Quantitative detection method and detection system for mesoscopic defects of soil-rock aggregate based on CT (Computed Tomography) image multi-threshold segmentation

Abstract

The invention provides a quantitative detection method and a quantitative detection system for mesoscopic defects of a soil-rock aggregate based on CT image multi-threshold segmentation, and belongs to the technical field of nondestructive testing and digital image processing of geotechnical engineering materials. The method comprises the following steps: firstly, carrying out multi-threshold segmentation on a CT image by adopting a fuzzy C-means clustering algorithm, and analyzing the image into four types of regions including a background region, a block stone region, a matrix soil body region and a mesoscopic defect region according to gray features; then, based on the heterogeneity characteristic of the material, the mesoscopic defects are divided into matrix defects and interface defects according to defect space distribution; the boundary contour of the block stone is extracted through morphological operation, and intelligent judgment of defect types is realized by combining spatial topology analysis of a defect binary image; and finally quantifying the volume ratio parameters of the two types of defects based on pixel statistics. According to the method, the limitation of a traditional homogeneous material defect analysis method is broken through, and a new technical means is provided for research on damage evolution of the earth-rock aggregate multiphase structure.

Description

Technical Field

[0001] The present invention relates to the technical field of non-destructive testing of geotechnical engineering materials and digital image processing, and particularly relates to a method and a detection system for quantitatively detecting mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images. Background Art

[0002] As a typical heterogeneous material widely existing in geological engineering such as landslides and accumulation layers, the mechanical properties deterioration of soil-rock mixtures is closely related to the evolution of internal mesoscopic defects (such as pores, fissures, etc.). Traditional geotechnical testing methods mostly rely on physical and mechanical tests or two-dimensional slice observations, which have limitations such as destroying the integrity of samples and being difficult to visualize the three-dimensional distribution of defects. Although computer tomography (CT) technology provides the possibility for non-destructive testing, when dealing with multi-phase media such as soil-rock mixtures, conventional image segmentation methods (such as single-threshold method, region growing method) are prone to misjudgment or missed detection of defects due to problems such as overlapping gray levels of material components and blurred interfaces, and it is difficult to accurately separate boulders, matrixes and defects, let alone distinguish the different effects of interface defects and matrix defects on mechanical properties.

[0003] In existing patents and literature, defect analysis based on CT images mostly focuses on homogeneous materials (such as concrete, rock), lacking dedicated methods for heterogeneous soil-rock mixtures. For example, traditional clustering algorithms do not consider the constraint of boulder boundary morphology on defect classification, resulting in insufficient accuracy in identifying interface defects; while defect quantification methods that solely rely on gray level statistics are difficult to establish quantitative correlations between mesoscopic defect parameters and macroscopic modulus, strength and other indicators. In addition, existing technologies often ignore the influence of the spatial distribution heterogeneity of defects on engineering stability, and cannot meet the requirements of defect type discrimination and damage evolution analysis in scenarios such as landslide monitoring and subgrade evaluation. Therefore, there is an urgent need to develop a defect detection method that integrates multi-modal image processing to solve the adaptability problem of defect analysis for heterogeneous materials. Summary of the Invention

[0004] The purpose of the present invention is to propose a method and a detection system for quantitatively detecting mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images, and through integrating fuzzy clustering algorithms, morphological operations and spatial topology analysis technologies, to achieve full-process innovation from image analysis to defect classification and evaluation.

[0005] The technical solution adopted by the present invention: In the first aspect, the present invention proposes a method for quantitatively detecting mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images, and the method includes:

[0006] Obtain the original CT image of the soil-rock mixture and extract its gray level data;

[0007] Perform multi-threshold segmentation on the grayscale data using the fuzzy C-means clustering algorithm to determine three grayscale thresholds Th1, Th2, and Th3. Among them, Th1 is the threshold for distinguishing the background from the target object, Th2 is the threshold for distinguishing the block stones from the matrix soil mass, and Th3 is the threshold for distinguishing the matrix soil mass from the mesoscopic defects;

[0008] According to the three grayscale thresholds, segment the original CT image into four regions: background, block stones, matrix soil mass, and mesoscopic defects, and generate a binary defect image;

[0009] Perform morphological processing on the block stone region, and extract the boundary contour of the block stones through dilation-erosion operations to generate a binary block stone boundary image;

[0010] Perform spatial topological analysis on the binary defect image and the binary block stone boundary image, and separate the interface defects and matrix defects through image subtraction operations. Among them, the interface defects are the defect regions overlapping with the block stone boundary, and the matrix defects are the defect regions independently distributed inside the matrix soil mass;

[0011] Statistically calculate the pixel proportion of the interface defects and the matrix defects, calculate their volume proportion in the overall image of the soil-rock mixture, and output the quantitative detection results.

