Transparent soil model analysis method and system and detection method

Abstract

The invention provides a transparent soil model analysis method and system and a detection method, and the method comprises the following steps: building a transparent soil-based initial test model having a test association relationship with a test target, and carrying out the distribution analysis of the initial test model, obtaining a first form distribution diagram composed of a plurality of initial distribution areas; performing a deformation experiment of the corresponding transparent soil on the initial test model based on any test item, and performing distribution analysis on the obtained current test model to obtain a second form distribution diagram composed of a plurality of current distribution areas; performing image comparison on the first form distribution diagram and the second form distribution diagram, and determining a region deviation value of an initial distribution region and a current distribution region corresponding to each same region position based on a comparison result; and generating a current result publicity graph corresponding to the test item based on all the region deviation values. The test accuracy is at least improved.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a transparent soil model analysis method, system and detection method. Background Art

[0002] In geotechnical engineering, soil deformation testing is a critical step in ensuring project safety and stability. Whether it's road construction, bridge construction, or high-rise building foundation design, it's crucial to accurately understand soil deformation characteristics under load, environmental changes, and other factors. Soil deformation testing can reveal key parameters like soil compressibility and shear strength, providing a reliable basis for engineering design. This allows for the rational planning of foundation treatment solutions and optimized structural design, avoiding problems such as building settlement and tilting, or road pavement cracking caused by excessive soil deformation. This effectively reduces the risk of engineering accidents, protects people's lives and property, and contributes to improving the economic efficiency and sustainability of engineering construction.

[0003] Currently, there are two main methods for soil deformation testing. One is traditional laboratory testing methods, such as confined compression tests and triaxial compression tests. These methods apply loads to soil samples under controlled laboratory conditions and measure their deformation. However, these methods have significant limitations. The soil samples obtained are not fully representative of the actual properties of the soil in situ, are subject to disturbance effects, and the test conditions differ from actual engineering environments, limiting the accuracy and applicability of the test results. The other is in situ testing methods, such as static cone penetration tests and standard penetration tests. While these methods can provide a certain degree of insight into the properties of the soil in situ, they often only capture partial parameters and fail to fully capture the soil's deformation process and internal structural changes. Furthermore, in situ testing is costly and inefficient, and testing in complex geological conditions can be challenging, including operational difficulties and data interpretation challenges. Furthermore, both laboratory and in situ testing methods struggle to visually observe the internal mechanisms of soil deformation, failing to meet the in-depth research needs of modern geotechnical engineering.

[0004] Therefore, there is an urgent need for a transparent soil model analysis method, system and detection method that can improve test accuracy and intuitively observe the internal structure changes of soil deformation. Summary of the Invention

[0005] Based on the above problems, the present invention is proposed to provide a transparent soil model analysis method, system and detection method that overcome the above problems or at least partially solve the above problems.

[0006] According to one aspect of the present invention, a transparent soil model analysis method is provided, comprising the following steps: Building an initial test model based on transparent soil that has a test association relationship with the test target, and performing distribution analysis on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas; Based on any test item, a deformation experiment of the transparent soil is performed on the initial test model, and a distribution analysis is performed on the obtained current test model to obtain a second morphological distribution map consisting of multiple current distribution areas; Performing an image comparison between the first morphological distribution map and the second morphological distribution map, and determining a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position based on the comparison result; A current result display diagram corresponding to the test item is generated based on all regional offset values.

[0007] Optionally, in the method according to the present invention, an initial test model based on transparent soil having a test association relationship with the test target is constructed, and a distribution analysis is performed on the initial test model to obtain a first morphological distribution map consisting of multiple initial distribution areas, including: Determine the test specifications corresponding to the test target, and fill the test cavity with transparent soil corresponding to the test specifications to obtain an initial test model; The control acquisition unit acquires images of the initial test model and performs image recognition on the obtained front view of the model to determine a soil area located in the front view of the model and indicating transparent soil; forming a rectangular outline circumscribing the soil area based on the front view of the model, and dividing the initial test model into an array corresponding to the rectangular outline to obtain initial spot areas; A distribution analysis is performed on each initial spot area to obtain a first morphological distribution map consisting of a plurality of initial distribution areas.

[0008] Optionally, in the method according to the present invention, distribution analysis is performed on each initial spot area to obtain a first morphological distribution map consisting of multiple initial distribution areas, including: Controlling the irradiation unit to irradiate light corresponding to the test pixel value toward the center point of each initial spot area, and controlling the acquisition unit to acquire an image of the initial spot area based on a frontal perspective to obtain a frontal image of the area; performing pixelation processing on the regional front image, and determining each image pixel point located in the regional front image, except for the image pixel point corresponding to the test pixel value, as a spot pixel point; An image coordinate system of the front view of the corresponding region is established with the center point of the image corresponding to the front view of the region as the origin, and the initial spot region is divided into regions based on the front view angle based on the X-axis and Y-axis of the corresponding image coordinate system to obtain each divided sub-region; Based on the distribution analysis of all the spot pixels included in each divided sub-region, a first morphological distribution map consisting of a plurality of initial distribution regions is obtained.

[0009] Optionally, in the method according to the present invention, a first morphological distribution map consisting of a plurality of initial distribution areas is obtained based on a distribution analysis of all spot pixels included in each divided sub-area, including: Performing pixel connections based on adjacent positions on all spot pixels located in the same divided sub-region based on the regional front view, and summarizing the obtained spot points into a regional division group corresponding to the divided sub-region; In response to the presence of different spot points having a connection relationship in the plurality of divided sub-regions, determining the point area corresponding to each spot point, and removing all spot points except the spot point with the largest point area from the area division group of the corresponding divided sub-region; generating a first initial distribution map, wherein the first initial distribution map includes an initial distribution area corresponding to each divided sub-area; The number of sub-regions of all the spot points in each region division group is determined, and the number of each sub-region is filled into the corresponding initial distribution region to obtain a first morphological distribution map.

