Separator screen surface separation detection method and system based on machine vision

Abstract

The invention relates to the technical field of machine vision, in particular to a separator screen surface separation detection method and system based on machine vision, and the method comprises the steps: deploying an image collection assembly based on a screen surface illumination fitness model, and carrying out the texture enhancement processing of a screen surface image collected by the image collection assembly, and obtaining a texture enhancement factor; extracting particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the whole screen surface state according to the particle distribution density; generating a significance weight according to the global separation consistency index; and generating a sieve surface defect judgment result according to the significance weight reconstruction and a preset sieve surface defect judgment threshold. According to the method, a multi-scale texture enhancement and dynamic partitioning mechanism is introduced, the texture features of the screen surface are enhanced under the background of complex particle motion, fine identification and defect type intelligent classification are carried out on the screen surface separation abnormal region, and the accuracy and stability of anomaly detection are improved.

Description

Technical Field

[0001] The present application relates to the field of machine vision technology, and in particular to a method and system for detecting separation of a separator screen surface based on machine vision. Background Art

[0002] A separator is a device used to physically or mechanically separate the different components in a mixture. It serves as a key piece of equipment in the material screening process. Examples include common industrial vibrating screen separators, centrifugal separators, or airflow classifiers. Examples include multi-layer vibrating screens used in coal preparation plants, and high-speed centrifugal screens used to classify fine particles in ore processing.

[0003] The operating status of the screen surface of the separator directly affects the separation efficiency and product quality. Existing separation detection methods generally rely on vibration sensors or fixed cameras for image acquisition, and use traditional image processing algorithms or simple threshold methods to identify screen surface anomalies, such as blockage, overloading, or screen hole occlusion. However, such methods generally have problems such as insufficient detection accuracy, poor anti-interference ability, and lack of dynamic trend analysis. Especially when dealing with screen surface scenarios with high dynamics, high occlusion, and high particle motion interference, existing technologies are often unable to accurately identify local anomalies, and it is difficult to provide timely warnings on the evolution trend of anomalies. In addition, the distribution of material particles on the screen surface is complex, and existing technologies have limited use of explicit features such as particle morphology and distribution stability. They lack targeted feature modeling and multi-source feature fusion mechanisms, resulting in weak ability to distinguish between anomaly types and a high misjudgment rate. It is difficult to meet the needs of fine perception and predictive control of the screen surface operating status under intelligent manufacturing conditions. Summary of the Invention

[0004] Based on this, it is necessary to provide a method and system for separator screen surface separation detection based on machine vision, which introduces multi-scale texture enhancement and dynamic blocking mechanism to enhance the texture characteristics of the screen surface under the background of complex particle movement, effectively improve the particle boundary recognition ability, and perform fine identification of abnormal separation areas on the screen surface and intelligent classification of defect types, thereby improving the accuracy and stability of abnormality detection, effectively distinguishing different types of screen surface abnormalities such as blockage and overloading, and solving the problems of lack of detailed classification of abnormalities and high misjudgment rate in the existing technology.

[0005] The technical solutions of the present invention are as follows: A method for detecting separation of a separator screen surface based on machine vision, the method comprising: Deploying an image acquisition component based on a screen surface illumination fitness model, and performing texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; Extracting particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state based on the particle distribution density; generating a significance weight according to the global separation consistency index; A screen surface defect judgment result is generated according to the significance weight reconstruction and a preset screen surface defect judgment threshold.

[0006] Specifically, an image acquisition component is deployed based on the screen surface illumination fitness model, and texture enhancement processing is performed on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor, including: Construct a screen surface illumination adaptability model under illumination reflection conditions, and adjust the light source angle and intensity of the image acquisition component based on the model to deploy the image acquisition component; The screen surface image acquired by the image acquisition component is subjected to dynamic block division and multi-scale texture enhancement processing to obtain a texture enhancement factor.

[0007] Specifically, the particle distribution density is extracted according to the texture enhancement factor, and based on the particle distribution density, a global separation consistency index of the entire screen surface state is constructed, including: Setting a dynamic texture weight model according to the texture enhancement factor, constructing a quantifiable image index for the separation state, and extracting the particle distribution density; Based on the extracted particle distribution density, a global separation consistency index of the entire screen surface state is constructed.

[0008] Specifically, generating a significance weight according to the global separation consistency index includes: Constructing a local anomaly significance assessment model based on the global separation consistency index; The local anomaly significance evaluation model outputs a significance weight based on the local anomaly significance evaluation model.

[0009] Specifically, the method further includes: Regions in the sieve surface image where severe separation abnormalities exist are identified based on the significance weights.

