Mineral aggregate crushing degree detection method based on image analysis

Abstract

The invention relates to the field of mineral aggregate detection, in particular to a mineral aggregate crushing degree detection method based on image analysis. The method comprises the following steps: acquiring a preprocessed mineral aggregate grey-scale map, calculating a gradient value of any pixel point in the mineral aggregate grey-scale map, and calculating the division difficulty of the mineral aggregate grey-scale map; obtaining edges in the mineral aggregate grey-scale map based on edge detection, constructing a minimum enclosing rectangle of a target edge, and counting the number of reference edges in the minimum enclosing rectangle to calculate the outer contour probability of the target edge; traversing to obtain outer contour probabilities of all edges in the mineral aggregate grey-scale map, determining an outer contour of the mineral aggregate, converting the mineral aggregate grey-scale map into a binary image according to the outer contour of the mineral aggregate, inputting the binary image into a preset neural network, obtaining a crushing result probability value, and completing crushing degree detection. According to the technical scheme, the precision and efficiency of the detection result of the crushing degree of the mineral aggregate can be improved.

Description

Technical Field

[0001] The present invention relates to the field of mineral material detection, and more specifically, to a method for detecting the degree of mineral material crushing based on image analysis. Background Art

[0002] Mineral crushing is a key link in mineral processing, especially in ore mining, mineral processing, building materials production and other industries. The degree of crushing directly affects the subsequent processing efficiency and product quality. The degree of crushing usually refers to the particle size distribution of the mineral during the crushing process, that is, the proportion of particles of various sizes crushed into particles. Particle size control is one of the core factors for optimizing the production process, improving energy efficiency and reducing costs. How to accurately and efficiently detect and monitor the degree of crushing of minerals has become an important issue in the industry.

[0003] With the improvement of industrial automation and intelligence, traditional manual detection methods can no longer meet the needs of high-speed production lines and high-precision detection. Traditional crushing degree detection mostly relies on manual sampling and screening methods, which is not only labor-intensive but also has large errors. In addition, due to the complexity and instability of the crushing process, delays and misjudgments often occur in the detection work, and the crushing status cannot be effectively fed back in real time, further affecting production scheduling and equipment maintenance.

[0004] The existing Chinese patent application document with publication number CN116823827A discloses a method for evaluating ore crushing effect based on image processing. The application document proposes to obtain an ore limescale image, use particle detection operators of different scales to generate an ore shape image, determine the particle offset distance and iteratively process the central particle, perform image segmentation, generate an ore particle image, extract the ore area and evaluate the crushing effect, and finally generate an evaluation result.

[0005] However, mineral materials usually present multiple edges and irregular shapes in images, which may be caused by different shapes of mineral particles, surface roughness or changes in lighting conditions. Multiple edges increase the complexity of image segmentation, making it impossible for image segmentation algorithms to accurately distinguish the actual boundaries of mineral materials, resulting in inaccurate detection results of crushing effects. Summary of the invention

[0006] In order to solve the problem of inaccurate detection results of crushing effect, the present invention proposes a method for detecting the degree of crushing of mineral materials based on image analysis.

[0007] The present invention discloses a method for detecting the degree of crushing of mineral materials based on image analysis, comprising: obtaining a preprocessed mineral material grayscale image, calculating the gradient value of any pixel point in the mineral material grayscale image, and calculating the difficulty of dividing the mineral material grayscale image; obtaining the edge in the mineral material grayscale image based on edge detection, taking any edge as a target edge, taking edges other than the target edge as reference edges, constructing a minimum circumscribed rectangle of the target edge, counting the number of reference edges in the minimum circumscribed rectangle, so as to calculate the outer contour probability of the target edge, and the outer contour probability satisfies the relationship: , represents the outer contour probability, Represents the mean value of the gradient values ​​of all pixels on the target edge. Represents the variance of the gradient values ​​of all pixels on the target edge, Indicates the number of pixels on the edge of the target. represents the number of reference edges, represents the adjustment factor, Indicates the difficulty of division, Represents a normalization function; traverses to obtain the outer contour probability of all edges in the mineral material grayscale image, determines the outer contour of the mineral material, converts the mineral material grayscale image into a binary image according to the outer contour of the mineral material, inputs the binary image into a preset neural network, obtains the probability value of the crushing result, and completes the crushing degree detection.