[0010] Optionally, in the method according to the present invention, the method further comprises: Grouping all initial spot regions in a horizontal direction and a vertical direction based on the rectangular outline to obtain horizontal region groups and vertical region groups; Merging all initial spot regions in the same horizontal region group and vertical region group, and controlling the acquisition unit to acquire images of the obtained horizontal merged region and vertical merged region based on the side view and the vertical view to obtain a regional side view and a regional vertical view; Based on the first morphological distribution map, a lateral distribution area having a lateral relationship with each initial distribution area and a vertical distribution area having a vertical relationship with each initial distribution area are formed to convert each initial distribution area from an initial two-dimensional form to a three-dimensional form; Dividing the initial spot area into regions based on the side view based on the regional side view, and filling the number of sub-regions of all spot points included in each divided sub-region into the side distribution region corresponding to the initial spot area located in the same horizontal region group; The initial spot area is divided into regions based on the vertical view angle based on the regional vertical map, and the number of sub-regions of all spot points included in each divided sub-region is filled into the vertical distribution area corresponding to the initial spot area located in the same vertical area group.

[0011] Optionally, in the method according to the present invention, performing image comparison on the first morphological distribution map and the second morphological distribution map, and determining, based on the comparison result, a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position, includes: Performing transparency processing on the image portions other than the number of each sub-region in the first morphological distribution map and the second morphological distribution map, and overlapping the obtained first processed map and second processed map; In response to any number of sub-regions in the first processing image completely overlapping with any number of sub-regions in the corresponding same region position in the second processing image, the region position is determined as a coincidence attribute, otherwise it is determined as an offset attribute; In response to any region position being an offset attribute, determining the number of subregions corresponding to the region position in the first processed image as a first comparison number, determining the number of subregions corresponding to the region position in the second processed image as a second comparison number, and calculating based on a difference between the first comparison number and the second comparison number to obtain a region offset value corresponding to the region position; In response to any region position being a matching attribute, the region offset value corresponding to the region position is determined to be 0.

[0012] Optionally, in the method according to the present invention, generating a current result display diagram corresponding to the test item based on all regional offset values ​​includes: generating an initial result public graph, wherein the initial result public graph includes an offset public region corresponding to each initial spot region; Establishing a regional coordinate system for the offset public area with the center point of the area corresponding to the offset public area as the origin, and dividing the offset public area based on the X-axis and Y-axis of the corresponding regional coordinate system to obtain each publicized sub-area; Based on the regional coordinate system and the image coordinate system, a digital association relationship is established between each public sub-region and the regional offset value corresponding to the same regional position; The area offset values ​​with digital associations in each publicized sub-area located in the same offset publicized area are aggregated into an offset pointing group, and an offset pointer located in the corresponding offset publicized area is generated based on each offset pointing group to obtain the current result publicized map.

[0013] Optionally, in the method according to the present invention, generating an offset pointer located in a corresponding offset disclosure area based on each offset pointing group includes: Determine the maximum regional offset value corresponding to the same offset pointing group as the ending offset value, and the minimum regional offset value as the starting offset value; and determine the regional center point of the public sub-region that has a numerical association with the ending offset value as the ending pointing point, and determine the regional center point of the public sub-region that has a numerical association with the starting offset value as the starting pointing point; Based on each offset orientation group, the largest regional offset value is determined as the maximum offset value, the corresponding minimum regional offset value is determined as the minimum offset value, and an offset interval consisting of the minimum offset value and the maximum offset value is determined; Dividing the offset interval into a number of intervals corresponding to a preset number of intervals, and configuring each obtained offset sub-interval corresponding to a different specification multiple; generating an offset pointer corresponding to a preset symbol specification and pointing from the starting pointing point to the ending pointing point, and performing an average calculation on the offset values ​​of all regions in the same offset pointing group; The offset mean obtained in response is located in any offset sub-interval, and the offset pointer corresponding to the offset pointing group is updated with a specification multiple corresponding to the offset sub-interval.

[0014] Optionally, in the method according to the present invention, the method further comprises: Determine any offset public announcement area as a primary public announcement area, and determine each other offset public announcement area surrounding the primary public announcement area as a secondary public announcement area; Determine the total number of secondary areas corresponding to all secondary public display areas, and multiply the total number of secondary areas by the retrieved preset offset ratio to obtain an allowable offset number; Comparing the direction of the offset pointer located in the primary public display area with the direction of the offset pointer located in each secondary public display area to obtain the direction offset degree corresponding to each secondary public display area; In response to the number of directional offsets corresponding to all directional offsets greater than the preset offset, which is greater than the allowed offset number, the main-level public area is marked as abnormal.

[0015] According to another aspect of the present invention, there is provided a transparent soil model analysis system, comprising: a first analysis module configured to construct an initial test model based on transparent soil having a test association relationship with a test target, and perform distribution analysis on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas; A second analysis module is configured to perform a deformation experiment of the transparent soil corresponding to the initial test model based on any test item, and perform a distribution analysis on the obtained current test model to obtain a second morphological distribution map consisting of multiple current distribution areas; an analysis and comparison module configured to perform an image comparison between the first morphological distribution map and the second morphological distribution map, and determine, based on the comparison result, a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position; The result display module is configured to generate a current result display diagram corresponding to the test item based on all regional offset values.

[0016] According to another aspect of the present invention, a method for performing compliance testing on the initial test model constructed above is provided, comprising the following steps: Dividing the initial test model into arrays based on the horizontal and vertical directions to obtain comparison areas; Controlling the irradiation unit to irradiate light corresponding to the test pixel value in the region extension direction of each comparison region, and controlling the acquisition unit to acquire an image of each comparison region to obtain a region comparison map; Determine each test pixel point located in each regional comparison image and having the same pixel value as the test pixel value, and determine other pixel points except the test pixel points as comparison pixel points; Determine any regional comparison graph as a first comparison graph, and determine a regional comparison graph corresponding to a region below the first comparison graph as a second comparison graph based on the initial test model; The first comparison image and the second comparison image are compared based on the comparison pixels, and a model state corresponding to the initial test model is determined based on the comparison result.