[0012] As a further improvement of the present invention, the objective function of the fuzzy C-means clustering algorithm is defined as:

[0013]

[0014] where N is the number of data points, C is the number of clusters, μ ij is the membership degree of pixel x i belonging to the j-th class, c j is the cluster center, m is the fuzzy weight coefficient, and for a given data point, the sum of its membership degrees belonging to each class is 1.

[0015] As a further improvement of the present invention, the kernel function of the dilation-erosion operation after morphological processing is a circular structural element, and the ratio of its radius to the average particle size of the block stones is 0.1 - 0.2.

[0016] As a further improvement of the present invention, the spatial topological analysis includes:

[0017] Perform a logical AND operation on the binary block stone boundary image and the binary defect image to extract the overlapping region as the interface defects;

[0018] Subtract the interface defect region from the binary defect image, and the remaining part is defined as the matrix defects.

[0019] As a further improvement of the present invention, the calculation formula for the volume proportion is:

[0020]

[0021] Among them, N defect is the number of pixels in the defect area, and N total is the total number of pixels in the effective area of the soil-rock mixture.

[0022] In a second aspect, the present invention also provides a mesoscopic defect quantitative detection system for soil-rock mixtures based on multi-threshold segmentation of CT images. The detection system includes

[0023] an image acquisition module for acquiring the original CT image of the soil-rock mixture and extracting its grayscale data;

[0024] an image processing module for performing multi-threshold segmentation on the grayscale data using the fuzzy C-means clustering algorithm to determine three grayscale thresholds Th1, Th2, and Th3. Among them, Th1 is the threshold for distinguishing the background from the target object, Th2 is the threshold for distinguishing the boulders from the matrix soil, and Th3 is the threshold for distinguishing the matrix soil from the mesoscopic defects; and segmenting the original CT image into four regions: background, boulders, matrix soil, and mesoscopic defects according to the three grayscale thresholds, and generating a defect binary image; and performing morphological processing on the boulder region, and extracting the boulder boundary contour through dilation-erosion operations to generate a boulder boundary binary image;

[0025] a defect classification module for performing spatial topological analysis on the defect binary image and the boulder boundary binary image, and separating the interface defects and matrix defects through image subtraction operations. Among them, the interface defects are the defect areas overlapping with the boulder boundary, and the matrix defects are the defect areas independently distributed inside the matrix soil;

[0026] a statistical output module for statistically calculating the pixel proportion of the interface defects and matrix defects, calculating their volume proportion in the overall image of the soil-rock mixture, and outputting the quantitative detection results.

[0027] In a third aspect, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the steps of the method described in any one of the above are implemented.

[0028] Compared with the prior art, the present invention has the following technical effects:

[0029] (1) The present invention uses the fuzzy C-means clustering algorithm to perform multi-threshold segmentation on the grayscale data of the CT image, so that each pixel has the highest membership degree in its corresponding category. The segmentation result is accurate and reliable, and the grayscale data is accurately divided into four categories of regions: background, boulders, matrix soil, and mesoscopic defects;

[0030] (2) Based on the different specific positions of the defects, the present invention performs morphological analysis and spatial topology operations on the CT images, further differentiating the mesoscopic defects of the soil-rock mixture into matrix defects and interface defects, which can more effectively reflect different mesoscopic damage mechanisms;

[0031] (3) Based on pixel statistics, the present invention quantifies the volume ratios of the two types of defects, which can be used as the basis for correlating the evolution of macroscopic mechanical properties;

[0032] (4) This method breaks through the limitations of the traditional defect analysis framework for homogeneous materials, significantly improving the accuracy and engineering applicability of defect detection for heterogeneous materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0034] Figure 1 is the image flowchart of the intelligent discrimination and quantitative detection method for mesoscopic defects of soil-rock mixture based on multi-threshold segmentation of CT images of the present invention;

[0035] Figure 2 is a schematic diagram of the defect classification result of the frozen-thawed soil-rock mixture specimen in the embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be described clearly and completely below. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention fall within the scope of protection of the present invention.

[0037]

Embodiment 1

[0038] The present invention proposes a quantitative detection method for mesoscopic defects of soil-rock mixture based on multi-threshold segmentation of CT images, including:

[0039] Obtain the original CT image of the soil-rock mixture and extract its gray-scale data.

[0040] Perform multi-threshold segmentation on the gray-scale data using the fuzzy C-means clustering algorithm to determine three gray-scale thresholds Th1, Th2, and Th3. Among them, Th1 is the threshold for distinguishing the background from the target object, Th3 is the threshold for distinguishing the block stones from the matrix soil, and Th3 is the threshold for distinguishing the matrix soil from the mesoscopic defects. Specifically, the objective function of the fuzzy C-means clustering algorithm is defined as:

[0041]

[0042] where N is the number of data points, C is the number of clusters, μij For pixel x i Degree of membership belonging to the j-th class, c j Is the cluster center, m is the fuzzy weight coefficient. For a given data point, the sum of the degrees of membership belonging to each class is 1.