[0008] By calculating the probability of the outer contour of the target edge and comparing it with the distribution of the reference edge, the outer contour of the mineral material can be distinguished from the complex internal edge, thereby reducing the error caused by the irregular shape of the mineral block or the perspective effect. Converting the outer contour information into a binary image and inputting it into the neural network can further improve the accuracy of the crushing degree detection and ensure that more reliable crushing results can be obtained during the mineral processing process.

[0009] Preferably, the preprocessing includes: using Gaussian filtering to denoise the mineral material grayscale image, and using adaptive histogram equalization to increase the contrast of the mineral material grayscale image.

[0010] Preferably, the division difficulty includes: constructing a gradient line graph with the gradient value as the horizontal coordinate and the number of pixel points as the vertical coordinate, and obtaining the maximum number of pixel points in the gradient line graph.

[0011] , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis.

[0012] Preferably, the division difficulty further includes: constructing a gradient line graph with the gradient value as the horizontal coordinate and the number of pixel points as the vertical coordinate, and obtaining the maximum number of pixel points in the gradient line graph.

[0013] , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis, Indicates the value of the horizontal coordinate corresponding to the maximum number of pixels. Indicates The value of the horizontal axis.

[0014] By comparing the ratio of the number of pixels corresponding to each gradient value to the maximum number of pixels in the image, as well as the degree of proximity between each gradient value and the maximum gradient value, the distribution of gradient changes in the image can be effectively evaluated. This calculation method not only reveals the pixel distribution characteristics of the image at different gradient values, but also reflects the richness of image details and the distribution law of edge strength.

[0015] Preferably, determining the outer contour of the mineral material comprises: clustering all edges to obtain a number of clusters, calculating the outer contour probability mean of all edges in any cluster, and taking the edge in the cluster with the largest outer contour probability mean as the outer contour of the mineral material.

[0016] It can not only reduce the misidentification caused by noise or interference from non-mineral areas, but also enhance the stability and accuracy of contour extraction through the probability mean method, ensuring that the selected edge best reflects the actual shape of the mineral.

[0017] Preferably, the preset neural network is a convolutional neural network, which extracts features from binary images to obtain image features, and classifies the image features to obtain probability values ​​of crushing results.

[0018] Preferably, it also includes: using the edge filling method to obtain the internal area of ​​the outer contour of the mineral material, and calculating the internal area correction value; traversing to obtain the internal area correction value of each outer contour of the mineral material, and calculating the mean internal area correction value of the mineral material grayscale image; using the mineral material grayscale image whose crushing degree meets the standard as the standard image, and using the mean internal area correction value of the standard image as the standard mean; in response to the mean internal area correction value being less than the standard mean, the crushing degree of the mineral material grayscale image is qualified.

[0019] By correcting the internal area of ​​the outer contour of the mineral material through the edge filling method, the error caused by the shape irregularity or image noise during the mineral material crushing process can be effectively eliminated, ensuring the accurate assessment of the degree of mineral material crushing. When the mean value of the internal area correction value of the mineral material grayscale image is less than the standard mean value, it indicates that the mineral material has achieved the expected crushing effect.

[0020] Preferably, the internal area correction value satisfies the relationship: , represents the internal area correction value, represents the internal area, It represents the Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image. Represents half of the diagonal of the mineral grayscale image.

[0021] Preferably, the Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image includes: constructing a position coordinate system for the mineral material grayscale image; respectively calculating the horizontal coordinate mean and the vertical coordinate mean of all pixel points in the outer contour of the mineral material as the position coordinates of the outer contour of the mineral material, and calculating the Euclidean distance between the position coordinates and the center point of the mineral material grayscale image.