[0017] Optionally, in the method according to the present invention, performing image comparison based on the comparison pixels on the first comparison image and the second comparison image, and determining the model state corresponding to the initial test model based on the comparison result, includes: Calculating a ratio of a first number of pixels corresponding to all comparison pixel points located in the first comparison image to a second number of pixels corresponding to all comparison pixel points located in the second comparison image, and determining a quantity comparison value of a corresponding quantity dimension based on the obtained quantity ratio; Placing the first comparison image on top of the second comparison image in such a manner that a center point of a first image of the first comparison image coincides with a center point of a second image of the second comparison image, thereby obtaining an overlapping comparison image; Obtaining the number of overlapping pixels corresponding to all comparison pixel points located in the overlap comparison image, and determining the largest one of the first number of pixels and the second number of pixels as the deviation from the reference number; Calculating a ratio of the deviation reference number to the number of overlapping pixels, and determining a deviation contrast value corresponding to the deviation dimension based on the obtained deviation ratio; The quantity comparison value and the deviation comparison value are weightedly summed up and calculated, and if the real-time detection value obtained in response is greater than a preset detection value, the model state is determined to be a qualified state, otherwise it is determined to be a defective state.

[0018] According to the solution of the present invention, firstly, in the process of building an initial test model based on transparent soil with a test correlation relationship with the test target and obtaining a first morphological distribution map, the optical transparency of the transparent soil is fully utilized, which can more realistically simulate the mechanical properties of the actual soil. Compared with traditional soil sample testing, the result deviation caused by soil sample disturbance and test condition differences is greatly reduced. The initial test model is subjected to a detailed distribution analysis to obtain a first morphological distribution map composed of multiple initial distribution areas, which provides an accurate initial data reference for subsequent tests and makes the data basis of the entire test process more reliable. Secondly, based on the test item, a deformation experiment is carried out on the initial test model to obtain a second morphological distribution map, which can closely simulate the soil deformation condition around the actual engineering needs. Taking advantage of transparent soil, which allows for intuitive observation of internal changes in the soil, this method can more comprehensively and accurately record soil deformation under different conditions and obtain rich deformation data, compared to in-situ testing, which is difficult to fully present the deformation process. Next, the first and second morphological distribution maps are compared and the regional offset value is determined. This quantitatively reflects the difference between the soil before and after deformation, allowing researchers and engineers to clearly grasp the specific location, extent, and trend of soil deformation, providing strong data support for in-depth research on soil deformation laws. Finally, a current result public diagram is generated based on the regional offset value, presenting complex data in a visual form. Compared with the rough data processing and visualization methods of traditional test, this public diagram can clearly and intuitively display subtle changes in soil deformation. Researchers and engineers can quickly obtain key information, facilitate accurate assessment of the impact of soil deformation on engineering structures, optimize engineering design plans, take targeted measures to prevent risks in advance, effectively reduce the probability of engineering accidents, ensure the safety and stability of the project, and improve the corresponding test accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a transparent soil model analysis method according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of a deviation disclosure area corresponding to a deviation mean is shown; Figure 3 A schematic diagram of an offset disclosure area corresponding to another offset mean is shown; Figure 4 A structural block diagram of a transparent soil model analysis system according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0020] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0021] To address the aforementioned problems in the prior art, the inventors have proposed the present invention. One embodiment of the present invention provides a transparent soil model analysis method that can be executed on a computing device, where the computing device can be understood as a terminal with data processing capabilities, such as a mobile phone or computer.

[0022] Figure 1 FIG. 4 shows a flow chart of a transparent soil model analysis method according to an embodiment of the present invention, wherein Figure 1 As shown, the method starts at step S1, and in step S1, includes the following contents: An initial test model based on transparent soil having a test association relationship with a test target is constructed, and a distribution analysis is performed on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas.

[0023] For example, in this embodiment, in order to conduct deformation tests on the corresponding soil based on transparent soil, it is necessary to pre-build an initial test model based on transparent soil that has a test association with the test target. Here, it can be explained that transparent soil is a special soil material. Because of its optical transparency, researchers can intuitively observe the changes inside the soil, which greatly improves the visualization and accuracy of the test compared to traditional soil tests. When building the initial test model, it is necessary to select suitable transparent soil materials according to the specific needs of the test target, such as the deformation of the test soil under specific loads and environmental conditions, and construct the model according to a certain method. For example, according to the stress distribution characteristics of the target test, the proportion, compaction degree and other parameters of the transparent soil are accurately controlled to make the initial test model simulate the actual soil as realistically as possible. The mechanical properties of the soil are determined by the initial test model, thereby ensuring that the test results can accurately reflect the actual situation and improving the reliability and effectiveness of the test. After the initial test model is built, it is necessary to perform a distribution analysis on it to obtain a first morphological distribution map composed of multiple initial distribution areas. It can be shown that the first morphological distribution map can clearly and intuitively display the characteristic distribution of the initial test model in each area, such as density distribution, stress distribution, etc. By observing these distributions, researchers can quickly understand the initial state of the model, providing important basic data for subsequent further research on the deformation law of the soil, helping researchers to more accurately grasp the deformation trend of the soil under different conditions, and then provide a reliable basis for related engineering design and construction, effectively avoiding engineering accidents caused by soil deformation problems, and ensuring the safety and stability of the project.

[0024] Furthermore, in this embodiment, the above-mentioned "building an initial test model based on transparent soil having a test association relationship with the test target, and performing distribution analysis on the initial test model to obtain a first morphological distribution map consisting of multiple initial distribution areas" may also include the following steps: Determine the test specifications corresponding to the test target, and fill the test cavity with transparent soil corresponding to the test specifications to obtain an initial test model; The control acquisition unit acquires images of the initial test model and performs image recognition on the obtained front view of the model to determine a soil area located in the front view of the model and indicating transparent soil; forming a rectangular outline circumscribing the soil area based on the front view of the model, and dividing the initial test model into an array corresponding to the rectangular outline to obtain initial spot areas; A distribution analysis is performed on each initial spot area to obtain a first morphological distribution map consisting of a plurality of initial distribution areas.