[0010] Specifically, the screen surface defect judgment threshold includes a first abnormality significance threshold, a second abnormality significance threshold and a morphological stability judgment threshold; Generating a screen surface defect judgment result according to the significance weight reconstruction and a preset screen surface defect judgment threshold includes: Constructing a classification function according to the significance weight and the first abnormal significance threshold, the second abnormal significance threshold, and the morphological stability determination threshold; A screen surface defect judgment result is generated according to the classification function.

[0011] Specifically, a separator screen surface separation detection system based on machine vision, the system includes: A texture enhancement factor generation module is used to deploy an image acquisition component based on a screen surface illumination fitness model, and perform texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; a consistency index generating module, configured to extract the particle distribution density according to the texture enhancement factor, and construct a global separation consistency index of the entire screen surface state according to the particle distribution density; A significance weight generation module, configured to generate a significance weight according to the global separation consistency index; The defect judgment result generating module is used to generate the screen surface defect judgment result according to the significance weight reconstruction and the preset screen surface defect judgment threshold.

[0012] Specifically, the texture enhancement factor generation module is further used to: A screen surface illumination adaptability model under illumination reflection conditions is constructed, and the light source angle and intensity of the image acquisition component are adjusted according to the model to deploy the image acquisition component; the screen surface image captured by the image acquisition component is dynamically segmented and multi-scale texture enhanced to obtain a texture enhancement factor.

[0013] Specifically, the consistency index generation module is also used to: set a dynamic texture weight model according to the texture enhancement factor, construct a quantifiable image index for the separation state, and extract the particle distribution density; based on the extracted particle distribution density, construct a global separation consistency index for the entire screen surface state.

[0014] Specifically, the significance weight generation module is further used to: construct a local anomaly significance evaluation model according to the global separation consistency index; and output a significance weight based on the local anomaly significance evaluation model.

[0015] Specifically, the significance weight generating module is further configured to: identify areas with severe separation abnormalities in the sieve surface image based on the significance weights.

[0016] Specifically, the screen surface defect judgment threshold includes a first abnormality significance threshold, a second abnormality significance threshold and a morphological stability judgment threshold; the defect judgment result generation module is also used to: construct a classification function based on the significance weight and the first abnormality significance threshold, the second abnormality significance threshold, and the morphological stability judgment threshold; and generate a screen surface defect judgment result based on the classification function.

[0017] Optionally, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps described in the above-mentioned separator screen surface separation detection method based on machine vision are implemented.

[0018] Optionally, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the steps described in the above-mentioned separator screen surface separation detection method based on machine vision are implemented.

[0019] The present invention relates to machine learning and deep learning technologies, and the technical effects achieved are as follows: (1) The machine vision-based separator screen surface separation detection method and system described in this application deploys an image acquisition component based on a screen surface illumination fitness model, and performs texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor. This achieves the goal of adjusting the image acquisition conditions by constructing a light fitness model, ensuring the illumination balance and detail clarity of the screen surface image, and significantly improving the image quality and the robustness of subsequent analysis. (2) By extracting the particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state according to the particle distribution density; generating a significance weight according to the global separation consistency index; generating a screen surface defect judgment result according to the significance weight reconstruction and the preset screen surface defect judgment threshold, introducing a multi-scale texture enhancement and dynamic blocking mechanism, it is possible to enhance the screen surface texture features under the background of complex particle movement, effectively improve the particle boundary recognition ability, and construct a screen surface separation consistency index based on the extracted particle distribution density, breaking the traditional isolated judgment method of local anomalies and realizing a quantitative description of the global state of the screen surface; (3) This application also combines the local anomaly significance evaluation model with the classification function to perform refined identification of abnormal separation areas on the screen surface and intelligent classification of defect types. This not only improves the accuracy and stability of anomaly detection, but also effectively distinguishes different types of screen surface anomalies such as blockage and overloading, solving the problem of lack of detailed classification of anomalies and high misjudgment rate in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 1. A schematic flow chart of a method for detecting separation of a separator screen surface based on machine vision in one embodiment; Figure 2 1 is a structural block diagram of a separator screen surface separation detection system based on machine vision in one embodiment; Figure 3 FIG. 1 is a structural block diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0021] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0022] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0023] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0024] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0025] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0027] In one embodiment, a terminal is provided, which is used to: deploy an image acquisition component based on a screen surface illumination fitness model, and perform texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; extract the particle distribution density according to the texture enhancement factor, and construct a global separation consistency index of the entire screen surface state according to the particle distribution density; generate a significance weight according to the global separation consistency index; and generate a screen surface defect judgment result according to the significance weight reconstruction and a preset screen surface defect judgment threshold.

[0028] The terminal may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices.

