Determination method for soil porosity and aggregate in karst region
Generating a three-dimensional model of soil in karst areas through CT scanning and machine learning solves the problem that existing methods are difficult to accurately determine soil porosity and agglomerates, and achieves high-precision soil structure analysis.
Patent Information
- Application Number
- CN202510450179.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-11
- Publication Date
- 2025-07-18
AI Technical Summary
Existing soil porosity and agglomerate determination methods are difficult to accurately and quickly reflect the true structure of the soil in karst areas, and conventional methods may destroy the soil structure or ignore important information, resulting in low research accuracy.
CT scanning technology is used to digitally model soil samples in karst areas, and combined with machine learning and data processing software VOLUMEGRAPHICSSTUDIOMAX, a three-dimensional model is generated, and the soil structure is reconstructed through filtered backprojection and convolutional neural network, calculating gravel and root contents are achieved to achieve accurate determination of soil porosity and agglomerates.
Without destroying the soil structure, the accuracy of soil porosity and agglomeration determination in the Karst area is improved, and the internal structure of the soil can be observed from multiple angles, and the influence of gravel and root systems are comprehensively considered to generate a clear and realistic three-dimensional structural model.
Smart Images
Figure CN120334257A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of soil detection in karst areas, and specifically to a method for measuring soil porosity and aggregates in karst areas. Background Technique
[0002] Soil porosity and soil aggregates are important components of soil structure and function, and they have important effects on soil water, air and nutrient flow, as well as the growth of plant roots and microbial activities. Soil porosity and soil aggregates are affected by various factors such as soil-forming factors and external disturbances, and have strong spatial heterogeneity. Under natural conditions, they are affected by soil properties such as soil texture and soil structure. The more sand grains, the larger the soil porosity and the fewer the soil aggregates. For different regions, due to various factors such as soil structure, plant root types, and human production activities, there are many uncertainties in the measurement of soil porosity and soil aggregates. At present, the methods for measuring soil porosity include the saturated water method, the gas replacement method, and the double-ring infiltrometer method. These methods cannot distinguish pores of different sizes and are also greatly affected by environmental factors. Soil aggregates mainly use wet sieving, dry sieving, and dispersion coefficient methods. These methods may damage fragile aggregates and may also ignore the information of water-stable aggregates. Therefore, innovating and developing a high-precision method for measuring soil porosity and aggregates helps to obtain accurate soil physical property data, thus providing support for scientific research and soil agricultural production practice.
[0003] China is the country with the most extensive distribution of karst areas in the world. Even excluding the buried karst, the karst area developed from the outcropping of carbonate rocks alone reaches 1.3 million km 2 , and the outcropping area of surface karst alone in the southwestern region centered on Guizhou reaches 510,000 km 2 , which is the area with strong karst development and the most concentrated distribution in the world, and is a typical area with fragile ecological environment. Under the dissolution action of these carbonate rocks, it is extremely easy to form unique karst geological environments such as pores and aggregates in the shallow surface layer. In karst areas, the rock exposure rate is high, the gravel content is large, and the surface is rugged and the topography is complex and diverse. The complex karst ecosystem makes it difficult to study soil porosity and aggregates, and there are many uncertainties in the research accuracy. Therefore, there are few reports on methods for simultaneously and rapidly measuring soil porosity and aggregates in karst areas.
[0004] Due to the complex terrain in karst areas, and the loss of soil through underground pores (fractures) and the soil being rich in high - content gravel. Currently, the methods used by researchers to study the porosity and aggregates of karst area soils mainly include steel trough simulation, uniform stone stacking simulation, field small - watershed monitoring, etc. These methods all have the common drawback that they cannot truly reveal the original properties of the soil, resulting in relevant research being in the basic simulation stage and having a large difference from the actual situation. Therefore, we propose a method that does not damage the soil structure and can observe the internal structure of samples from different angles in three - dimensional space to accurately and quickly measure the soil porosity and aggregates in karst areas. Summary of the Invention
[0005] The technical problem to be solved by the present invention is to overcome the existing defects and provide a method for measuring the soil porosity and aggregates in karst areas. By digitally modeling the soil samples collected from karst areas, users can observe the internal structure of the samples from different angles in three - dimensional space to know the characteristics of soil aggregates and soil pores inside the samples, which is convenient for users to empower the soil in karst areas and can effectively solve the problems in the background technology.