[0043] The original CT image is segmented into four regions: background, block stones, matrix soil mass, and meso-defects according to three gray thresholds, and a binary defect image is generated.

[0044] Morphological processing is performed on the block stone region. The boundary contour of the block stones is extracted through dilation-erosion operations to generate a binary block stone boundary image. Specifically, the kernel function of the dilation-erosion operation for morphological processing is a circular structuring element, and the ratio of its radius to the average particle size of the block stones is 0.1 - 0.2.

[0045] Spatial topological analysis is performed on the binary defect image and the binary block stone boundary image. Interface defects and matrix defects are separated through image subtraction operations. Among them, the interface defects are the defect regions overlapping with the block stone boundary, and the matrix defects are the defect regions independently distributed inside the matrix soil mass. The spatial topological analysis includes: performing a logical AND operation on the binary block stone boundary image and the binary defect image to extract the overlapping region as the interface defects; subtracting the interface defect region from the binary defect image, and the remaining part is defined as the matrix defects.

[0046] Statistical analysis of the pixel proportion of the interface defects and the matrix defects is carried out, and the volume ratio of them in the overall image of the soil-rock mixture is calculated, and the quantitative detection results are output. Specifically, the calculation formula for the volume ratio is:

[0047]

[0048] Where, N defect Is the number of pixels in the defect region, N total Is the total number of pixels in the effective region of the soil-rock mixture.

[0049] Figure 1 Shows the complete process and key steps of the meso-defect quantitative detection method for soil-rock mixtures based on multi-threshold segmentation of CT images in the present invention, specifically including:

[0050] 1. Input of the original CT image (a): The three-dimensional CT image of the soil-rock mixture specimen after freeze-thaw cycles, with a gray level range of 0 - 255.

[0051] 2. Gray level frequency distribution and threshold segmentation image (b - c):

[0052] The horizontal axis represents the gray value (0 - 255), and the vertical axis represents the pixel frequency; the three-level thresholds (Th1, Th2, Th3) are determined by the fuzzy C-means clustering algorithm, and the image is segmented into four types of regions: background (gray value ≥ Th1), block stones (Th2 ≤ gray value < Th1), matrix soil (Th3 ≤ gray value < Th2), and mesoscopic defects (gray value < Th3).

[0053] 3. Morphological processing and defect classification:

[0054] The binary image of the block stones (d) extracts the boundary (e) through dilation - erosion operations;

[0055] The binary image of the defects (g) performs a logical operation with the boundary of the block stones to separate the interface defects (f) and the matrix defects (h);

[0056] 4. Pixel statistics and result output:

[0057] Quantitatively display the proportion of matrix defects and the proportion of interface defects.

[0058] Figure 2 Shows the defect classification results of the soil - rock mixture specimen after 6 freeze - thaw cycles in the embodiment. Among them, Figure 2 (a) is the distribution map of interface defects. The red highlighted area is the defects (pores / fissures) at the soil - rock interface, which are extracted by the logical "AND" operation of the binary image of the block stone boundary and the binary image of the defects, reflecting the damage concentration phenomenon in the contact zone between the block stones and the matrix soil. Figure 2 (b) is the distribution map of matrix defects. The blue area is the independent defects inside the matrix soil, which is obtained by subtracting the interface defects from the total defect area, characterizing the deterioration of the internal structure of the soil caused by freeze - thaw cycles.

[0059]

Embodiment 2

[0060] The present invention proposes a mesoscopic defect quantitative detection system for soil - rock mixtures based on multi - threshold segmentation of CT images, including:

[0061] An image acquisition module for acquiring the original CT image of the soil - rock mixture and extracting its gray - scale data;

[0062] An image processing module for performing multi - threshold segmentation on the gray - scale data using the fuzzy C - means clustering algorithm to determine three gray - scale thresholds Th1, Th2, and Th3. Among them, Th1 is the threshold for distinguishing the background from the target object, Th2 is the threshold for distinguishing the block stones from the matrix soil, and Th3 is the threshold for distinguishing the matrix soil from the mesoscopic defects; and segmenting the original CT image into four regions: background, block stones, matrix soil, and mesoscopic defects according to the three gray - scale thresholds, and generating a binary image of the defects; and performing morphological processing on the block - stone region, extracting the boundary contour of the block stones through dilation - erosion operations, and generating a binary image of the block - stone boundary;

[0063] A defect classification module for performing spatial topology analysis on the defect binary image and the block stone boundary binary image, separating the interface defect and the matrix defect through image subtraction operation, where the interface defect is the defect area overlapping with the block stone boundary, and the matrix defect is the defect area independently distributed inside the matrix soil body;

[0064] A statistical output module for calculating the pixel proportion of the interface defect and the matrix defect, calculating their volume proportion in the overall image of the soil-rock mixture, and outputting the quantitative detection result.