[0029] In one embodiment, Figure 1 As shown, a method for detecting separation of a separator screen surface based on machine vision is provided, the method comprising: Step S100: deploying an image acquisition component based on a screen surface illumination fitness model, and performing texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; Step S200: extracting the particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state according to the particle distribution density; Step S300: generating a significance weight according to the global separation consistency index; Step S400: generating a screen surface defect judgment result according to the significance weight reconstruction and a preset screen surface defect judgment threshold.

[0030] This embodiment deploys an image acquisition component based on the screen surface illumination fitness model, and performs texture enhancement processing on the screen surface image collected by the image acquisition component to obtain a texture enhancement factor; extracts the particle distribution density based on the texture enhancement factor, and constructs a global separation consistency index of the entire screen surface state based on the particle distribution density; generates a significance weight based on the global separation consistency index; and generates a screen surface defect judgment result based on the significance weight reconstruction and a preset screen surface defect judgment threshold. This application introduces a multi-scale texture enhancement and dynamic blocking mechanism to enhance the texture features of the screen surface in the context of complex particle movement, and performs refined identification of abnormal screen surface separation areas and intelligent classification of defect types, thereby improving the accuracy and stability of anomaly detection.

[0031] In one embodiment, step S100: deploying an image acquisition component based on a screen surface illumination fitness model, and performing texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor, includes: Step S110: constructing a screen surface illumination adaptability model under illumination reflection conditions, and adjusting the light source angle and intensity of the image acquisition component according to the model to deploy the image acquisition component; Step S120: performing dynamic blocking and multi-scale texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor.

[0032] In this embodiment, to provide input image data for subsequent screen surface feature extraction, the core objective is not simply to capture images; it also requires parameterized modeling of illumination conditions, combining the screen surface material and the reflective properties of the material particles. Therefore, a model of the screen surface's illumination adaptability under these conditions is constructed, and the image acquisition component's light source angle and intensity are adjusted based on this model to facilitate deployment.

[0033] The screen surface illumination fitness model uses the average brightness of the screen surface area, the particle boundary clarity and the noise-background ratio to construct a screen surface illumination fitness index. The screen surface illumination fitness model is specifically as follows:

[0034] in, is the light adaptability of the screen surface. It is the average grayscale value of the screen surface image per unit area, which is calculated by statistically averaging the grayscale values ​​of pixels in the unit area and then using a grayscale image or converting a color image into a grayscale image. Used to reflect the overall brightness level of the area and measure the uniformity of lighting; This is the image edge clarity coefficient. It uses the Sobel or Canny operator to detect edges in an image and calculates the average intensity of edge responses within a unit area. This reflects the clarity of particle or sieve boundaries within the area and affects image separability. The signal-to-noise ratio is calculated by comparing the standard deviation of the background area with the grayscale value of the main image area. It indicates the degree to which non-target areas in the image interfere with the recognition effect. The more complex the background, the higher the value. It is a tiny constant set to prevent the denominator from being zero. It is pre-set by those skilled in the art to a fixed constant much smaller than 1, such as 0.001. It is only used to ensure the numerical stability of the model calculation and does not depend on the image content.

[0035] The image acquisition component will be based on the maximum The light source angle and intensity are adjusted according to the criterion, and the image state that is most conducive to subsequent image segmentation and feature extraction is selected to ensure that the quality of subsequent input images reaches the optimal state. The existing technology often ignores the reflection differences between screen surface materials such as metal and rubber and material particles such as coal powder and gravel at different lighting angles, resulting in blurred texture details in the image or difficulty in edge recognition, especially in scenes with strong reflections or high noise backgrounds. In contrast, in this embodiment, the screen surface illumination fitness model automatically performs a round of illumination scanning after the image acquisition component is deployed, that is, adjusts different light source angles and intensities, collects corresponding images, and calculates the illumination fitness index for each frame of the image in real time. ; Then, according to The size of determines whether the current configuration is optimal. If the value does not reach the preset threshold, adjust the parameters and continue collecting until After reaching the optimum or convergence stability, lock the current lighting conditions for formal sampling. The corresponding preset threshold can be set through experimental calibration, i.e., collecting multiple sets of images under typical sieve surface and particle conditions, and calculating the lower limit of illumination fitness that meets the requirements of clear edge recognition and low noise interference. In practical applications, it can be set to an empirical value between 15.0 and 20.0. The specific value is determined by those skilled in the art and is not specifically limited in this application.

[0036] The method disclosed in this embodiment can dynamically adjust parameters based on real-time lighting changes in complex dusty environments such as coal mines, sand and gravel plants, and metallurgical workshops, ensuring image acquisition stability. The system can optimize image quality in extremely reflective environments, such as metal screens, or in low-light conditions, such as those associated with nighttime operations, by adjusting the light source array directionality, fill light ratio, and exposure parameters of the acquisition device.