[0022] It can reflect the overall distribution position and shape characteristics of the mineral material, and the Euclidean distance from the center point of the mineral material grayscale image can obtain the position offset of the mineral material in the image.

[0023] Beneficial effects of the present invention: The present invention can effectively detect the degree of crushing of mineral materials by comprehensively applying image processing technology and neural network models. First, by preprocessing and gradient analysis of the grayscale image of the mineral material, combining edge detection with the construction of the minimum circumscribed rectangle, the outer contour of the mineral material is accurately identified, and then the outer contour probability of each target edge is calculated to reflect the degree of fineness of the mineral material crushing. The partition difficulty assessment introduced by this method helps to quantify the complexity of the grayscale image of the mineral material, thereby providing a more accurate analysis of the crushing effect. In addition, by extracting and classifying the binary image through a convolutional neural network, combined with the analysis of the geometric characteristics of the outer contour of the mineral material and the internal area correction value, the uniformity and quality of the mineral material crushing can be more comprehensively evaluated, ensuring the high accuracy and reliability of the detection results. In practical applications, it can provide an efficient and accurate method for the automated monitoring of the ore crushing process, improve the quality control level of mineral material processing, help reduce energy waste and production costs, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 The present invention is a flowchart of a method for detecting the degree of crushing of mineral materials based on image analysis.

[0025] Figure 2 It is a grayscale image of a mineral material according to a method for detecting the degree of mineral material crushing based on image analysis in an embodiment of the present invention.

[0026] Figure 3 It is a gradient line graph of a method for detecting the degree of mineral material crushing based on image analysis in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

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

[0029] The present invention provides a method for detecting the degree of crushing of mineral materials based on image analysis. Figure 1 As shown, a method for detecting the degree of mineral material crushing based on image analysis includes steps S1 to S3, which are described in detail below.

[0030] S1, obtaining a preprocessed mineral material grayscale image, calculating the gradient value of any pixel point in the mineral material grayscale image, and calculating the difficulty of dividing the mineral material grayscale image.

[0031] In one embodiment, an industrial camera is used to capture an RGB image of the ore after the ore is crushed, and then the image is converted into an initial ore grayscale image through grayscale processing. Next, in order to remove noise in the image, a Gaussian filter is used to smooth the initial grayscale image, thereby reducing the impact of noise on subsequent processing. Subsequently, in order to enhance the contrast of the image, an adaptive histogram equalization technique is used to improve the detail performance of the initial ore grayscale image, especially in areas where the grayscale value distribution is relatively concentrated or the contrast is low, thereby obtaining a preprocessed ore grayscale image, such as Figure 2 The figure shows the grayscale image of the mineral material after preprocessing.

[0032] The gradient value of each pixel in the mineral grayscale image is obtained through the Sobel operator, so as to calculate the difficulty of dividing the mineral grayscale image. The division difficulty includes: constructing a gradient line graph with the gradient value as the horizontal coordinate and the number of pixel points as the vertical coordinate, and obtaining the maximum number of pixel points in the gradient line graph.

[0033] like Figure 3 As shown in FIG. 1 , a gradient line graph is constructed, in which the vertical axis pixel frequency, that is, the number of times the pixel appears, is equal to the number of pixels.

[0034] The partition difficulty satisfies the relationship: , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis.

[0035] The number of pixels corresponding to each gradient value in the mineral grayscale image is counted, for example, Figure 3 , the number of pixels corresponding to the gradient value of 23 is 11000, and the number of pixels corresponding to the gradient value of 254 is the largest, which is 42000. Therefore, the maximum number of pixels is 42000 corresponding to the gradient value of 254.

[0036] It can reflect the distribution of pixel gradient values. The larger the difference, the more concentrated the gradient values ​​of most pixels are. The image contains less texture and edge information, and the more difficult it is to divide the outer contour.