[0025] For example, in this embodiment, in a deformation test of a corresponding soil based on transparent soil, the specific process of building an initial test model based on transparent soil having a test association relationship with the test target and obtaining a first morphological distribution map is as follows: First, the test specifications corresponding to the test target need to be determined. Here, since the deformation of the test soil under different loads and different environmental humidity requires different test specifications, it is necessary to clarify the test specifications corresponding to the test target. After the test specifications are clarified, transparent soil of the corresponding specifications can be filled into the test cavity. Based on the above content, it can be seen that the physical and chemical properties of transparent soil can simulate the characteristics of real soil under the corresponding working conditions. During the filling process, by controlling the filling uniformity, density and other parameters, an initial test model that can accurately reflect the test target is obtained, making the subsequent test results more authentic and valuable for reference, laying a reliable foundation for soil deformation research. Next, the acquisition unit is controlled to acquire images of the initial test model. Here, the acquisition unit may be an optical camera that can clearly capture the appearance information of the initial test model. After the acquisition of the front view of the model is completed, image recognition technology can be used to analyze and process it to accurately determine the soil area located in the front view of the model and indicating transparent soil, thereby avoiding errors caused by manual judgment, greatly improving the accuracy and reliability of data acquisition, and enabling subsequent analysis to be carried out based on accurate data. Then, a rectangular outline circumscribing the soil area is formed based on the model front view. It can be shown that the rectangular outline can completely cover the soil area, facilitating subsequent unified division processing; Then, the initial test model can be divided into an array corresponding to the rectangular outline, that is, the area within the rectangular outline is divided into multiple small initial spot areas according to certain rules. Based on the array division, the analysis of the initial test model can be more detailed and comprehensive, and subtle differences in different positions of the model can be captured, which helps to gain a deeper understanding of the distribution of soil characteristics. Finally, a distribution analysis is performed on each initial spot area. For example, the physical properties of each initial spot area, such as density, are quantitatively analyzed to obtain a first morphological distribution map composed of multiple initial distribution areas. Here, the first morphological distribution map intuitively presents the characteristic distribution status of each area in the initial test model. Researchers can quickly locate areas with abnormal characteristics in the model by observing the distribution map, providing key initial data support for subsequent research on the deformation laws of soil under different conditions, helping researchers to more accurately predict soil deformation trends, and providing a scientific and effective basis for engineering design and construction, thereby improving the project's ability to deal with soil deformation problems and ensuring the safety and stability of the project.

[0026] Furthermore, in this embodiment, the above-mentioned “performing a distribution analysis on each initial spot area to obtain a first morphological distribution map consisting of a plurality of initial distribution areas” may further include the following steps: Controlling the irradiation unit to irradiate light corresponding to the test pixel value toward the center point of each initial spot area, and controlling the acquisition unit to acquire an image of the initial spot area based on a frontal perspective to obtain a frontal image of the area; performing pixelation processing on the regional front image, and determining each image pixel point located in the regional front image, except for the image pixel point corresponding to the test pixel value, as a spot pixel point; An image coordinate system of the front view of the corresponding region is established with the center point of the image corresponding to the front view of the region as the origin, and the initial spot region is divided into regions based on the front view angle based on the X-axis and Y-axis of the corresponding image coordinate system to obtain each divided sub-region; Based on the distribution analysis of all the spot pixels included in each divided sub-region, a first morphological distribution map consisting of a plurality of initial distribution regions is obtained.

[0027] For example, in this embodiment, when conducting soil deformation testing based on transparent soil, a distribution analysis is performed on each initial spot area to obtain a first morphological distribution map. The specific steps are as follows: First, the irradiation unit is controlled to irradiate the corresponding test pixel value in a manner of irradiating the area toward the center point of each initial spot area. It can be explained that the irradiation unit here can use a controllable light source. By accurately irradiating the light toward the center point of the area, it can ensure that the initial spot area is evenly illuminated, thereby avoiding test data deviation caused by uneven illumination. At the same time, the acquisition unit is controlled to acquire an image of the initial spot area based on a frontal perspective to obtain a frontal map of the area. Here, the acquisition unit acquires the image at a frontal perspective, which can minimize the error caused by perspective deformation, thereby obtaining image data that truly reflects the state of the initial spot area, providing reliable original data for subsequent analysis, and ensuring the accuracy and credibility of the analysis results. Next, it can be explained that when the transparent soil located in the initial test model is illuminated by light corresponding to the test pixel value, due to its unique optical transparency, the light corresponding to the test pixel value will undergo a change in the corresponding pixel value when passing through the transparent soil. Therefore, the transparent soil can be identified based on the difference in pixel values. That is, the front view of the region can be pixelated to decompose the image into individual pixels, and each image pixel located in the front view of the region except for the image pixel corresponding to the test pixel value can be further determined as a spot pixel (i.e., a pixel corresponding to the transparent soil). This can effectively screen out key pixel information related to the test target and eliminate interference from irrelevant pixels, allowing subsequent analysis to focus on core data, greatly improving analysis efficiency and accuracy, and avoiding the influence of redundant data on the judgment of transparent soil characteristics. Subsequently, an image coordinate system of the front view of the corresponding area is established with the center point of the image corresponding to the front view of the area as the origin. Based on the X-axis and Y-axis of the image coordinate system, the initial spot area is divided into regions based on the front view to obtain each sub-region. It can be shown that by establishing a coordinate system for precise division, the initial spot area can be systematically decomposed into multiple sub-regions with clear position identifications, which facilitates researchers to conduct targeted analysis of each sub-region, helps to deeply explore the characteristic differences of transparent soil at different locations and fully understand its distribution law. Finally, based on the above content, it can be seen that the spot pixels are the pixels corresponding to the transparent soil. Therefore, a distribution analysis can be performed based on all the spot pixels included in each divided sub-area. For example, these spot pixels can be quantitatively analyzed and feature extracted to obtain a first morphological distribution map composed of multiple initial distribution areas. Here, it can be explained that the first morphological distribution map presents the characteristic distribution of transparent soil in the initial spot area in a visual manner. Researchers can use this map to quickly identify the changing trends of the density, stress and other characteristics of transparent soil in different areas, providing intuitive and accurate data support for in-depth research on soil deformation laws, helping researchers to more scientifically evaluate soil deformation risks in actual projects, optimize engineering design plans, effectively reduce the probability of engineering accidents caused by soil deformation, and ensure safe and stable operation of projects.