[0037] Furthermore, in order to solve the problem in the prior art that the use of a unified enhancement strategy for the entire image may easily lead to information loss in local areas of the image due to particle accumulation, occlusion, local reflection, etc. in a complex screening environment, and unified enhancement often leads to noise amplification or areas with blurred edges and cannot effectively improve image separability, therefore, in step S120, the screen surface image collected by the image acquisition component is dynamically segmented and multi-scale texture enhanced to obtain a texture enhancement factor.

[0038] Specifically, dynamic block processing dynamically divides the overall task into smaller "blocks" for independent processing based on data characteristics or runtime resource availability to improve efficiency, adapt to changes, or enhance robustness. In this embodiment, a weighted judgment is made based on the brightness change rate, particle distribution gradient, and grayscale contrast to accurately segment different areas.

[0039] In each image block , the corresponding texture enhancement factor as follows:

[0040] in, It is the screen surface illumination adaptability outputted in step S110 and is the stability weight at the whole image level.

[0041] It is a region The grayscale gradient amplitude average reflects the intensity of local texture changes. Apply the Sobel operator, calculate the gradient amplitude of each pixel and then average it. Used to indicate the intensity of texture changes. It is the skewness of the regional grayscale distribution, which measures whether the regional texture distribution is balanced. It counts the skewness of the grayscale values ​​of all pixels in the region and reflects the symmetry of the grayscale distribution. It is usually implemented by the third-order center distance formula. It is the texture enhancement factor, which indicates whether the region should be enhanced in texture extraction and is the basis for deciding whether the region should be enhanced.

[0042] By calculating each area ,like If the value is greater than the texture response threshold set by the system, the area is judged as a "weak texture area" and needs to be enhanced. The enhancement operation includes local contrast stretching, edge response weighting, and directional filter (such as Gabor) application. All enhancement operations are performed in The texture response threshold can be set according to the actual device resolution. Alternatively, the setting of the texture response threshold can be based on the statistical analysis of sample images, such as calculating the texture response threshold for a large number of screen surface images in different states. The threshold is selected based on the distribution of values. The minimum value that covers more than 80% of the areas with clear texture is used as the threshold to ensure that only areas with weak texture are enhanced. In practice, if the device resolution is 1080p and the grain size is medium, the threshold can be set to 1.2-1.5. For example, if it is set to 1.3, the system will only enhance areas below this value, ensuring that enhancement resources are concentrated on low-quality image blocks.

[0043] In addition, the use of this model can also enable the system to intelligently decide "enhanced budget allocation" based on the image status, that is, to concentrate limited computing resources on The area with high value is selected to improve the overall computing efficiency and image quality.

[0044] In summary, this step further expands the dimensions of screen surface image processing, focusing not only on global image quality but also on fine-grained structural analysis of local complexity. For example, when the screen surface is partially clumped, accumulated, or obscured by wet material, conventional enhancement processing often fails to effectively restore details. However, the method in step S120 can identify these low-texture active areas and perform weighted enhancement, making subsequent feature extraction and defect identification more accurate.

[0045] In one embodiment, step S200: extracting the particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state according to the particle distribution density, including: Step S210: setting a dynamic texture weight model according to the texture enhancement factor, constructing a quantifiable image index for the separation state, and extracting the particle distribution density; Step S220: constructing a global separation consistency index of the entire screen surface state based on the extracted particle distribution density.

[0046] In this embodiment, to address the problem that most existing methods for identifying screen surface status only perform rough statistics on screen surface particle distribution through simple pixel binarization or edge recognition, which cannot fully reflect complex working conditions such as local accumulation, hole blockage, or uneven loading, a dynamic texture weight model is set according to the texture enhancement factor to construct a quantifiable image indicator for the separation state and extract the particle distribution density. Based on the extracted particle distribution density, a global separation consistency indicator for the entire screen surface state is then constructed.

[0047] The dynamic texture weight model is as follows:

[0048] in, It is the texture enhancement factor, which reflects the image quality and information richness of the area. It is the number of particles identified in the area and can be output by the machine vision detection module, that is, the image acquisition component. It is the area of ​​the region, representing the unit area range, that is, the area range corresponding to an image block area. It indicates the actual particle distribution density in the area and is also used to indirectly estimate the sieve porosity of the area. The mapping relationship is converted into the proportion of the sieve holes blocked, thereby indirectly estimating the sieve hole porosity. Estimate the available void ratio of the sieve area and determine whether there are any abnormal screening conditions such as particle accumulation, pore blockage or material overloading in the area. After the image is divided into blocks, count the particles in each area. , and combined with the area of ​​the region and the texture enhancement factor from step S2, calculate The larger the value, the denser the particle coverage in the area, and the image quality is good enough, with high analysis reliability. Next, set a sieve visual saturation threshold. If If the threshold is exceeded, it is initially judged as a "blocked or overloaded" area, otherwise it is a normal screening area. This method realizes density weighted estimation based on image readability, which has higher stability and accuracy than the traditional method based on quantitative statistics alone. Among them, the sieve visual saturation threshold can be obtained by measuring the sieve under normal screening conditions. Perform statistics and take 90%-95% of the upper limit as the critical value, for example, set it to 0.045 particles / pixel 2 If the value exceeds this value, it is judged as a local blockage or overload area.

