Method for rapidly testing two-dimensional shape characteristics of aggregate

Abstract

The invention provides a method for rapidly testing two-dimensional shape characteristics of aggregate, and relates to the field of civil engineering materials. The method comprises the following steps: screening to obtain an enough amount of aggregate with a specified particle size; acquiring image information of the aggregate through an image acquisition device; the aggregate image is processed through gray level conversion, threshold segmentation and binarization methods, so that subsequent calculation is facilitated; equivalent ellipse information corresponding to the aggregate image is obtained by calculating a covariance matrix, and information such as the length-width ratio of an equivalent ellipse is further obtained; and calculating the perimeter and the area of the polygon through a contour edge identification method and a Green's formula to obtain a roundness value. The method not only can process aggregate information on a large scale at the same time to obtain key two-dimensional shape features of the batch of aggregate, but also has the advantages of simplicity in operation, short test period, accurate result and the like, and provides an efficient and convenient technical means for analyzing the influence of the aggregate form on the performance of cement concrete and asphalt concrete.

Description

Technical Field

[0001] The invention relates to the field of civil engineering materials, and in particular to a method for quickly testing two-dimensional shape characteristics of aggregates. Background Art

[0002] Aggregate is an indispensable component of asphalt and cement concrete, and plays a vital role in the overall performance of concrete. Specifically, the shape, edges, and texture of aggregate particles will affect the interaction between particles and their interaction with other materials (such as asphalt, cement, and lime), thereby affecting the durability, shear resistance, tensile strength, stiffness, and other properties of the asphalt concrete pavement surface layer, base layer, and cement concrete. Therefore, accurate testing and quantification of aggregate shape characteristics is of great significance for controlling the quality of aggregate processing and production and improving the performance of asphalt or cement concrete.

[0003] However, for a long time, the collection and analysis of aggregate shape characteristics have mainly relied on traditional testing methods, that is, relying on manual visual observation or simple measuring tools to judge the shape of aggregates. This method has the disadvantages of low efficiency and inability to fully quantify the shape characteristics of aggregates. For example, it is difficult to accurately determine the content of needle-like aggregates through manual testing, and too much needle-like aggregates will significantly weaken the workability and mechanical properties of the mixture. In other descriptions of aggregate shape, manual testing is even more difficult to achieve, and is easily affected by human subjective factors, resulting in great experimental deviations. In addition, the time-consuming and labor-intensive problems of traditional methods have greatly hindered the accurate measurement of aggregate shape characteristics.

[0004] With the vigorous development of digital technology, computer vision, image processing and other fields have made great progress. In the field of aggregate research, aggregate information collection technology has emerged. The popularization of high-resolution imaging equipment, such as industrial-grade high-definition cameras, microscope cameras, etc., can clearly capture the image information of aggregate particles. The collection methods of aggregate information can be divided into two-dimensional collection and three-dimensional collection. Three-dimensional information is more comprehensive, but this method is expensive, the digital image processing process is relatively complex, and it is difficult to promote. In contrast, the two-dimensional collection method is more time-saving and convenient, and can quickly obtain key information on aggregate shape characteristics. However, how to accurately extract key features based on the collected image information to obtain specific shape indicators still lacks a unified processing and analysis method.