[0028] Furthermore, in this embodiment, the above-mentioned “performing a distribution analysis on all the spot pixels included in each divided sub-region to obtain a first morphological distribution map consisting of a plurality of initial distribution regions” may further include the following steps: Performing pixel connections based on adjacent positions on all spot pixels located in the same divided sub-region based on the regional front view, and summarizing the obtained spot points into a regional division group corresponding to the divided sub-region; In response to the presence of different spot points having a connection relationship in the plurality of divided sub-regions, determining the point area corresponding to each spot point, and removing all spot points except the spot point with the largest point area from the area division group of the corresponding divided sub-region; generating a first initial distribution map, wherein the first initial distribution map includes an initial distribution area corresponding to each divided sub-area; The number of sub-regions of all the spot points in each region division group is determined, and the number of each sub-region is filled into the corresponding initial distribution region to obtain a first morphological distribution map.

[0029] For example, in this embodiment, when conducting soil deformation testing based on transparent soil, obtaining a first morphological distribution map from the spot pixels of the divided sub-regions can be specifically implemented based on the following method steps: First, based on the front view of the region, all the spot pixels located in the same divided sub-region are connected based on adjacent positions. It can be shown that in this way, discrete spot pixels can be connected in series according to the spatial position relationship to form a coherent spot point set, and then the obtained spot points are aggregated into the regional division group corresponding to the divided sub-region. This operation enables the originally scattered pixel information to be integrated and classified, providing an orderly data structure for subsequent systematic analysis, facilitating researchers to quickly locate and process key data in each divided sub-region, avoiding data confusion, and improving analysis efficiency. Next, when different spot points with a connection relationship are detected in multiple divided sub-areas, the point area corresponding to each spot point is determined, and all spot points except the spot point with the largest point area are removed from the area division group of the corresponding divided sub-area, thereby effectively solving the problem of data overlap or fuzzy attribution between adjacent divided sub-areas. In actual testing, due to the complexity of transparent soil characteristics, there may be blurred boundaries between different areas. Therefore, by determining the spot point with the largest point area and removing all spot points except the spot point from the area division group of the corresponding divided sub-area, the uniqueness and accuracy of the data can be guaranteed, so that the data of each area can more accurately reflect the true characteristics of the transparent soil in the area, avoid analysis deviation due to data duplication, and further improve the reliability of the test results. Subsequently, a first initial distribution map is generated. The map includes initial distribution areas corresponding to each sub-region. This initial distribution map establishes a preliminary visualization framework, presenting the distribution of each sub-region in an intuitive graphical form. This allows researchers to gain a holistic understanding of the approximate distribution of the transparent soil in the initial test model. Compared to simply listing data, this allows researchers to more quickly grasp the overall picture and provides a clear guide for subsequent in-depth analysis. Finally, the number of sub-regions of all the spot points in each regional division group can be determined, and the number of each sub-region can be filled into the corresponding initial distribution area to obtain the first morphological distribution map. Here, by quantifying the number of spot point sub-regions in each area and filling them into the distribution map, the originally abstract data can be converted into visual graphic parameters, so that the first morphological distribution map can not only show the distribution range of transparent soil, but also intuitively reflect the degree of change and differences in transparent soil characteristics in different areas through the difference in the number of sub-regions. By observing the first morphological distribution map, researchers can quickly locate areas where transparent soil characteristics change significantly, providing an accurate visualization basis for in-depth research on soil deformation laws, helping researchers to more accurately assess soil deformation risks in actual projects, formulate more scientific and reasonable engineering design plans, effectively reduce the hidden dangers of engineering accidents caused by soil deformation, and ensure the safe and stable operation of the project.

[0030] In addition, in this embodiment, the method further includes: Grouping all initial spot regions in a horizontal direction and a vertical direction based on the rectangular outline to obtain horizontal region groups and vertical region groups; Merging all initial spot regions in the same horizontal region group and vertical region group, and controlling the acquisition unit to acquire images of the obtained horizontal merged region and vertical merged region based on the side view and the vertical view to obtain a regional side view and a regional vertical view; Based on the first morphological distribution map, a lateral distribution area having a lateral relationship with each initial distribution area and a vertical distribution area having a vertical relationship with each initial distribution area are formed to convert each initial distribution area from an initial two-dimensional form to a three-dimensional form; Dividing the initial spot area into regions based on the side view based on the regional side view, and filling the number of sub-regions of all spot points included in each divided sub-region into the side distribution region corresponding to the initial spot area located in the same horizontal region group; The initial spot area is divided into regions based on the vertical view angle based on the regional vertical map, and the number of sub-regions of all spot points included in each divided sub-region is filled into the vertical distribution area corresponding to the initial spot area located in the same vertical area group.