[0006] To achieve the above object, the present invention provides the following technical solution: A method for measuring the soil porosity and aggregates in karst areas, including the following steps:
[0007] S1 Sample Scanning: Fix the prepared sample on the stage of the CT device, gently shake the sample by hand, observe whether the sample shakes accordingly to ensure the quality of the scanned image and reduce human error. After confirming that the sample is fixed, observe the position of the sample, adjust the stage so that the sample is in the center of the field of view, and adjust the parameters of the device, such as voltage, current, number of scan sheets, number of superimposed sheets, sensitivity, etc. When the scan resolution reaches the preset value, start scanning;
[0008] S2 Data Reconstruction: After the scanning is completed, select the automatic geometric correction to optimize the image, and adjust the values to reduce the influence caused by beam hardening. After the image quality of the representative layer reaches the standard, generate the positions of gravel and roots in the soil structure, and reconstruct the three - dimensional structure model based on the gravel and roots in the soil based on the filtered back - projection, Fourier transform, and convolutional neural network algorithms;
[0009] S3 Data Processing: Use machine learning to calculate the number of gravel and roots, generate a calculus function to correct the soil porosity and aggregates, and consider the inhomogeneity and depth changes of the soil to analyze and process the data of the reconstructed three - dimensional model using the data processing software VOLUMEGRAPH ICSSTUD IOMAX:
[0010] (1) Sample CT Scanning: Complete the CT scanning according to the sample scanning part in the experimental process;
[0011] (2) Sample digital model visualization: Through volume graph I CSSTUD IOMAX, using pictures and videos, the two-dimensional projection images and the overall three-dimensional model in the three directions of up-down, left-right, and front-back are displayed layer by layer;
[0012] (3) Three-dimensional internal display: By cutting, rotating or unfolding the sample through VOLUMEGRAPH I CSSTUDIOMAX, the internal structure of the sample can be observed from different angles in three-dimensional space to obtain the characteristics of soil aggregates and soil pores inside the sample.
[0013] Furthermore, the samples to be scanned in step (S1) need to be in-situ soil columns collected from designated karst areas, and the samples cannot be stored for more than one week before scanning.
[0014] Furthermore, after resampling, the sample soil column should be wrapped with tin foil to prevent its moisture from evaporating.
[0015] Furthermore, before scanning, the sample scanned in step (S1) needs to select a part of the soil column sample for analysis, and the analysis size is determined to be a 10mm×10mm area in the X and Y directions. In addition, since the interval between continuous slices is 0.5mm, 362 tiff images are selected in the Z direction to form a sample grid of 10mm×10mm×10mm.
[0016] Furthermore, in step (S2), when correcting and optimizing the image, since the original image is dark, the image contrast needs to be further enhanced. There are also a series of ring artifacts in the scanned original image, which need to be removed by polar coordinate conversion method (coordinate system conversion-Fourier transform-Butterworth filter-inverse Fourier transform-polar coordinate to rectangular coordinate) to make the original image meet the analysis needs.
[0017] Furthermore, after the image meets the analysis requirements, it is imported into ImageJ software and binarized according to the gray value map. Since different threshold segmentation methods have a great influence on the research results, this study uses a variety of methods for comparison and determines the maximum inter-class variance method (Otsu 2007) to perform binary segmentation on the image.
[0018] Furthermore, in step (S3), the CT scanning equipment radiation instrument is composed of a radiation source and a detector, which can obtain the internal structure, composition, material and defect status of the scanned sample, and use CT scanning visualization technology to analyze the changes in the applied soil structure characteristics, mainly including soil aggregate characteristics and soil pore characteristics.
