a key image;

[0113] This step automatically identifies key images within the image sequence that have undergone significant changes by determining whether the pixel difference in the target area exceeds a preset threshold. The second preset threshold is an empirical value used to quantify the significance of image changes and can be adjusted based on specific application scenarios and requirements. When the pixel difference between the target area of ​​a molten plastic image at a given moment and its adjacent image exceeds this threshold, the system marks it as a key image, assuming it contains important thermal disturbance information and requires focused analysis and feature extraction.

[0114] In specific implementations, the system can control the sensitivity of key image recognition by setting a threshold parameter. The smaller the threshold, the more key images are recognized and the more subtle the thermal disturbance changes captured. Conversely, the larger the threshold, the fewer key images are recognized and the more significant the thermal disturbance changes captured. The system can dynamically adjust the threshold value based on actual needs to balance computational efficiency and recognition accuracy. Furthermore, considering that the characteristics of changes in different regions may vary, the system can also set different thresholds for different target sub-regions to achieve more refined key image recognition.

[0115] In actual applications, due to the complex and changeable flow state of molten plastic, a single pixel difference threshold judgment may not be able to accurately identify all key images. To improve the robustness of recognition, the system can comprehensively consider multiple image change characteristics and construct a composite judgment criterion. For example, the system can combine multiple indicators such as pixel difference, gradient change, and texture features, and form a more reliable key image discrimination model through weighted fusion or logical combination. At the same time, the system can also introduce machine learning algorithms to automatically optimize the parameters of the discrimination model through learning and training on a large amount of historical sample data, further improving the accuracy and adaptability of key image recognition.

[0116] S204: Update the short-term image distortion feature according to the key image.

[0117] The system updates the short-time image distortion features according to the key image, specifically including: performing pixel comparison between the key image and the molten plastic image at two adjacent moments to obtain a grayscale difference matrix; determining the continuous pixel area with the largest difference according to the grayscale difference matrix to obtain the coordinates of the distortion focus area; calculating the grayscale change vectors of adjacent pixel points within the coordinate range of the distortion focus area to obtain a distortion direction matrix; calculating the main distortion direction and distortion amplitude of the distortion focus area according to the distortion direction matrix to obtain updated distortion feature parameters; replacing the parameters of the corresponding area in the short-time image distortion features with the updated distortion feature parameters to obtain updated short-time image distortion features.

[0118] This step uses the identified key images to dynamically update and optimize the existing short-term image distortion features, ensuring they reflect changes in the molten plastic's flow state. By continuously incorporating the latest thermal disturbance information into the feature model, the system enables real-time monitoring and optimized control of the injection molding process, improving product quality and production efficiency. The short-term image distortion features here can be a set of parameters extracted in the previous step, which quantify attributes such as the type, direction, and magnitude of image distortion.

[0119] As described in the paper, the system first compares the key image with images of the molten plastic at two adjacent moments, generating a grayscale difference matrix. Based on this matrix, the system identifies a region of continuous pixels with the most significant grayscale changes, which serves as the region of interest for distortion. The system then calculates the grayscale change vectors of adjacent pixels within this region to generate a direction matrix reflecting the local distortion direction. By performing statistical analysis on this direction matrix, the system can determine key parameters such as the primary distortion direction and magnitude of the region of interest. These parameters are then used to update the original short-term image distortion features, enabling them to more accurately describe the current thermal disturbance state.

[0120] In its implementation, the system can choose different strategies to balance computational efficiency and update accuracy. For example, the system can accumulate distortion information from multiple key images and periodically batch-update short-term image distortion features. Alternatively, the system can immediately update the distortion features of the corresponding region upon identifying each key image, enabling real-time tracking. Furthermore, the system can introduce a forgetting factor to perform a weighted fusion of new and old distortion information, enabling the feature model to take into account both historical and current states, improving the continuity and stability of thermal disturbance prediction.