[0031] For example, in this embodiment, during soil deformation testing based on transparent soil, in order to more comprehensively and three-dimensionally understand the distribution of transparent soil characteristics, after obtaining the first morphological distribution map, the three-dimensional information of the model can be further improved. The specific steps are as follows: First, based on the rectangular outlines obtained above, all initial spot areas can be grouped in the horizontal and vertical directions to obtain horizontal and vertical area groups. Then, the scattered initial spot areas can be classified in an orderly manner according to their spatial position relationships, making the originally messy regional data clear and organized, avoiding data confusion, and improving analysis efficiency. At the same time, it also lays the foundation for studying the characteristics of transparent soil from multiple dimensions. Next, all the initial spot areas in the same horizontal area group and vertical area group are merged, and the acquisition unit is controlled to acquire images of the obtained horizontal merged area and vertical merged area based on the side view and the vertical view, so as to obtain the side view of the area and the vertical view of the area. It can be shown that the area merging operation can reduce the complexity of data processing, so that the acquisition unit can more efficiently obtain image information from different perspectives. The image acquisition from the side view and the vertical view makes up for the limitation of observing only from the front view, and obtains the morphological characteristics and characteristic distribution information of the transparent soil in different spatial dimensions, so that researchers can have a more comprehensive understanding of the spatial structure of the transparent soil and improve the integrity and accuracy of the test results. Subsequently, based on the first morphological distribution map, lateral distribution areas with lateral relationships and vertical distribution areas with vertical relationships with each initial distribution area are formed to convert each initial distribution area from its initial two-dimensional form to a three-dimensional form. Here, by expanding the distribution information on the two-dimensional plane to three-dimensional space, the characteristic distribution of the transparent soil is made more three-dimensional and intuitive. Compared with simple two-dimensional analysis, three-dimensional form can more realistically simulate the spatial structure of actual soil, helping researchers discover characteristic variation patterns that may be overlooked from a two-dimensional perspective, providing a more realistic model reference for in-depth research on soil deformation mechanisms, and improving the scientificity and practicality of the research. Finally, based on the regional side view, the initial spot area is divided based on the side perspective, and the number of sub-areas of all spot points included in each divided sub-area is filled into the side distribution area corresponding to the initial spot area in the same horizontal area group; at the same time, the initial spot area can also be divided based on the vertical perspective based on the regional vertical view, and the number of sub-areas of all spot points included in each divided sub-area is filled into the vertical distribution area corresponding to the initial spot area in the same vertical area group. Here, the operations of region division and data filling based on images of different perspectives can give specific quantitative information to each part of the three-dimensional morphology, so that the side distribution area and the vertical distribution area can accurately reflect the characteristic differences and change trends of transparent soil in different spatial directions. Through these detailed three-dimensional distribution information, researchers can more accurately analyze the deformation law of soil when subjected to stress in different directions, provide a more reliable basis for the rational planning of foundation treatment, slope protection, etc. in engineering design, effectively reduce the engineering risks caused by soil deformation, and ensure the safety and stability of engineering structures.

[0032] In step S2, the following contents are included: Based on any test item, a deformation experiment of the corresponding transparent soil is performed on the initial test model, and a distribution analysis is performed on the obtained current test model to obtain a second morphological distribution map composed of multiple current distribution areas.

[0033] For example, in this embodiment, when performing soil deformation testing based on transparent soil, after completing the construction of the initial test model and obtaining the first morphological distribution map, it is necessary to further explore its deformation characteristics under actual working conditions. At this time, the initial test model can be subjected to a deformation experiment corresponding to the transparent soil based on any test item. It can be explained that multiple different test items can be included for the same test target. These test items can simulate various conditions that soil may face in real engineering scenarios, such as the application of loads of different intensities, etc. By applying specific test conditions to the initial test model, it causes corresponding deformation. Based on the above content, it can be seen that the optical transparency of transparent soil plays a key role in this process. It allows researchers to directly observe the occurrence and development of internal deformation of the soil. Compared with traditional opaque soil tests, it greatly improves the intuitiveness and accuracy of deformation observation; and after completing the deformation experiment to obtain the current test model, it can be subjected to distribution analysis. Here, the distribution analysis of the current test model can adopt similar technical means to the analysis of the initial test model, and then obtain a second morphological distribution map composed of multiple current distribution areas.

[0034] It can be explained that the second form distribution map shows the distribution of transparent soil characteristics after deformation during the test. In the subsequent process, by comparing the first and second form distribution maps, researchers can clearly and intuitively see the deformation differences of transparent soil under different test conditions, including the location of the deformation area, changes in size, and changes in characteristic parameters. This helps to deeply understand the deformation patterns of soil under actual working conditions and provide accurate data support for engineering design and construction. For example, in the design of building foundations, based on these distribution maps, the foundation structure can be more reasonably planned, avoiding problems such as building settlement and tilting caused by soil deformation, effectively ensuring project safety, reducing project risks and maintenance costs, and also providing a valuable reference basis for subsequent similar projects, improving the entire engineering field's research and response capabilities to soil deformation issues.

[0035] In step S3, the following contents are included: An image comparison is performed between the first morphological distribution map and the second morphological distribution map, and based on the comparison result, a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position is determined.

[0036] For example, in this embodiment, in order to accurately quantify the degree of soil deformation during soil deformation testing based on transparent soil, an in-depth analysis of the morphological distribution before and after the test is required. That is, an image comparison is required between the first morphological distribution map obtained based on the initial test model and the second morphological distribution map corresponding to the current test model after the deformation experiment. It can be explained that since these two distribution maps intuitively present the characteristic distribution state of the transparent soil before and after the test, the two can be compared, and then the regional offset value between the initial distribution area and the current distribution area corresponding to each identical regional position can be determined based on the comparison results. It can be explained that the regional offset value can be mainly used to measure the deformation displacement of the transparent soil in each region. Based on these values, researchers can quickly locate the area with significant deformation, analyze the deformation trend of different regions, and further explore the deformation mechanism of the soil under different working conditions. In actual engineering applications, such data can help researchers more scientifically evaluate the stability of engineering structures such as foundations and slopes, provide a strong basis for optimizing engineering design and preventing safety hazards caused by soil deformation, effectively reduce engineering risks, and ensure the safe and stable operation of the project. At the same time, it also provides a more reliable data foundation for research and development in the field of soil mechanics.

[0037] Furthermore, in this embodiment, the above-mentioned “performing image comparison between the first morphological distribution map and the second morphological distribution map, and determining the regional offset value between the initial distribution area and the current distribution area corresponding to each identical regional position based on the comparison result” may further include the following steps: Performing transparency processing on the image portions other than the number of each sub-region in the first morphological distribution map and the second morphological distribution map, and overlapping the obtained first processed map and second processed map; In response to any number of sub-regions in the first processing image completely overlapping with any number of sub-regions in the corresponding same region position in the second processing image, the region position is determined as a coincidence attribute, otherwise it is determined as an offset attribute; In response to any region position being an offset attribute, determining the number of subregions corresponding to the region position in the first processed image as a first comparison number, determining the number of subregions corresponding to the region position in the second processed image as a second comparison number, and calculating based on a difference between the first comparison number and the second comparison number to obtain a region offset value corresponding to the region position; In response to any region position being a matching attribute, the region offset value corresponding to the region position is determined to be 0.