[0049] Therefore, this embodiment introduces a texture enhancement factor into particle density modeling, taking into account not only the number of particles but also regional image quality and legibility as weighted indicators, thereby constructing a dynamic texture weighting model. Furthermore, while traditional models often overlook the visual usability of the sieve apertures, this method, based on particle identification, further estimates the porosity of the regional sieve apertures—the degree to which the sieve surface is covered by particles. This porosity estimation provides a quantitative basis for subsequent screening efficiency evaluation and anomaly detection, possessing high practical value.

[0050] Next, in order to solve the problem that the existing screen surface detection methods mostly stay in the local area of ​​particle accumulation identification or simple average value statistics, and cannot systematically evaluate whether the entire screen surface operation status has global separation unevenness, material flow deviation or screen hole partition blockage, the regional particle density index output in the previous step is fused and analyzed across the entire screen surface, and the global consistency index is used to calculate the particle density of the entire screen surface. Quantify the stability and balance of material separation on the screen surface. By statistically analyzing the deviation of each area's density from the overall average density, this model not only detects blockage but also identifies early signs of screen imbalance, enabling more proactive fault warning and separation quality control. Compared to traditional methods that only measure total particle count or local maxima, this model is more robust and has greater applicability.

[0051] The global separation consistency index is used to determine whether there are problems such as uneven separation, local blockage, or uneven distribution. Specifically, the global separation consistency index is as follows:

[0052] in, is the particle distribution density. Yes all The average value is the particle distribution benchmark of the entire screen surface. is the number of regions into which the image is divided, indicating the number of regions into which the image is divided in the vertical direction (rows), N indicates the number of regions into which the image is divided in the horizontal direction (columns), and M·N is the total number of regions into which the image is divided. is the variance-weighted consistency evaluation index of particle distribution density, that is, the global separation consistency index, and The dimension of is the same. The global separation consistency index indirectly reflects the uniformity of the material distribution during the operation of the separator screen. If it is significantly greater than the threshold, it means there is a risk of regional particle accumulation or local pore blockage, and then the next stage is to perform abnormal significance modeling and screening defect classification. The corresponding threshold can be calculated by multiple normal screen surface sample images The value distribution is calculated, and the mean plus double the standard deviation is taken as the uneven separation warning threshold, which can be set to 0.006 particles / pixel. 2 If the value is exceeded, it is determined that there is obvious particle accumulation or unevenness on the screen surface.

[0053] In this embodiment, special consideration is given to the non-ideal conditions in industrial operation, for example, part of the screen surface may have a serious deviation in particle density due to material adhesion, screen hole wear or uneven feeding. The indicators can effectively detect whether abnormal areas have systematic deviations or developmental accumulation trends, achieving the transition from local analysis to global judgment.

[0054] In one embodiment, step S300: generating a significance weight according to the global separation consistency index includes: Step S310: constructing a local anomaly significance assessment model based on the global separation consistency index; Step S320: Outputting a significance weight based on the local anomaly significance assessment model.

[0055] Step S330: Identifying areas with severe separation abnormalities in the sieve surface image based on the significance weights.

[0056] In this embodiment, in order to solve the problem that traditional screen surface abnormality detection is mostly based on a single particle density or texture feature, it is difficult to take into account the separation deviation and image complexity of the abnormal area at the same time, resulting in a high false alarm rate and insufficient abnormality recognition ability under complex screen surface working conditions, a local abnormality significance evaluation model is constructed according to the global separation consistency index in this embodiment; and the significance weight is output based on the local abnormality significance evaluation model.

[0057] Specifically, the local anomaly significance evaluation model is as follows:

[0058] in, is the particle distribution density. yes The average value of . is a global separation consistency indicator. It is a very small number that prevents the denominator from being zero. It is an indicator of regional texture complexity, which is calculated by the change of local directional gradient. Specifically, it is obtained by counting the image gradient direction in the region, such as the Sobel direction angle, or the variance of the local structure tensor. The larger the value, the more complex the texture and the more uneven the structure, indicating the richness of image details and texture diversity. is the significance weight, which quantifies the confidence and risk level of the anomaly in the area. Classify each area of ​​the screen surface and extract abnormal areas with severe separation. First, calculate the relative deviation of the particle density of each area from the overall average density, and divide the result by the overall consistency index to obtain the standardized abnormal deviation coefficient. Then, multiply the deviation coefficient by the texture complexity index. , comprehensively consider the particle distribution anomaly and image texture characteristics, thereby improving the reliability of anomaly recognition. The risk of each area is graded based on the size of the risk factor, and high-risk abnormal areas are screened out.