[0121] In the above embodiment, by acquiring images of molten plastic in a continuous time series to form a time-series image group, calculating the pixel difference of the target area of ​​adjacent images and setting a threshold to filter the key images, the moment when the state of the molten plastic changes significantly can be quickly located in a large amount of image data. This key image recognition method based on pixel difference can promptly capture abnormal changes in the flow process of molten plastic. By analyzing the key images and updating the short-term image distortion features, the system can dynamically track the impact of thermal disturbances on image quality and ensure the accuracy of the image processing results. This adaptive image feature update mechanism enables the system to have the ability to continuously learn and optimize, can adapt to various dynamic changes in the injection molding process, can better deal with various interference factors in actual production, and ensure the long-term stable operation of the system.

[0122] The following describes the system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the physical device structure of an injection molding machine control system based on a visual algorithm provided in an embodiment of the present application.

[0123] It should be noted that Figure 3 The structure of the system shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0124] like Figure 3As shown, the system includes a CPU 301, which can perform various appropriate actions and processes according to the programs stored in the ROM 302 or the programs loaded from the storage unit 308 into the RAM 303, such as executing the methods in the above-mentioned embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An I / O interface 305 is also connected to the bus 304.

[0125] The following components are connected to the I / O interface 305: an input section 306 including a camera, infrared sensor, and the like; an output section 307 including a liquid crystal display (LCD) and speakers; a storage section 308 including a hard disk and the like; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. Removable media 311, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is installed in the drive 310 as needed, so that computer programs read from the media can be installed in the storage section 308 as needed.

[0126] In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present invention includes a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 309 and / or installed from the removable medium 311. When the computer program is executed by the CPU 301, the various functions defined in the present invention are performed.

[0127] It should be noted that the computer-readable medium described in the embodiments of the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium may include a data signal transmitted in baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take any of a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two boxes shown in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0129] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the system described in the above embodiments, or may exist independently and not incorporated into the system. The storage medium carries one or more computer programs, and when executed by a processor of a system, the system implements the methods provided in the above embodiments.

[0130] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0131] As used in the above embodiments, the term “when…” may be interpreted as “if…” or “after…” or “in response to determining…” or “in response to detecting…”, depending on the context. Similarly, the phrases “upon determining…” or “if (stated condition or event) is detected” may be interpreted as “if determining…” or “in response to determining…” or “upon detecting (stated condition or event)” or “in response to detecting (stated condition or event)”, depending on the context.

[0132] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive).

[0133] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A control method for an injection molding machine based on a visual algorithm, characterized in that: include: Acquire an image of molten plastic at a first moment and an image of molten plastic at a second moment at a nozzle of an injection molding machine, wherein a time interval between the first moment and the second moment is less than a preset threshold; determining a pixel change amount of a target area in the molten plastic image at the first moment and the molten plastic image at the second moment; Determining the short-term image distortion characteristics caused by thermal disturbance based on the pixel change amount of the target area, specifically comprising: constructing a time series data sequence based on the pixel change amount of the target area; performing Fourier transform on the time series data sequence to obtain a frequency domain characteristic spectrum; determining a characteristic spectrum greater than a preset frequency threshold as a high-frequency characteristic spectrum; performing spatial distribution analysis on the high-frequency characteristic spectrum to obtain a distortion direction and a distortion amplitude; constructing a distortion feature vector based on the distortion direction and the distortion amplitude; and mapping the distortion feature vector to an image space to obtain the short-term image distortion characteristics caused by thermal disturbance; Separating the image distortion portion caused by thermal disturbance and the image change portion caused by molten plastic flow from the real-time molten plastic image based on the short-term image distortion feature, specifically comprising: using the short-term image distortion feature as a filter template, performing a convolution operation on the molten plastic image at the first moment and the molten plastic image at the second moment to obtain the image distortion portion caused by thermal disturbance; subtracting the molten plastic image at the first moment and the molten plastic image at the second moment from the image distortion portion to obtain the image change portion caused by molten plastic flow; Determining the thermal disturbance intensity at the nozzle of the injection molding machine according to the distribution law of the image distortion portion caused by the thermal disturbance; Calculating the actual flow state of the molten plastic according to the image change portion caused by the flow of the molten plastic and the thermal disturbance intensity; The process parameters of the injection molding machine are controlled according to the actual flow state.