[0005] Patent CN117929211A discloses a method for evaluating the three-dimensional morphological characteristics of coarse aggregate particles, which divides the regions and classifies the shapes of shotcrete coarse aggregates of different particle sizes. However, this method only evaluates through the particle size and axis ratio indicators, ignoring the different evaluation results that may be caused by other key indicators. Patent CN106780457A calculates the sphericity index, angularity index and texture index of aggregates through AIMS, but does not involve the calculation of indicators such as roundness and equivalent ellipse aspect ratio. The two-dimensional indicator only has the angularity index, which makes it difficult to comprehensively evaluate the two-dimensional shape characteristics of aggregates. Patent CN117740675A discloses an aggregate morphology analysis device and method based on line scanning, which can realize large-scale automated processing of aggregate samples, but this method cannot accurately segment and identify each aggregate on the conveyor belt, and the evaluation parameter is the sphericity parameter, which does not consider the shape parameters such as angularity that also affect the performance of asphalt or cement concrete. Patent CN105136622A discloses a method for testing the flatness of coarse aggregates. A batch of aggregate images are taken by a camera, and the aggregate images are processed by Image-Pro Plus to calculate the ratio of the major and minor axes of each aggregate. Although this method can process multiple aggregates at one time, the camera shooting accuracy is lower than that of the image acquisition design, and other two-dimensional shape features cannot be accurately calculated, which has limitations. Patent CN118052117A discloses a method for estimating the void ratio of aggregates based on two-dimensional images. The two-dimensional shape features of aggregates calculated by this method include roundness, convexity, aspect ratio and angularity. The focus is on exploring the estimation method of the void ratio of aggregates, and the processing method of the two-dimensional image of aggregates and the calculation method of shape features are not explored in detail. In comparison, the shape feature used in the present invention is the aspect ratio of the equivalent ellipse, rather than the aspect ratio directly calculated based on the major and minor axes of the aggregates. The equivalent ellipse has the same first-order moment and second-order moment as the original aggregate, and its shape feature is more equivalent to the original aggregate, and the result is more reasonable. Patent CN119131259A discloses a method for characterizing the three-dimensional morphological characteristics of coarse aggregate. This method can calculate the shape, edges, texture and other characteristics of coarse aggregate, but the calculation amount is relatively complex and requires testing through a three-dimensional laser scanner. It has high requirements for equipment and can only test aggregate characteristics one by one. For a large number of samples, the test workload is large and time-consuming. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention provides a method for quickly testing the two-dimensional shape characteristics of aggregates, which can realize the rapid testing of aggregate shape characteristics and has the advantages of simple operation, high detection efficiency, and high result accuracy. It provides a quick and efficient technical means for exploring the influence of aggregate shape characteristics on the performance of asphalt concrete, cement concrete, and inorganic binder stabilization layer.

[0007] To achieve the above objectives, the technical solution of the present invention is implemented through the following technical solutions:

[0008] A method for quickly testing the two-dimensional shape characteristics of aggregates, the method comprising the following steps:

[0009] S1. Clean the aggregate, dry it and cool it to room temperature;

[0010] S2, using an aggregate image measurement system to obtain aggregate image information;

[0011] S3, processing the aggregate image: including grayscale conversion, threshold segmentation, image binarization, and image inversion to obtain a processed aggregate image;

[0012] S4, searching for aggregate polygons in the processed aggregate image and obtaining polygon boundaries;

[0013] S5, obtaining the centroid of the aggregate polygon, calculating the first-order moment and second-order moment parameters, and obtaining the major and minor axis information of the equivalent ellipse through covariance;

[0014] S6. Calculate the area and perimeter information of the aggregate polygon;

[0015] S7. Calculate the major axis length, minor axis length, angularity, roundness, equivalent ellipse aspect ratio and other characteristic information of all tested aggregates to obtain comprehensive aggregate two-dimensional shape characteristic indicators.

[0016] Preferably, the aggregate particle size range used in step S1 is 0.075 mm-26.5 mm.

[0017] Preferably, in step S2, an aggregate image measurement system is used to obtain aggregate image information. After the test, two-dimensional shape index data including major axis length, minor axis length, and angularity of the aggregate can be directly obtained, and an aggregate image with a file suffix of tif can be obtained.

[0018] Preferably, in step S3, the image is binarized by an algorithm, and the polygon is set as the foreground, that is, the polygon value is 1 and the background value is 0.

[0019] Preferably, the Canny edge detection algorithm is used in step S4 to read multiple aggregate image information in one image at one time. The specific implementation process is as follows:

[0020] S4-1. Use Gauss filtering to reduce the noise of the image. In Gauss smoothing, pixels at different positions will be assigned different weights, so that adjacent pixels have larger weights. Use Gauss filtering with a kernel size of 3*3 or 5*5 for processing. The kernel size must be N*N and an odd number greater than 1.

[0021] S4-2, use the Sobel operator to calculate the gradient of the image, apply the convolution template to the horizontal and vertical directions respectively, and obtain the calculated gradient value and direction. The gradient direction can be divided into four directions: 0°, 45°, 90° and 135°;

[0022] S4-3, adopt the non-maximum suppression strategy, among the pixels that may be detected as edges, only the pixels with the largest local gradient are retained, and the edge response corresponding to the non-maximum pixels is suppressed to 0, so as to obtain a more accurate edge with a single pixel width. When the gradient direction is 45°, the detection formula is:

[0023]

[0024] The edge of the aggregate polygon is determined based on double threshold detection, where the high threshold is set to 25% of the peak value of the gradient amplitude, and the low threshold is set to 40% of the high threshold, that is, 10% of the peak value of the gradient amplitude.