[0037] In one embodiment, dividing the difficulty further includes: constructing a gradient line graph with the gradient value as the horizontal coordinate and the number of pixel points as the vertical coordinate, and obtaining the maximum number of pixel points in the gradient line graph.

[0038] The partition difficulty satisfies the relationship: , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis, Indicates the value of the horizontal coordinate corresponding to the maximum number of pixels. Indicates The value of the horizontal axis.

[0039] Introduction , when the gradient values ​​of two pixels are and hour, The larger the value, the easier it is to distinguish the two pixels.

[0040] By comparing the ratio of the number of pixels corresponding to each gradient value to the maximum number of pixels in the image, as well as the proximity of each gradient value to the maximum gradient value, the distribution of gradient changes in the image can be effectively evaluated. This calculation method not only reveals the pixel distribution characteristics of the image at different gradient values, but also reflects the richness of image details and the distribution law of edge strength. If the number of pixels corresponding to certain gradient values ​​is large, it indicates that the image has more obvious edge or texture features in these areas; if the gradient value distribution is more uniform or is far from the maximum gradient value, it may mean that the image is relatively smooth or lacks obvious structural changes.

[0041] S2, based on edge detection, obtains the edge in the mineral grayscale image, constructs the minimum bounding rectangle of the target edge, and counts the number of reference edges in the minimum bounding rectangle to calculate the outer contour probability of the target edge.

[0042] It should be noted that, as the shape and size of the particles change during the crushing process, the grayscale image of the material will show different edge features, which reflect the degree of crushing of the material. The grayscale image of the material not only contains the outer contour of the material, but also may contain internal edges caused by surface cracks, contact surfaces between particles, and different material areas. These internal edges will interfere with the accurate detection of the degree of crushing.

[0043] In one embodiment, edges in a mineral grayscale image are obtained through edge detection, any edge is taken as a target edge, edges other than the target edge are taken as reference edges, the number of pixels in the target edge is counted, and the minimum enclosing rectangle of the target edge is constructed to obtain the number of reference edges in the minimum enclosing rectangle.

[0044] Calculate the outer contour probability, the outer contour probability satisfies the relationship: , represents the outer contour probability, Represents the mean value of the gradient values ​​of all pixels on the target edge. Represents the variance of the gradient values ​​of all pixels on the target edge, Indicates the number of pixels on the edge of the target. represents the number of reference edges, represents the adjustment factor, Indicates the difficulty of division, Represents a normalization function. The purpose of the adjustment factor is to prevent the result from being 0, because there is a situation where the target edge is an outer contour, but there is no reference contour in the minimum enclosing rectangle of the target edge.

[0045] It should be noted that the outer contour of the mineral material is usually more obvious and stronger than the inner edge, because the outer contour is often affected by the shape and size of the mineral material and shows a higher gradient value; while the inner edge of the mineral material, such as the contact surface between cracks or particles, usually has a lower gradient value and is more blurred. Therefore, by calculating the average value of the gradient values ​​of all pixels corresponding to the edge, the outer contour and the inner edge can be effectively distinguished. When the average gradient value of the edge is large, it indicates that the gradient value of the pixel corresponding to the edge is strong and is more likely to be the outer contour of the mineral material; conversely, if the average gradient value is small, it means that the gradient value of the edge is weak, which may be the performance of the inner edge of the mineral material.

[0046] The outer contour of the mineral material usually manifests as a significant grayscale jump from the mineral material to the background or adjacent mineral material. Therefore, the gradient value in this area changes relatively smoothly and strongly. The stability of this grayscale change makes the gradient values ​​of the corresponding pixel points of the outer contour of the mineral material usually consistent, showing a higher gradient value.

[0047] The edges inside the mineral material, especially the texture changes caused by factors such as surface irregularities, cracks or contact between particles, show more complex and irregular edge features. The gradient changes of these internal edges are more complex and show certain fluctuations in the image, resulting in large differences between the gradient values ​​of corresponding pixels.