[0038] For example, in this embodiment, in a soil deformation test based on transparent soil, in order to accurately obtain the changes in soil property distribution before and after the test, it is necessary to carefully compare the first morphological distribution map with the second morphological distribution map and determine the regional offset value. The specific implementation process can be described as follows: First, the image portions of the first and second morphological distribution maps, except for the number of each sub-region, are made transparent. This means that image editing techniques are used to remove interfering information in the image that is unrelated to the number of sub-regions, such as background patterns and borders. This allows subsequent analysis to focus on the key data—the number of sub-regions. This effectively avoids redundant information from interfering with the comparison results, improving analysis efficiency and accuracy. Next, by overlapping the first and second processed images obtained after processing, the two images are placed in the same reference coordinate system, creating conditions for accurate comparison, and then the attributes of each regional position can be determined. If the number of any sub-regions located in the first processed image completely overlaps with the number of any sub-regions located in the corresponding same regional position in the second processed image, the regional position is determined as a matching attribute, otherwise it is determined as an offset attribute. This allows for rapid and intuitive identification of areas where the distribution of transparent soil properties has not changed and has changed before and after the test, providing a clear classification for subsequent targeted analysis, allowing researchers to quickly lock in key areas of focus and saving a lot of analysis time. Finally, the regional offset value can be calculated for regional positions with different attributes, for example, including: in response to any regional position being an offset attribute, the number of sub-regions corresponding to the regional position in the first processing diagram is determined as a first comparison number, and the number of sub-regions corresponding to the regional position in the second processing diagram is determined as a second comparison number, and based on the difference between the first comparison number and the second comparison number, the regional offset value corresponding to the regional position is obtained, that is, by quantifying the difference in the number of sub-regions, the degree of change in the distribution of transparent soil characteristics is converted into a specific numerical value, so as to provide researchers with detailed and accurate deformation data, and help them to deeply analyze the laws and causes of soil deformation; and in response to any regional position To match the properties, the regional offset value corresponding to the regional position is determined to be 0, which simply and clearly indicates that the characteristic distribution of the region has not changed before and after the test, ensuring the integrity and logic of the data; so that the final regional offset value can comprehensively and accurately reflect the changes in the characteristic distribution of transparent soil before and after the test. In actual engineering applications, it can help researchers accurately evaluate the impact of soil deformation on engineering structures, such as predicting foundation settlement and slope displacement trends, thereby optimizing engineering design plans and taking preventive measures in advance, effectively reducing the risk of engineering accidents caused by soil deformation, and ensuring the safety and stability of the project. At the same time, it also provides a reliable quantitative basis for soil mechanics research and promotes the further development of research in related fields.

[0039] In step S4, the following contents are included: A current result display diagram corresponding to the test item is generated based on all regional offset values.

[0040] For example, in this embodiment, after conducting soil deformation tests based on transparent soil, in order to convert the abstract regional offset values ​​into intuitive and easy-to-understand visualization results to assist engineering practice and academic research, it is necessary to generate a current result display diagram of the corresponding test items based on all regional offset values.

[0041] Furthermore, in this embodiment, the above-mentioned “generating a current result display diagram corresponding to the test item based on all regional offset values” may further include the following steps: generating an initial result public graph, wherein the initial result public graph includes an offset public region corresponding to each initial spot region; Establishing a regional coordinate system for the offset public area with the center point of the area corresponding to the offset public area as the origin, and dividing the offset public area based on the X-axis and Y-axis of the corresponding regional coordinate system to obtain each publicized sub-area; Based on the regional coordinate system and the image coordinate system, a digital association relationship is established between each public sub-region and the regional offset value corresponding to the same regional position; The area offset values ​​with digital associations in each publicized sub-area located in the same offset publicized area are aggregated into an offset pointing group, and an offset pointer located in the corresponding offset publicized area is generated based on each offset pointing group to obtain the current result publicized map.

[0042] For example, in this embodiment, after conducting a soil deformation test based on transparent soil, generating a current result display diagram of the corresponding test item based on all regional offset values ​​can be achieved through the following specific process: First, an initial result diagram is generated. It contains offset display areas corresponding to each initial spot area, building a basic visualization framework that maps initial spot areas to corresponding offset display areas. This provides a clear visualization platform for the previously scattered regional offset value data, allowing researchers to quickly establish a connection between data and graphics, and form a preliminary understanding of the overall test results. Compared to simply listing data, this greatly improves the efficiency of information acquisition. Next, a regional coordinate system corresponding to the offset public area is established with the regional center point corresponding to the offset public area as the origin, and the offset public area is divided into regions based on the X-axis and Y-axis of the corresponding regional coordinate system to obtain each publicized sub-region. In other words, by establishing a precise regional coordinate system and dividing the publicized sub-regions, the offset public area is further refined, providing an orderly spatial structure for subsequent precise labeling and analysis of regional offset values. Each publicized sub-region has a clear location identifier, which facilitates researchers to conduct targeted research on deformation conditions at different locations, avoids data confusion caused by ambiguous regional divisions, and improves the accuracy and reliability of analysis. Subsequently, based on the regional coordinate system and the image coordinate system, a numerical correlation relationship is established between each public sub-region and the regional offset value corresponding to the same regional position. It can be shown that by combining the two coordinate systems, the accurate docking of the public sub-region and the actual deformation data is achieved, ensuring that the regional offset value displayed by each public sub-region truly reflects the deformation of its corresponding actual position, thereby providing accurate data support for subsequent visualization, enabling researchers to intuitively obtain deformation information of each position through graphics, and improving the readability and usability of the data; Finally, the regional offset values ​​of each public sub-area with a numerical association in the same offset public area are summarized into an offset pointing group, and an offset pointer located in the corresponding offset public area is generated based on each offset pointing group to obtain the current result public map. It can be explained that in this embodiment, by summarizing the regional offset values ​​to form an offset pointing group and then generating an offset pointer, the abstract numerical data can be converted into an intuitive graphic identifier. Here, the offset pointer can clearly show the deformation trend and degree difference in each offset public area, so that researchers can quickly locate the area with significant deformation by observing the current result public map, and intuitively compare the deformation conditions of different areas. In actual engineering applications, this public map can help researchers quickly evaluate the impact of soil deformation on engineering structures, provide an intuitive and reliable basis for engineering decisions such as foundation reinforcement and slope protection, effectively reduce the engineering risks caused by soil deformation, and ensure the safe and stable operation of the project. At the same time, it also provides an efficient visualization analysis tool for soil deformation related research.