[0059] Specifically, first count all regions The numerical distribution of risk is then divided into risk levels based on quantile thresholds, such as 85% and 95%. For example, areas with values ​​above the 95th percentile are defined as "high risk," those between 85% and 95% as "medium risk," and those below 85% as "low risk." Furthermore, the location of the areas on the screen surface, such as concentration at the discharge end or in the center, can be combined for further weighting and optimization, accurately screening out high-risk abnormal areas. These serve as key input for reconstructing the screen surface's operating status and subsequent automatic diagnosis.

[0060] Therefore, this embodiment, through steps S310-S320, can accurately locate abnormal areas on the screen surface and classify their risk levels. By introducing texture complexity weights, the model is not only sensitive to deviations in particle distribution but also dynamically adjusts the confidence level of abnormalities based on image details, significantly improving the resistance to interference and accuracy of abnormality screening, greatly enhancing the practical value of abnormality diagnosis.

[0061] In one embodiment, the screen surface defect judgment threshold includes a first abnormality significance threshold, a second abnormality significance threshold and a morphological stability judgment threshold; Step S400: generating a screen surface defect judgment result according to the significance weight reconstruction and a preset screen surface defect judgment threshold, including: Step S410: constructing a classification function according to the significance weight and the first abnormal significance threshold, the second abnormal significance threshold, and the morphological stability determination threshold; Step S420: generating a screen surface defect judgment result according to the classification function.

[0062] In this embodiment, in order to solve the problem that in the prior art, screen surface defects mostly adopt simple threshold judgment or empirical rule classification, lack in-depth analysis of the morphological characteristics of abnormal areas, resulting in inaccurate defect type judgment and poor adaptability, this embodiment constructs a classification function based on the significance weight and the first abnormal significance threshold, the second abnormal significance threshold, and the morphological stability judgment threshold; and then generates a screen surface defect judgment result based on the classification function.

[0063] The classification function uses a defect classification model based on block association cluster density, which is as follows:

[0064] in, is the significance weight, reflecting the abnormality degree of the region. is the regional particle morphology stability. The first abnormal significance threshold and the second anomaly significance threshold , Set as abnormal area 80% quantile or above, The 95% quantile of the normal area of ​​all samples is usually taken. This is the threshold for determining morphological stability. The coefficient of variation (CV) of particle outlines across all image regions, such as the ratio of the standard deviation of area and perimeter to the mean, is calculated. The median or 70th percentile of this value, calculated under normal screening conditions, is used as the threshold. Values ​​below this threshold indicate relatively consistent particle morphology and high morphological stability. is the final defect classification label.

[0065] The classification function determines the defect type based on the regional abnormal significance weight and particle morphology stability. In the defect classification model, the regional particle morphology stability The coefficient of variation calculation based on the particle contour reflects the regularity and consistency of the particle morphology. Specifically, it can be obtained by calculating the coefficient of variation of the contour area of ​​all identified particles in the region, that is, dividing the standard deviation of the particle area by its average value. The smaller the value, the more consistent the particle morphology and the more regular the distribution.

[0066] Anomaly significance threshold middle, Used to distinguish the severity of anomalies; final defect classification label In the figure, 1 indicates accumulation blockage, 2 indicates eccentric load dispersion, and 3 indicates normal area.

[0067] This classification function determines defect types based on both the significance weight of the regional anomaly and the particle morphological stability. Regions with high anomaly weights and low morphological stability are classified as accumulation and blockage (Type 1), indicating a backlog of material with irregular morphology. Regions with high anomaly weights and high morphological stability are classified as eccentric load dispersion (Type 2), reflecting material displacement but relatively regular particle distribution. Regions with low anomaly weights are classified as normal (Type 3). This dual-metric approach effectively avoids misjudgments caused by a single metric, enhancing the accuracy and stability of defect identification.

[0068] In this embodiment, the method is also provided with a continuous time frame analysis mechanism, which incorporates the change process of the screen surface image within a time window into the analysis logic, and is used to judge the dynamic stability and trend evolution characteristics of the screening state, so as to achieve early warning of "potential fault precursors".

[0069] The continuous time-series frame analysis mechanism does not determine the degree of abnormality in a single frame of image, but rather determines whether the "evolution trend" of the abnormality is continuously amplified. To this end, in a continuous time period, the system records and analyzes the distribution of significant areas, the fluctuation in the number of defect labels, and the spatial migration trajectory of the center point of the abnormal area. Specific methods include: First, trend variation judgment: by recording the changes in abnormal labels in the same area in consecutive images, if the abnormal labels continue to increase, it is considered as potential unstable evolution.