[0019] High-resolution X-ray computed tomography (CT) technology is a method for rapidly obtaining the sectional structure of a sample using X-rays (X-rays, also known as Roentgen rays or X-rays, are a type of electromagnetic wave with a wavelength range between 0.01 nanometers and 10 nanometers); the X-ray instrument of the CT scanning device consists of a radiation source and a detector, which can accurately obtain two-dimensional tomographic images of an object, and accurately and intuitively obtain the internal structure, composition, material, and defect status of the object; the main working principle is that when X-rays with intensity I(x) pass through a thin layer of homogeneous medium with thickness dx, the rays will undergo absorption, scattering, or reflection, and their speed, frequency, etc. will change, and finally they will emerge with intensity I + dI. The physical principle of CT technology is based on the interaction between rays and matter; when the ray beam penetrates an object, due to the interaction between photons and matter, a considerable part of the incident photons are converted into electrons or disappear, so the ray intensity will weaken in the incident direction. When the rays are transmitted, the ray intensities passing through different types of minerals and different-sized storage spaces are different, and horizontal, longitudinal translation, and vertical lifting movements can be performed; when moving longitudinally, the closer to the X-ray source, the greater the magnification, the internal details are magnified, and the three-dimensional image is clearer, but at the same time the detectable area will be correspondingly reduced; on the contrary, the closer to the detector, the smaller the magnification, the lower the image resolution, but the detectable area increases. The horizontal translation and vertical lifting are used to change the scanning area but do not change the image resolution. The stage placed during the CT scanning of the sample itself can rotate. During CT scanning, the turntable drives the rotation. After each rotation of a small angle, a two-dimensional projection image at this position is obtained by X-ray irradiation. After rotating 360 degrees, a series of two-dimensional projection images can be obtained, and the two-dimensional projection images are data-recombined to obtain a three-dimensional image.
[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: The method for the soil pores and aggregates in this karst area has the following advantages:
[0021] The present invention collects undisturbed soil samples from the karst area, performs CT scanning and image processing on the collected soil samples after collection, and then generates a calculus function through machine learning to correct the data, generating a three-dimensional digital model. Through the three-dimensional digital model, the internal structure of the soil sample can be observed from different angles in three-dimensional space, and the characteristics of soil aggregates and soil pores inside the sample can be known. The influence of gravel content and roots on soil aggregates and pores in the karst area is comprehensively considered. It has the following beneficial effects:
[0022] (1) Without destroying the structure of the undisturbed soil, the present invention calculates the gravel content and root content in the soil by scanning the internal structure of the undisturbed soil column. Without destroying the undisturbed soil, using precise measuring tools and data analysis models, a calculus model for gravel and root content is generated through machine learning, improving the measurement accuracy. This method comprehensively considers multiple factors, including external factors such as gravel and plant roots in the soil, making the measurement more comprehensive. Through variance analysis and regression analysis, accurate measurement of soil porosity and aggregates is ensured.
[0023] (2) Aiming at the problem that the existing generation model cannot retain the distribution law of the pore characteristics of the original soil aggregates, the present invention comprehensively considers the non-uniformity and depth change of the soil, and makes an adaptive improvement to the generative adversarial network framework, which can automatically and quickly generate a large number of relatively clear, visually realistic and diverse three-dimensional structures of soil aggregate pores.
[0024] (3) The original generative adversarial network generates at one time, and it is easy to ignore the influence of many gravels and roots on the soil structure. This method is changed to a progressive generation process model. The distribution characteristics of the internal structure and data of the soil pore structure and aggregates show stronger overall regularity compared with ordinary RGB image data. The information content of a single pore structure sub-sample is relatively small, while the overall distribution of multiple sub-samples of the same soil type and at the same resolution shows a high degree of consistency. The progressive generation process can capture the overall distribution characteristics of the pore structure and aggregates at a specific resolution, so as to present the effect of the spatial heterogeneity of karst soil from multiple angles. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 is a flowchart of this method;
[0026] Figure 2 is a schematic diagram of the X-ray CT imaging chain of the present invention;
[0027] Figure 3 is a schematic diagram of the soil plane structure after the ray penetrates through the soil of the present invention;
[0028] Figure 4 is a diagram of the gravel content of the internal structure obtained after scanning the soil column, and the white quantity is the quantity diagram of the gravel;
[0029] Figure 5 is a distribution diagram of the root quantity obtained after scanning the soil column;
[0030] Figure 6 is a schematic diagram of the visualization, quantitative analysis and three-dimensional data recombination based on the soil gravel content and root images. DETAILED DESCRIPTION OF THE INVENTION
[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0032] This embodiment provides a technical solution: a method for measuring soil porosity and aggregates in karst areas, including the following steps:
[0033] S1 Sample Scanning: Fix the prepared sample on the stage of the CT device, gently shake the sample by hand, observe whether the sample shakes accordingly, ensure the quality of the scanned image, reduce human error. After confirming that the sample is fixed, observe the position of the sample, adjust the stage to make the sample in the center of the field of view, and adjust the parameters of the device, such as voltage, current, number of scans, number of overlays, sensitivity, etc. When the scanning resolution reaches the preset value, start scanning. The scanned sample is an in-situ soil column collected from a specified karst area. The storage time of the sample before scanning cannot exceed one week. After sampling, the soil column of the sample should be wrapped with tinfoil to avoid water evaporation. Before scanning, a part of the area needs to be selected from the soil column sample for analysis, and the analysis size is determined to be a 10 mm × 10 mm area in the X and Y directions. Additionally, since the continuous section interval is 0.5 mm, 362 tiff images are selected in the Z direction to form a 10 mm × 10 mm × 10 mm sample grid.