[0048] The outer contour of the mineral material may contain internal edges such as cracks or textures, and these internal edges generally have a smaller span and will be included in the minimum circumscribed rectangle of the outer contour.

[0049] The calculation of the outer contour probability is adjusted by the difficulty of division. The greater the difficulty of division, the smaller the negative correlation mapping result of the exponential function will be, and the adjustment of the outer contour probability will become larger to widen the gap between the outer contour and the internal edge of the mineral material.

[0050] S3, traverse to obtain the outer contour probability of all edges in the mineral material grayscale image, determine the outer contour of the mineral material, convert the mineral material grayscale image into a binary image according to the outer contour of the mineral material, input the binary image into a preset neural network, obtain the probability value of the crushing result, and complete the crushing degree detection.

[0051] In one embodiment, determining the outer contour of the mineral material includes: clustering all edges to obtain a number of clusters, calculating the outer contour probability mean of all edges in any cluster, and taking the edge in the cluster with the largest outer contour probability mean as the outer contour of the mineral material.

[0052] According to the outer contour of the mineral material, the grayscale image of the mineral material is converted into a binary image, wherein the pixel points of the outer contour of the mineral material are marked as 1, and the pixel points of the non-outer contour are marked as 0. The binary image is input into a preset neural network, which is a convolutional neural network. The convolutional neural network extracts features from the binary image to obtain image features, and classifies the image features to obtain the probability value of the crushing result, wherein the probability of a normal crushing result and the probability of an abnormal crushing result are calculated for the image features, that is, they are divided into a normal crushing category and an abnormal crushing category, and the probability of a normal crushing result and the probability of an abnormal crushing result are compared to determine the probability value of the crushing result of the mineral material grayscale image, and complete the crushing degree detection.

[0053] Among them, the training process of the preset neural network includes: taking the binary image in the history as input information, and taking the true value of the probability value of the crushing result of the binary image as a label, to obtain a set of training data, and inputting the training data into the preset neural network to obtain the output result; calculating the loss value of the preset neural network through the output result and the label, and the loss function uses the cross entropy loss function, back-propagating the error signal according to the loss value, updating the network parameters of the preset neural network to make the loss value smaller; iteratively updating the network parameters of the preset neural network, and when the preset neural network reaches the set maximum number of training times or the loss value is less than the set loss value, stop updating to obtain a trained preset neural network.

[0054] Exemplarily, updating is stopped when the preset neural network is trained 100 times or the loss value is less than 0.001.

[0055] In one embodiment, the inner area of ​​the outer contour of the mineral material is obtained by using the edge filling method. It should be explained that the outer contour is completed by using the edge filling method, and the number of pixels in the completed outer contour is used as the inner area.

[0056] Calculate the internal area correction value, the internal area correction value satisfies the relationship: , represents the internal area correction value, represents the internal area, It represents the Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image. Represents half of the diagonal of the mineral grayscale image.

[0057] Among them, the Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image includes: constructing a position coordinate system for the mineral material grayscale image; calculating the mean of the horizontal coordinate and the mean of the vertical coordinate of all pixel points in the outer contour of the mineral material as the position coordinates of the outer contour of the mineral material, and calculating the Euclidean distance between the position coordinates and the center point of the mineral material grayscale image.

[0058] It should be noted that when collecting the grayscale image of the mineral material, due to the existence of perspective effect, the size of the mineral material in the image will change with the distance from the camera, that is, the mineral material closer to the camera appears larger in the image, while the mineral material farther away appears smaller. This perspective change will cause the outer contour area of ​​mineral materials of the same size in the image to be different, thus affecting the accurate detection of the degree of mineral crushing. By analyzing the relative position of the mineral block in the image and combining the area occupied by the outer contour of the mineral block, correction processing is performed.

[0059] The internal area correction value of each mineral material outer contour is obtained by traversing, and the internal area correction value mean of the mineral material grayscale image is calculated. The mineral material grayscale image with a crushing degree that meets the standard is used as the standard image, and the internal area correction value mean of the standard image is used as the standard mean.