2. The method according to claim 1, characterized in that The determining of the pixel change amount of the target area in the molten plastic image at the first moment and the molten plastic image at the second moment specifically includes: Dividing the molten plastic image at the first moment and the molten plastic image at the second moment into corresponding multiple grid units respectively; Calculating a pixel grayscale mean value for each grid cell in the molten plastic image at the first moment to obtain a first grayscale distribution matrix; calculating a pixel grayscale mean value for each grid cell in the molten plastic image at the second moment to obtain a second grayscale distribution matrix; Calculating grayscale differences between grid cells corresponding to the first grayscale distribution matrix and the second grayscale distribution matrix to obtain a grayscale difference matrix; performing feature point matching on corresponding grid cells in the molten plastic image at the first moment and the molten plastic image at the second moment to obtain a feature point matching result; Calculate the pixel position offset matrix of the corresponding grid unit according to the feature point matching result; Determining an area where the value of the grayscale difference matrix is ​​greater than a first preset threshold as a thermal disturbance affected area; Calculating the gradient value of the pixel position offset matrix within the thermal disturbance influence area; The pixel change amount in the target area is calculated according to the gradient value.

3. The method according to claim 1, characterized in that The calculating the actual flow state of the molten plastic according to the image change portion caused by the flow of the molten plastic and the thermal disturbance intensity specifically includes: Establishing a spatial distortion compensation coefficient based on the thermal disturbance intensity; constructing a three-dimensional distortion correction matrix according to the spatial distortion compensation coefficient; Applying the three-dimensional distortion correction matrix to the image change portion caused by the flow of the molten plastic to obtain a corrected flow image; extracting the flow boundary contour of the molten plastic in the corrected flow image; Calculating displacement vectors of the flow boundary contours at adjacent moments; Calculating the flow velocity field and flow direction field of the molten plastic according to the displacement vector; The actual flow state of the molten plastic is obtained based on the flow velocity field and the flow direction field.

4. The method according to claim 1, wherein The process parameters of the injection molding machine are controlled according to the actual flow state, specifically including: Calculating the shear rate distribution of the molten plastic according to the flow velocity field in the actual flow state; determining a shear stress distribution of the molten plastic based on the shear rate distribution; Calculating the apparent viscosity of the molten plastic according to the shear stress distribution and the shear rate distribution, and determining adjustment parameters of injection pressure and injection speed according to a comparison result of a change trend of the apparent viscosity with a preset viscosity curve; The injection molding pressure and injection molding speed of the injection molding machine are synchronously adjusted according to the adjustment parameters to obtain adjusted process parameters; and the injection molding process of the injection molding machine is controlled according to the adjusted process parameters.

5. The method according to claim 1, characterized in that After controlling the process parameters of the injection molding machine according to the actual flow state, the method further includes: Between the first moment and the second moment, acquiring a molten plastic image at every preset sampling time interval to obtain a time sequence image group; Calculating a pixel difference value of a target area of ​​two adjacent molten plastic images in the time-series image group; When it is determined that the pixel difference value of the target area is greater than a second preset threshold, the molten plastic image at the corresponding moment is used as a key image; The short-term image distortion feature is updated according to the key image.

6. The method according to claim 5, characterized in that The updating of the short-term image distortion feature according to the key image specifically includes: Comparing the key image with the molten plastic images at two adjacent moments to obtain a grayscale difference matrix; Determine the continuous pixel area with the largest difference according to the grayscale difference matrix, and obtain the coordinates of the distortion focus area; Calculating the grayscale change vectors of adjacent pixel points within the coordinate range of the distortion focus area to obtain a distortion direction matrix; Calculating the main distortion direction and distortion amplitude of the distortion focus area according to the distortion direction matrix to obtain updated distortion feature parameters; The updated distortion feature parameters are used to replace the parameters of the corresponding area in the short-time image distortion feature to obtain an updated short-time image distortion feature.

7. An injection molding machine control system based on a visual algorithm, characterized in that: The system comprises: One or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the system to execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are run on the system, The system is capable of executing the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is run on a system, the system is caused to perform the method according to any one of claims 1 to 6.

Images