[0034] S2 Data Reconstruction: After scanning, select the automatic geometric correction to optimize the image, and adjust the values to reduce the influence caused by beam hardening. After the image quality of the representative layer reaches the standard, generate the positions of gravel and roots in the soil structure, and reconstruct a three-dimensional structure model based on gravel and roots in the soil by using algorithms such as filtered back projection, Fourier transform, and convolutional neural network. When correcting and optimizing the image, since the original image is darker, it is necessary to further enhance the image contrast. There are also a series of circular artifacts in the original scanned image, which need to be removed by the polar coordinate transformation method (coordinate system transformation - Fourier transform - Butterworth filtering - inverse Fourier transform - polar coordinate to rectangular coordinate) to make the original image meet the analysis requirements. After the image meets the analysis requirements, import the image into the ImageJ software, and perform binary processing on the image according to the grayscale value map. Since different threshold segmentation methods have a great influence on the research results, in this study, multiple methods are selected for comparison, and the maximum inter-class variance method (Otsu 2007) is determined to perform binary segmentation on the image.
[0035] S3 Data Processing: Use machine learning to calculate the number of gravel and roots, generate a calculus function to correct soil porosity and aggregates, and consider the soil inhomogeneity and depth changes. Use the data processing software VOLUMEGRAPH ICSSTUD IOMAX to analyze and process the reconstructed three-dimensional model data:
[0036] (1) CT Scanning of Samples: Complete the CT scanning according to the sample scanning part in the experimental process;
[0037] (2) Digital Model Visualization of Samples: Through VOLUMEGRAPHICSSTUDIOMAX, use pictures and videos to layer by layer display the two-dimensional projection images and the overall three-dimensional model in the up-down, left-right, and front-back three directions;
[0038] (3) Internal Display of Three-Dimensional Views: Through VOLUMEGRAPH ICSSTUDIOMAX, cut, rotate or unfold the samples, and observe the internal structure of the samples from different angles in three-dimensional space to know the soil aggregate characteristics and soil pore characteristics inside the samples;
[0039] The ray instrument of the CT scanning device consists of a radiation source and a detector, which is used to obtain the internal structure, composition, material and damage condition of the scanned sample. Using CT scanning visualization technology, analyze the changes in the characteristics of the applied soil structure, mainly including soil aggregate characteristics and soil pore characteristics.
[0040] Testing Principle:
[0041] The physical principle of CT technology is based on the interaction between rays and matter. When the ray beam penetrates an object, due to the interaction between photons and matter, a considerable part of the incident photons are converted into electrons or disappear, so the ray intensity will be weakened in the incident direction. As Figure 1 shown, for different densities (A - B) and different thicknesses (B - C), the attenuation degree of the rays is different.
[0042] When the rays are radiographed, the ray intensities passing through different types of minerals and different-sized reservoir spaces are different. According to the above characteristics, design a CT imaging chain as Figure 2 shown. The X-ray source and the detector are respectively placed on both sides of the turntable. The X-ray penetrates the object placed on the turntable and is then received by the detector. The detector generates images with different gray-scale signals, and the components and internal reservoir spaces are analyzed through the different gray-scales of the images.
[0043] It can perform horizontal, vertical translation and vertical lifting movements. When moving longitudinally, the closer it is to the X-ray source, the greater the magnification, the internal details are magnified, and the three-dimensional image is clearer. However, the detectable area will be correspondingly reduced; on the contrary, the closer it is to the detector, the smaller the magnification and the lower the image resolution, but the detectable area increases. Horizontal translation and vertical lifting are used to change the scanning area without changing the image resolution.