[0025] Preferably, in step S5, by performing eigenvalue decomposition on the covariance matrix, the major axis and minor axis information of the equivalent ellipse is obtained, and the equivalent ellipse has the same first-order moment and second-order moment as the original aggregate, and the decomposition formula is:

[0026] Σ=QΛQ -1

[0027] Among them, Σ is the covariance matrix, Q is the orthogonal matrix composed of eigenvectors, and Λ is the diagonal matrix;

[0028] The contour of a two-dimensional polygon can be viewed as a set of points on a plane, that is, a two-dimensional data set, with points (x i ,y i ), where i = 1, 2, 3...n; the covariance matrix Σ is expressed as follows:

[0029]

[0030] Among them, Cov(x,x) is the variance of variable x, Cov(y,y) is the variance of variable y, and Cov(x,y)=Cov(y,x) is the covariance of x and y;

[0031] In the two-dimensional case, two eigenvalues ​​can be obtained. The eigenvector corresponding to the eigenvalue represents the extension direction of the contour point on the plane. The center of the equivalue ellipse of the aggregate polygon is located at the mean point of the data. The directions of the major axis and the minor axis are determined by the eigenvector of the covariance matrix. The major axis direction corresponds to the eigenvector corresponding to the larger eigenvalue, and the minor axis direction corresponds to the eigenvector corresponding to the smaller eigenvalue.

[0032] Preferably, in step S6, the area of ​​the polygon is calculated using Green's formula;

[0033] Since the aggregate polygonal contour can be regarded as a continuous closed curve with a plane area of ​​D and a boundary of C, the polygonal area can be calculated using Green's formula, as shown below:

[0034]

[0035] Among them, P and Q are two functions defined on the open area containing the plane area and its boundary; in order to more conveniently calculate the area of ​​the polygon, take P = -y / 2, Q = x / 2, then according to Green's formula, the area of ​​the polygon can be expressed as:

[0036] S=∮ C xdy-ydx.

[0037] Preferably, in step S7, the formula for calculating the two-dimensional shape index roundness R and the equivalent ellipse aspect ratio AR is:

[0038]

[0039] Where p is the perimeter of the two-dimensional projection of the aggregate particle, and A is the area of ​​the two-dimensional projection of the aggregate particle;

[0040]

[0041] Among them, L max is the major axis length of the isovalue ellipse, L min is the length of the minor axis of the isoellipse.

[0042] Preferably, the method can not only process single image information, but also process multiple image information at one time.

[0043] The present invention provides a method for quickly testing the two-dimensional shape characteristics of aggregates, which has the following advantages over the prior art:

[0044] (1) The present invention realizes efficient analysis of aggregate morphological characteristics through an aggregate image measurement system. While accurately measuring basic geometric parameters such as the major axis and minor axis of the aggregate, digital image information of the aggregate is synchronously collected. Through grayscale conversion, threshold segmentation, image binarization, image inversion and other operations, an image that can be used for information extraction is obtained, and the perimeter, area and other information of the aggregate polygon are calculated in combination with the Canny algorithm, Green's formula and the like, and then the morphological characteristic information such as roundness and equivalent ellipse aspect ratio is calculated. The present invention can quickly and accurately obtain sufficient key indicators of aggregate shape characteristics, and has the advantages of simple operation, high detection efficiency, and high result accuracy. It provides a quick and efficient technical means for quantitatively analyzing the impact of aggregate shape characteristics on the performance of asphalt concrete and cement concrete;

[0045] (2) The present invention proposes a systematic and comprehensive method for calculating the two-dimensional shape characteristic parameters of aggregates. The present invention uses an image acquisition device to obtain high-precision two-dimensional image information of aggregates, and obtains key information such as the major axis length, minor axis length, angularity, roundness, and equivalent ellipse aspect ratio of aggregates through steps such as image binarization processing, contour detection, and eigenvalue covariance calculation. Compared with the existing methods, the present invention innovatively proposes the equivalent ellipse aspect ratio index to evaluate the shape characteristics of aggregates, and comprehensively characterizes the two-dimensional characteristics of aggregates by combining multiple indicators, and finally forms an efficient aggregate image processing method and a two-dimensional shape characteristic index calculation method. This method is more effective, more accurate, and more sensitive in quantifying and distinguishing the shape characteristics of aggregates, and has the characteristics of fast processing speed and high result accuracy. It is of great significance for efficiently and accurately obtaining the shape characteristics of aggregates, and then exploring their influence on the performance of asphalt concrete and cement concrete. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the test process of the present invention;