[0060] In response to the internal area correction value mean being less than the standard mean, the crushing degree of the mineral material grayscale image is qualified.

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

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

Claims

1. A method for detecting the degree of crushing of mineral materials based on image analysis, characterized in that: include: Obtain the preprocessed mineral material grayscale image, calculate the gradient value of any pixel point in the mineral material grayscale image, and calculate the division difficulty of the mineral material grayscale image; Based on edge detection, the edges in the grayscale image of the mineral material are obtained, any edge is taken as the target edge, and the edges other than the target edge are taken as the reference edge. The minimum bounding rectangle of the target edge is constructed, and the number of reference edges in the minimum bounding rectangle is counted to calculate the outer contour probability of the target edge. The outer contour probability satisfies the relationship: , represents the outer contour probability, Represents the mean value of the gradient values ​​of all pixels on the target edge. Represents the variance of the gradient values ​​of all pixels on the target edge, Indicates the number of pixels on the edge of the target. represents the number of reference edges, represents the adjustment factor, Indicates the difficulty of division, represents the normalization function; The outer contour probability of all edges in the mineral material grayscale image is traversed to determine the outer contour of the mineral material. The mineral material grayscale image is converted into a binary image according to the outer contour of the mineral material. The binary image is input into the preset neural network to obtain the probability value of the crushing result and complete the crushing degree detection.

2. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: The pre-processing comprises: Gaussian filtering is used to denoise the mineral grayscale image, and adaptive histogram equalization is used to increase the contrast of the mineral grayscale image.

3. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: The division difficulty includes: Construct a gradient line graph with the gradient value as the horizontal coordinate and the number of pixels as the vertical coordinate, and obtain the maximum number of pixels in the gradient line graph; , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis.

4. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: The division difficulty also includes: Construct a gradient line graph with the gradient value as the horizontal coordinate and the number of pixels as the vertical coordinate, and obtain the maximum number of pixels in the gradient line graph; , Indicates the difficulty of division, Indicates the maximum number of pixels. Indicates The number of pixels on the horizontal axis, Indicates the value of the horizontal coordinate corresponding to the maximum number of pixels. Indicates The value of the horizontal axis.

5. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: Determining the outer contour of the mineral material comprises: All edges are clustered to obtain several clusters, the mean outer contour probability of all edges in any cluster is calculated, and the edge in the cluster with the largest outer contour probability mean is taken as the outer contour of the mineral material.

6. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: The preset neural network is a convolutional neural network, which extracts features from binary images to obtain image features, and classifies the image features to obtain crushing result probability values.

7. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 1, characterized in that: Also includes: The internal area of ​​the outer contour of the mineral material is obtained by using the edge filling method, and the internal area correction value is calculated; Traverse to obtain the internal area correction value of each mineral material outer contour, and calculate the mean internal area correction value of the mineral material grayscale image; The grayscale image of the mineral material whose crushing degree meets the standard is taken as the standard image, and the mean of the internal area correction value of the standard image is taken as the standard mean; In response to the internal area correction value mean being less than the standard mean, the crushing degree of the mineral material grayscale image is qualified.

8. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 7, characterized in that: The internal area correction value satisfies the relationship: , represents the internal area correction value, represents the internal area, It represents the Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image. Represents half of the diagonal of the mineral grayscale image.

9. The method for detecting the degree of crushing of mineral materials based on image analysis according to claim 8, characterized in that: The Euclidean distance between the outer contour of the mineral material and the center point of the mineral material grayscale image includes: Construct a position coordinate system for the mineral material grayscale image; The mean values ​​of the horizontal coordinates and the vertical coordinates of all the pixel points in the outer contour of the mineral material are calculated respectively as the position coordinates of the outer contour of the mineral material, and the Euclidean distance between the position coordinates and the center point of the mineral material grayscale image is calculated.

Images