[0065]

Example 3

[0066] An embodiment of the present invention provides a computer-readable storage medium with a computer program stored thereon. When the computer program is executed, the steps of any one of the above methods are realized.

[0067] This method aims at the original CT image of the soil-rock mixture, separates the interface defect and the matrix defect through the fusion of the fuzzy clustering algorithm, morphological operation and spatial topology analysis, then calculates the pixel proportion of the interface defect and the matrix defect, and calculates their volume proportion in the overall image of the soil-rock mixture, realizing the full-process innovation from image analysis to defect classification and evaluation.

[0068] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited thereto. All changes that can be made within the knowledge scope of those skilled in the art without departing from the purpose of the present invention are within the protection scope of the claims of the present invention.

Claims

1. A quantitative detection method for mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images, characterized in that The method includes: Obtain the original CT image of the soil-rock mixture and extract its grayscale data; Perform multi-threshold segmentation on the grayscale data using the fuzzy C-means clustering algorithm to determine three grayscale thresholds Th1, Th2, and Th3, where Th1 is the threshold for distinguishing the background from the target object, Th2 is the threshold for distinguishing the block stones from the matrix soil, and Th3 is the threshold for distinguishing the matrix soil from the mesoscopic defects; According to the three grayscale thresholds, segment the original CT image into four regions: background, block stones, matrix soil, and mesoscopic defects, and generate a defect binary image; Perform morphological processing on the block stone region, and extract the boundary contour of the block stones through dilation-erosion operations to generate a block stone boundary binary image; Perform spatial topological analysis on the defect binary image and the block stone boundary binary image, and separate the interface defects and matrix defects through image subtraction operations, where the interface defects are the defect regions overlapping with the block stone boundary, and the matrix defects are the defect regions independently distributed inside the matrix soil; Statistically calculate the pixel proportion of the interface defects and the matrix defects, calculate their volume proportion in the overall image of the soil-rock mixture, and output the quantitative detection result.

2. The quantitative detection method for mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images according to claim 1, characterized in that The objective function of the fuzzy C-means clustering algorithm is defined as: where N is the number of data points, C is the number of clusters, μ ij is the membership degree of pixel x i belonging to the j-th class, c j is the cluster center, m is the fuzzy weight coefficient, and for a given data point, the sum of the membership degrees belonging to each class is 1.

3. The quantitative detection method for mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images according to claim 1, characterized in that, The kernel function of the dilation-erosion operation after morphological processing is a circular structural element, and the ratio of its radius to the average particle size of the block stones is 0.1 - 0.

2.

4. The quantitative detection method for mesoscopic defects of soil-rock mixtures based on multi-threshold segmentation of CT images according to claim 1, characterized in that, The spatial topological analysis includes: Perform a logical AND operation on the block stone boundary binary image and the defect binary image to extract the overlapping region as the interface defects; Subtract the interface defect region from the defect binary image, and the remaining part is defined as the matrix defects.

5. The method for quantitatively detecting mesoscopic defects of soil-rock mixture based on multi-threshold segmentation of CT images according to claim 1, characterized in that, The calculation formula for the volume proportion is: Among them, N defect is the number of pixels in the defective area, and N total is the total number of pixels in the effective area of the soil-rock mixture.

6. A mesoscopic defect quantitative detection system for soil-rock mixtures based on multi-threshold segmentation of CT images, characterized in that, The detection system includes An image acquisition module for obtaining the original CT image of the soil-rock mixture and extracting its grayscale data; An image processing module for performing multi-threshold segmentation on the grayscale data using the fuzzy C-means clustering algorithm to determine three grayscale thresholds Th1, Th2, and Th3, where Th1 is the threshold for distinguishing the background from the target object, Th2 is the threshold for distinguishing the block stones from the matrix soil, and Th3 is the threshold for distinguishing the matrix soil from the mesoscopic defects; and segmenting the original CT image into four regions: background, block stones, matrix soil, and mesoscopic defects according to the three grayscale thresholds, and generating a defect binary image; and performing morphological processing on the block stone region, and extracting the boundary contour of the block stones through dilation-erosion operations to generate a block stone boundary binary image; A defect classification module for performing spatial topological analysis on the defect binary image and the block stone boundary binary image, and separating the interface defects and matrix defects through image subtraction operations, where the interface defects are the defect regions overlapping with the block stone boundary, and the matrix defects are the defect regions independently distributed inside the matrix soil; A statistical output module for statistically calculating the pixel proportion of the interface defects and the matrix defects, calculating their volume proportion in the overall image of the soil-rock mixture, and outputting the quantitative detection result.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the steps of the method according to any one of claims 1 - 5.

Images