[0047] Figure 2 It is a scanning diagram of the aggregate in Example 1 of the present invention;

[0048] Figure 3 is the aggregate map after processing in Example 1 of the present invention;

[0049] Figure 4 This is a scanning diagram of the aggregate in Example 2 of the present invention;

[0050] Figure 5 This is the aggregate diagram after processing in Example 2 of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention is clearly and completely described below in combination with 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Embodiment 1:

[0053] A method for quickly testing the two-dimensional shape characteristics of aggregates comprises the following steps:

[0054] S1. Obtain a sufficient amount of aggregates within a specified particle size range, clean the aggregates, dry them, and cool them to room temperature. The aggregate particle size used in this embodiment is 9.5-13.2 mm, and the number of aggregate information collected is 50.

[0055] S2. Use the aggregate image measurement system (AIMS) to obtain aggregate image information, obtain the major axis length, minor axis length and image file of this batch of aggregates, and the obtained image file is in tif format. In the obtained image file, each image contains only one aggregate information, and a total of 50 image files are obtained. The obtained aggregate image is as follows: Figure 2 shown.

[0056] S3, the aggregate image is processed, including grayscale conversion, threshold segmentation, image binarization, image inversion, etc., to obtain the processed aggregate image (in order to better obtain the aggregate image information, the image is binarized through the algorithm, and the polygon is set as the foreground, that is, the polygon value is 1 and the background value is 0). The aggregate image after segmentation processing is as follows Figure 3 shown.

[0057] S4, find the aggregate polygon in the current image and obtain the polygon boundary. The image is processed by the Canny algorithm. In the aggregate image in this embodiment, each image contains only one aggregate polygon information:

[0058] Through the Canny edge detection algorithm, it is possible to read multiple aggregate image information in an image at one time. The specific implementation process is as follows:

[0059] Use Gauss filtering to reduce the noise of the image. In Gauss smoothing, pixels at different positions will be assigned different weights, so that adjacent pixels have larger weights. Gauss filtering with a kernel size of 3*3 or 5*5 can be used for processing. The kernel size must be N*N and an odd number greater than 1.

[0060] The Sobel operator is used to calculate the gradient of the image. The convolution template is applied in the horizontal and vertical directions to obtain the calculated gradient value and direction. The gradient direction can be divided into four directions: 0°, 45°, 90°, and 135°.

[0061] The non-maximum suppression strategy is adopted. Among the pixels that may be detected as edges, only the pixels with the largest local gradient are retained, and the edge responses corresponding to the non-maximum pixels are suppressed to 0, thereby obtaining a more accurate edge with a single pixel width. When the gradient direction is 45°, the detection formula is:

[0062]

[0063] The edge of the aggregate polygon is determined based on double threshold detection, where the high threshold is set to 25% of the peak value of the gradient amplitude, and the low threshold is set to 40% of the high threshold, that is, 10% of the peak value of the gradient amplitude.

[0064] S5. Get the centroid of the aggregate polygon, calculate the first-order moment and second-order moment parameters, and obtain the major and minor axis information of the equivalent ellipse through covariance:

[0065] By performing eigenvalue decomposition on the covariance matrix, the major and minor axis information of the equivalent ellipse is obtained. The equivalent ellipse has the same first-order moment and second-order moment as the original aggregate. The decomposition formula is:

[0066] Σ=QΛQ -1

[0067] Among them, Σ is the covariance matrix, Q is the orthogonal matrix composed of eigenvectors, and Λ is the diagonal matrix.

[0068] The contour of a two-dimensional polygon can be viewed as a set of points on a plane, that is, a two-dimensional data set, with points (x i ,y i ), where i = 1, 2, 3, ... n. The covariance matrix Σ is expressed as follows:

[0069]

[0070] Among them, Cov(x,x) is the variance of variable x, Cov(y,y) is the variance of variable y, and Cov(x,y)=Cov(y,x) is the covariance of x and y.