[0070] Secondly, spatial drift feature extraction: track the movement trajectory of the center point of each abnormal area over time to determine whether there is a continuous trend of concentration toward the edge or middle of the screen surface.

[0071] Next, the dynamic intervention mechanism recommends output: Based on the dynamic stability results, the system makes adjustment suggestions, such as reducing the feed rate, temporarily pausing the vibration module, checking the response of the eccentric load sensor, and other operational suggestions.

[0072] Compared with traditional image analysis systems that mostly focus on static judgment, this application has the ability to identify and respond to temporal trends; it can achieve early detection of minor but continuously aggravating screen surface faults, such as the continuous spread of small blockages and the evolution of large-scale accumulation; it constructs a closed-loop logic path between image recognition results and control instructions, which is conducive to the subsequent formation of an intelligent control system.

[0073] Therefore, this embodiment achieves intelligent identification and classification of various screen surface defect types through inter-regional correlation clustering. This not only improves the accuracy of defect classification, but also effectively distinguishes different fault modes such as accumulation blockage and uneven load dispersion. This provides a more targeted basis for automatic adjustment and maintenance decision-making during the screening process, significantly enhancing the system's intelligence and practical value.

[0074] In general, the separator screen surface separation detection method and system based on machine vision described in the present application, by deploying an image acquisition component based on the screen surface illumination fitness model, and performing texture enhancement processing on the screen surface image acquired by the image acquisition component, obtains a texture enhancement factor, and adjusts the image acquisition conditions by constructing an illumination fitness model to ensure the illumination balance and detail clarity of the screen surface image, thereby significantly improving the image quality and the robustness of subsequent analysis; by extracting the particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state according to the particle distribution density; generating a significance weight according to the global separation consistency index; and reconstructing and presetting the screen surface defect according to the significance weight. The screen surface defect judgment result is generated by the trap judgment threshold, and the multi-scale texture enhancement and dynamic blocking mechanism are introduced, which can enhance the texture characteristics of the screen surface under the background of complex particle movement, effectively improve the particle boundary recognition ability, and construct the screen surface separation consistency index based on the extracted particle distribution density, breaking the traditional isolated judgment method of local anomalies and realizing the quantitative description of the global state of the screen surface; in addition, the application also combines the local anomaly significance evaluation model and classification function to perform fine identification of the screen surface separation abnormal area and intelligent classification of defect types, which not only improves the accuracy and stability of anomaly detection, but also can effectively distinguish different screen surface abnormality types such as blockage and overloading, solving the problems of lack of detailed classification of anomalies and high misjudgment rate in the existing technology.

[0075] In one embodiment, Figure 2 As shown, a separator screen surface separation detection system based on machine vision is also provided, and the system includes: A texture enhancement factor generation module is used to deploy an image acquisition component based on a screen surface illumination fitness model, and perform texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; a consistency index generating module, configured to extract the particle distribution density according to the texture enhancement factor, and construct a global separation consistency index of the entire screen surface state according to the particle distribution density; A significance weight generation module, configured to generate a significance weight according to the global separation consistency index; The defect judgment result generating module is used to generate the screen surface defect judgment result according to the significance weight reconstruction and the preset screen surface defect judgment threshold.

[0076] In one embodiment, the texture enhancement factor generation module is further configured to: A screen surface illumination adaptability model under illumination reflection conditions is constructed, and the light source angle and intensity of the image acquisition component are adjusted according to the model to deploy the image acquisition component; the screen surface image captured by the image acquisition component is dynamically segmented and multi-scale texture enhanced to obtain a texture enhancement factor.

[0077] In another embodiment, the consistency index generation module is also used to: set a dynamic texture weight model according to the texture enhancement factor, construct a quantifiable image index for the separation state, and extract the particle distribution density; based on the extracted particle distribution density, construct a global separation consistency index for the entire screen surface state.

[0078] In another embodiment, the significance weight generation module is further configured to: construct a local anomaly significance evaluation model according to the global separation consistency index; and output a significance weight based on the local anomaly significance evaluation model.

[0079] In another embodiment, the significance weight generation module is further configured to: identify regions with severe separation abnormalities in the screening surface image based on the significance weights.

[0080] In another embodiment, the screen surface defect judgment threshold includes a first abnormality significance threshold, a second abnormality significance threshold and a morphological stability judgment threshold; the defect judgment result generation module is also used to: construct a classification function based on the significance weight and the first abnormality significance threshold, the second abnormality significance threshold, and the morphological stability judgment threshold; and generate a screen surface defect judgment result based on the classification function.