[0044] The placed stage itself can rotate. During CT scanning, the turntable drives the rotation. After each small rotation angle, a two-dimensional projection image at this position is obtained by X-ray irradiation. After rotating 360 degrees, a series of two-dimensional projection images can be obtained. The two-dimensional projection images are data-recombined to obtain a three-dimensional image, as Figure 3 shown Figure 3 In it, 1-4 respectively correspond to the two-dimensional projection planes in the 1-4 directions. By recombining 4 two-dimensional projection images, the three-dimensional spatial position information of the color square can be obtained.
[0045] The above are only the embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.
Claims
1. A method for measuring soil porosity and aggregates in karst areas, characterized in that: It includes the following steps: S1 Sample Scanning: Fix the prepared sample on the stage of the CT device, gently shake the sample by hand, observe whether the sample shakes accordingly, ensure the quality of the scanned image, reduce human error. After confirming that the sample is fixed, observe the position of the sample, adjust the stage to make the sample in the center of the field of view, and adjust the parameters of the device, such as voltage, current, number of scans, number of overlays, sensitivity, etc. When the scanning resolution reaches the preset value, start scanning; S2 Data Reconstruction: After the scanning is completed, select the automatic geometric correction to optimize the image, and adjust the values to reduce the influence caused by beam hardening. After the image quality of the representative layer meets the standard, generate the positions of gravel and roots in the soil structure, and reconstruct the three-dimensional structure model of gravel and roots in the soil based on the filtered back projection, Fourier transform, and convolutional neural network algorithms; S3 Data Processing: Use machine learning to calculate the number of gravel and roots, generate a calculus function to correct the soil porosity and aggregates, and consider the inhomogeneity and depth variation of the soil. Use the data processing software VOLUMEGRAPHICS STUDIOMAX to analyze and process the data of the reconstructed three-dimensional model: (1) Sample CT Scanning: Complete the CT scanning according to the sample scanning part in the experimental process; (2) Sample Digital Model Visualization: Through VOLUMEGRAPHICS STUDIOMAX, use pictures and videos to layer by layer display the two-dimensional projection images and the overall three-dimensional model in the up-down, left-right, and front-back three directions; (3) Internal Display of the Three-Dimensional View: Through VOLUMEGRAPHICS STUDIOMAX, cut, rotate, or unfold the sample, and observe the internal structure of the sample from different angles in the three-dimensional space to know the characteristics of soil aggregates and soil pores inside the sample.
2. The method for measuring soil porosity and aggregates in a karst area according to claim 1, wherein: The sample for sample scanning in step (S1) needs to be the in-situ soil column collected from the designated karst area, and the storage time of the sample before scanning cannot exceed one week.
3. The method for measuring soil porosity and aggregates in a karst area according to claim 2, characterized in that: After the sample soil column is resampled, wrap the soil column with tin foil paper to avoid water evaporation.
4. The method for measuring soil porosity and aggregates in a karst area according to claim 2, characterized in that: Before scanning the sample in step (S1), a part of the area needs to be selected from the soil column sample for analysis. Determine that the analysis size is a 10 mm × 10 mm area in the X and Y directions. In addition, since the continuous section interval is 0.5 mm, 362 tiff pictures are selected in the Z direction to form a 10 mm × 10 mm × 10 mm sample grid.
5. The method for measuring soil porosity and aggregates in a karst area according to claim 1, characterized in that: When correcting and optimizing the image in step (S2), since the original image is darker, it is necessary to further enhance the image contrast. There are also a series of circular artifacts in the scanned original image, and the circular artifacts need to be removed by the polar coordinate transformation method (coordinate system transformation - Fourier transform - Butterworth filtering - inverse Fourier transform - polar coordinate to rectangular coordinate) to make the original image meet the analysis requirements.
6. The method for measuring soil porosity and aggregates in a karst area according to claim 5, characterized in that: After the image meets the analysis requirements, it is imported into ImageJ software and binarized according to the gray value map. Since different threshold segmentation methods have a great influence on the research results, this study uses a variety of methods for comparison and selection, and determines the maximum inter-class variance method (Otsu 2007) to perform binary segmentation on the image.
7. A method for measuring soil porosity and aggregates in a karst area according to claim 1, characterized in that: In step (S3), the CT scanning equipment radiation instrument is composed of a radiation source and a detector, which obtains the internal structure, composition, material and defect status of the scanned sample, and uses CT scanning visualization technology to analyze the changes in the applied soil structure characteristics, mainly including soil aggregate characteristics and soil pore characteristics.