[0071] In the two-dimensional case, two eigenvalues ​​can be obtained, and the eigenvectors corresponding to the eigenvalues ​​represent the extension direction of the contour points on the plane. The center of the equivalue ellipse of the aggregate polygon is located at the mean point of the data. The directions of the major and minor axes are determined by the eigenvectors of the covariance matrix. The major axis direction corresponds to the eigenvector corresponding to the larger eigenvalue, and the minor axis direction corresponds to the eigenvector corresponding to the smaller eigenvalue.

[0072] S6. Calculate the area, perimeter and other information of the aggregate polygon:

[0073] Use Green's formula to calculate the area of ​​polygons;

[0074] Since the aggregate polygonal contour can be regarded as a continuous closed curve with a plane area of ​​D and a boundary of C, the polygonal area can be calculated using Green's formula, as shown below:

[0075]

[0076] Among them, P and Q are two functions defined on the open area containing the plane area and its boundary. In order to more conveniently calculate the area of ​​the polygon, take P = -y / 2, Q = x / 2, then according to Green's formula, the area of ​​the polygon can be expressed as:

[0077] S=∮ C xdy-ydx

[0078] S7. Calculate the roundness, aspect ratio and other information of all tested aggregates:

[0079] The formula for calculating the two-dimensional shape index roundness R and the equivalent ellipse aspect ratio AR is:

[0080]

[0081] Where p is the perimeter of the two-dimensional projection of the aggregate particle, and A is the area of ​​the two-dimensional projection of the aggregate particle.

[0082]

[0083] Among them, L max is the major axis length of the isovalue ellipse, L min is the length of the minor axis of the isoellipse.

[0084] Finally, the shape characteristic information of all aggregates in this batch was obtained. Table 1 shows the shape characteristic results of 20 particles in this batch.

[0085] Table 1

[0086]

[0087] Embodiment 2:

[0088] A method for quickly testing the two-dimensional shape characteristics of aggregates comprises the following steps:

[0089] S1. Obtain a sufficient amount of aggregates within a specified particle size range, clean the aggregates, dry them, and cool them to room temperature. The aggregate particle size used in this embodiment is 1.18-2.36 mm, and the number of aggregate information collected is 100.

[0090] S2. Use the aggregate image measurement system (AIMS) to obtain aggregate image information, obtain the major axis length, minor axis length and image file of this batch of aggregates, and the obtained image file is in tif format. In the image file obtained in this embodiment, each image contains 2-5 aggregate information, and a total of 32 image files are obtained. The image containing multiple aggregate information is as follows: Figure 4 shown.

[0091] S3, processing the aggregate image, including grayscale conversion, threshold segmentation, image binarization, image inversion, etc., to obtain the processed aggregate image. The aggregate image after segmentation is as follows: Figure 5 shown.

[0092] S4, searching for aggregate polygons in the current image and obtaining polygon boundaries. The image is processed by the Canny algorithm. In the aggregate images in this embodiment, each image contains multiple aggregate polygon information. This method can identify different polygons in the image.

[0093] S5. Obtain the centroid of the aggregate polygon, calculate the first-order moment and second-order moment parameters, and obtain the major and minor axis information of the equivalent ellipse through covariance.

[0094] S6. Calculate the area, perimeter and other information of the aggregate polygon;

[0095] S7. Calculate the roundness, aspect ratio and other information of all tested aggregates.

[0096] Finally, the shape characteristic information of all aggregates in this batch was summarized, and Table 2 shows the shape characteristic results of 20 particles in this batch.

[0097] Table 2

[0098]

[0099]

[0100] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, 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 embodiments of the present invention.

Claims

1. A method for quickly testing the two-dimensional shape characteristics of aggregates, characterized in that: The method comprises the following steps: S1. Clean the aggregate, dry it and cool it to room temperature; S2, using an aggregate image measurement system to obtain aggregate image information; S3, processing the aggregate image: including grayscale conversion, threshold segmentation, image binarization, and image inversion to obtain a processed aggregate image; S4, searching for aggregate polygons in the processed aggregate image and obtaining polygon boundaries; S5, obtaining the centroid of the aggregate polygon, calculating the first-order moment and second-order moment parameters, and obtaining the major and minor axis information of the equivalent ellipse through covariance; S6. Calculate the area and perimeter information of the aggregate polygon; S7. Calculate the major axis length, minor axis length, angularity, roundness, equivalent ellipse aspect ratio and other characteristic information of all tested aggregates to obtain comprehensive aggregate two-dimensional shape characteristic indicators.