[0081] In one embodiment, Figure 3 A computer device is also provided, including a memory and a processor. The memory stores a computer program and an operating system, and the processor executes the computer program to implement the steps of the above-mentioned machine vision-based separator screen surface separation detection method. The computer device also includes a system bus, internal memory, a network structure, a display screen, and an input device.

[0082] In one embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned separator screen surface separation detection method based on machine vision are implemented.

[0083] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0084] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0085] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0086] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0087] An embodiment of the present application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the computer program.

[0088] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0089] An embodiment of the present application provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned various method embodiments when executing the computer program product.

[0090] If the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a camera / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electric carrier signals or telecommunication signals.

[0091] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0092] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0093] In the embodiments provided in this application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0094] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0095] The above-described 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

[0096] An embodiment of the present application also provides a computer device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps in any embodiment of the above method when executing the computer program.

[0097] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above description is an example of a computer device and does not limit the computer device. The computer device may include more or fewer components than described above, or a combination of certain components, or different components. For example, the computer device may also include input / output devices, network access devices, etc.

[0098] The processor may be a central processing unit (CPU), and the processor 0 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0099] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped with the computer device. Furthermore, the memory may include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been output or is about to be output.

[0100] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and such modifications and improvements are intended to fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for detecting separation of a separator screen surface based on machine vision, characterized in that: The method comprises: Deploying an image acquisition component based on a screen surface illumination fitness model, and performing texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; Extracting particle distribution density according to the texture enhancement factor, and constructing a global separation consistency index of the entire screen surface state based on the particle distribution density; generating a significance weight according to the global separation consistency index; A screen surface defect judgment result is generated according to the significance weight reconstruction and a preset screen surface defect judgment threshold.

2. The method for detecting separation of a separator screen surface based on machine vision according to claim 1, characterized in that: An image acquisition component is deployed based on the screen surface illumination fitness model, and texture enhancement processing is performed on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor, including: Construct a screen surface illumination adaptability model under illumination reflection conditions, and adjust the light source angle and intensity of the image acquisition component based on the model to deploy the image acquisition component; The screen surface image acquired by the image acquisition component is subjected to dynamic block division and multi-scale texture enhancement processing to obtain a texture enhancement factor.

3. The method for detecting separation of a separator screen surface based on machine vision according to claim 1, characterized in that: The particle distribution density is extracted according to the texture enhancement factor, and a global separation consistency index of the entire screen surface state is constructed based on the particle distribution density, including: Setting a dynamic texture weight model according to the texture enhancement factor, constructing a quantifiable image index for the separation state, and extracting the particle distribution density; Based on the extracted particle distribution density, a global separation consistency index of the entire screen surface state is constructed.

4. The method for detecting separation of a separator screen surface based on machine vision according to claim 1, characterized in that: Generating a significance weight according to the global separation consistency index includes: Constructing a local anomaly significance assessment model based on the global separation consistency index; The local anomaly significance evaluation model outputs a significance weight based on the local anomaly significance evaluation model.

5. The method for detecting separation of a separator screen surface based on machine vision according to claim 1, characterized in that: The method further comprises: Regions in the sieve surface image where severe separation abnormalities exist are identified based on the significance weights.

6. The method for detecting separation of a separator screen surface based on machine vision according to claim 1, characterized in that: The screen surface defect judgment threshold includes a first abnormality significance threshold, a second abnormality significance threshold and a morphological stability judgment threshold; Generating a screen surface defect judgment result according to the significance weight reconstruction and a preset screen surface defect judgment threshold includes: Constructing a classification function according to the significance weight and the first abnormal significance threshold, the second abnormal significance threshold, and the morphological stability determination threshold; A screen surface defect judgment result is generated according to the classification function.

7. A separator screen surface separation detection system based on machine vision, characterized in that: The system comprises: A texture enhancement factor generation module is used to deploy an image acquisition component based on a screen surface illumination fitness model, and perform texture enhancement processing on the screen surface image acquired by the image acquisition component to obtain a texture enhancement factor; a consistency index generating module, configured to extract the particle distribution density according to the texture enhancement factor, and construct a global separation consistency index of the entire screen surface state according to the particle distribution density; A significance weight generation module, configured to generate a significance weight according to the global separation consistency index; The defect judgment result generating module is used to generate the screen surface defect judgment result according to the significance weight reconstruction and the preset screen surface defect judgment threshold.

8. The machine vision-based separator screen surface separation detection system according to claim 7, characterized in that: The texture enhancement factor generation module is further used for: A screen surface illumination adaptability model under illumination reflection conditions is constructed, and the light source angle and intensity of the image acquisition component are adjusted according to the model to deploy the image acquisition component; the screen surface image captured by the image acquisition component is dynamically segmented and multi-scale texture enhanced to obtain a texture enhancement factor.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images