2. The method according to claim 1, characterized in that: The aggregate particle size range used in step S1 is 0.075 mm-26.5 mm.

3. The method according to claim 1, characterized in that: In step S2, an aggregate image measurement system is used to obtain aggregate image information. After the test, two-dimensional shape index data including the major axis length, minor axis length, and angularity of the aggregate can be directly obtained, and an aggregate image with a file suffix of tif can be obtained.

4. The method according to claim 1, characterized in that: In the step S3, the image is binarized by an algorithm, and the polygon is set as the foreground, that is, the polygon value is 1 and the background value is 0.

5. The method according to claim 1, characterized in that In step S4, the Canny edge detection algorithm is used to read multiple aggregate image information in one image at one time. The specific implementation process is as follows: S4-1. Use Gauss filtering to reduce the noise of the image. In Gauss smoothing, pixels at different positions will be assigned different weights, so that adjacent pixels have larger weights. Use Gauss filtering with a kernel size of 3*3 or 5*5 for processing. The kernel size must be N*N and an odd number greater than 1. S4-2, use the Sobel operator to calculate the gradient of the image, apply the convolution template to the horizontal and vertical directions respectively, and obtain the calculated gradient value and direction. The gradient direction can be divided into four directions: 0°, 45°, 90° and 135°; S4-3, adopt the non-maximum suppression strategy, among the pixels that may be detected as edges, only the pixels with the largest local gradient are retained, and the edge response corresponding to the non-maximum pixels is suppressed to 0, so as to obtain a more accurate edge with a single pixel width. When the gradient direction is 45°, the detection formula is: The edge of the aggregate polygon is determined based on double threshold detection, where the high threshold is set to 25% of the peak value of the gradient amplitude, and the low threshold is set to 40% of the high threshold, that is, 10% of the peak value of the gradient amplitude.

6. The method according to claim 1, characterized in that: In step S5, by performing eigenvalue decomposition on the covariance matrix, the major axis and minor axis information of the equivalent ellipse is obtained. The equivalent ellipse has the same first-order moment and second-order moment as the original aggregate. The decomposition formula is: S=QΛQ -1 Among them, Σ is the covariance matrix, Q is the orthogonal matrix composed of eigenvectors, and Λ is the diagonal matrix; The contour of a two-dimensional polygon can be viewed as a set of points on a plane, that is, a two-dimensional data set, with points (x i ,y i ), where i = 1, 2, 3...n; the covariance matrix Σ is expressed as follows: Among them, Cov(x,x) is the variance of variable x, Cov(y,y) is the variance of variable y, and Cov(x,y)=Cov(y,x) is the covariance of x and y; In the two-dimensional case, two eigenvalues ​​can be obtained. The eigenvector corresponding to the eigenvalue represents the extension direction of the contour point on the plane. The center of the equivalue ellipse of the aggregate polygon is located at the mean point of the data. The directions of the major axis and the minor axis are determined by the eigenvector of the covariance matrix. The major axis direction corresponds to the eigenvector corresponding to the larger eigenvalue, and the minor axis direction corresponds to the eigenvector corresponding to the smaller eigenvalue.

7. The method according to claim 1, characterized in that: In step S6, the area of ​​the polygon is calculated using Green's formula; Since the aggregate polygonal contour can be regarded as a continuous closed curve with a plane area of ​​D and a boundary of C, the polygonal area can be calculated using Green's formula, as shown below: Among them, P and Q are two functions defined on the open area containing the plane area and its boundary; in order to more conveniently calculate the area of ​​the polygon, take P = -y / 2, Q = x / 2, then according to Green's formula, the area of ​​the polygon can be expressed as:

8. The method according to claim 1, characterized in that: In step S7, the formula for calculating the two-dimensional shape index roundness R and the equivalent ellipse aspect ratio AR is: Where p is the perimeter of the two-dimensional projection of the aggregate particle, and A is the area of ​​the two-dimensional projection of the aggregate particle; Among them, L max is the length of the major axis of the isovalue ellipse, L min is the length of the minor axis of the isoellipse.

9. The method according to any one of claims 1 to 8, characterized in that: The method can not only process single image information, but also process multiple image information at one time.

Images