[0043] Furthermore, in this embodiment, the above-mentioned "generating an offset pointer located in a corresponding offset public area based on each offset pointer group" may further include the following steps: Determine the maximum regional offset value corresponding to the same offset pointing group as the ending offset value, and the minimum regional offset value as the starting offset value; and determine the regional center point of the public sub-region that has a numerical association with the ending offset value as the ending pointing point, and determine the regional center point of the public sub-region that has a numerical association with the starting offset value as the starting pointing point; Based on each offset orientation group, the largest regional offset value is determined as the maximum offset value, the corresponding minimum regional offset value is determined as the minimum offset value, and an offset interval consisting of the minimum offset value and the maximum offset value is determined; Dividing the offset interval into a number of intervals corresponding to a preset number of intervals, and configuring each obtained offset sub-interval corresponding to a different specification multiple; generating an offset pointer corresponding to a preset symbol specification and pointing from the starting pointing point to the ending pointing point, and performing an average calculation on the offset values ​​of all regions in the same offset pointing group; The offset mean obtained in response is located in any offset sub-interval, and the offset pointer corresponding to the offset pointing group is updated with a specification multiple corresponding to the offset sub-interval.

[0044] For example, in this embodiment, after performing soil deformation testing based on transparent soil, in order to more intuitively display the deformation of each area, it is necessary to generate an offset pointer based on the offset pointer group. The specific implementation method can be based on the following content: First, the corresponding maximum regional offset value in the same offset pointing group can be determined as the end offset value, and the corresponding minimum regional offset value can be determined as the start offset value. The regional center point of the public sub-region with a numerical correlation relationship with the end offset value can be determined as the end pointing point, and the regional center point of the public sub-region with a numerical correlation relationship with the start offset value can be determined as the start pointing point. Here, it can be explained that by clarifying the start and end offset values ​​and their corresponding pointing points, the basic direction and endpoints of the offset pointer are determined, and a clear logical starting point is provided for subsequent visualization, allowing researchers to quickly understand the deformation range represented by each offset pointing group, avoiding confusion during data visualization, and improving the accuracy of information transmission. Next, based on each offset orientation group, the largest regional offset value can be determined as the maximum offset value, the corresponding smallest regional offset value can be determined as the minimum offset value, and an offset interval consisting of the minimum offset value and the maximum offset value can be determined. It can be explained that determining the offset interval provides a range for subsequent quantitative analysis of deformation degree, providing a unified comparison benchmark for deformation degrees of different offset orientation groups, thereby enhancing the readability and usability of the data. Subsequently, the offset interval is divided into intervals corresponding to the preset number of intervals, and each obtained offset sub-interval is configured with corresponding specification multiples. Here, the offset interval is subdivided and the specification multiples are configured to convert the continuous deformation data into discrete and clearly distinguishable levels, so that the offset pointer can intuitively display the difference in deformation degree through different specifications. When viewing the current result display map in actual engineering, researchers do not need to check the numerical values ​​in detail, but can quickly judge the severity of soil deformation in each area based on the specifications of the offset pointer, which greatly improves the data visualization effect and information acquisition efficiency. Next, an offset pointer corresponding to a preset symbol specification is generated, pointing from the starting pointing point to the ending pointing point. The mean offset value of all regional offsets in the same offset pointing group is calculated. The initially generated offset pointer provides a visual representation of the deformation direction, while the calculated offset mean further explores the average characteristics of regional deformation. This allows multiple data points to be condensed into a representative value, facilitating subsequent comparison with offset subintervals. This also lays the foundation for accurately adjusting the offset pointer specifications, ensuring that the visualization results truly reflect the actual regional deformation. Finally, when the offset mean value obtained in response is in any offset sub-interval, the offset pointer corresponding to the offset pointing group is updated with the specification multiple corresponding to the offset sub-interval. The offset pointer specifications are dynamically updated according to the interval where the offset mean value is located, and accurate mapping of deformation data and visualization graphics is achieved. In the current result display diagram, offset pointers of different specifications can clearly show the differences in soil deformation in various regions. Researchers can quickly locate areas with significant deformation and analyze soil deformation laws by intuitively comparing the offset pointer specifications. For researchers, this visualization result can help them more scientifically evaluate the impact of soil deformation on engineering structures, thereby formulating more reasonable engineering plans, effectively reducing engineering risks, and ensuring safe and stable operation of the project. It also provides an efficient and intuitive analysis tool for soil deformation research.

[0045] For example, in this embodiment, Figure 2 as well as Figure 3 A schematic diagram showing the offset pointers located in different offset disclosure areas generated when the corresponding offset mean values ​​are located in different offset subintervals is shown, where it can be seen that: Figure 2 The public sub-areas A1, B1, C1 and D1 are included, wherein the offset pointer is from C1 to A1, and Figure 3 It includes public sub-areas A2, B2, C2 and D2, wherein the offset pointer points from B2 to D2, and Figure 3 The included offset pointer is significantly smaller than Figure 2 The specifications of the included offset pointers (since each offset pointer points from the starting pointing point to the ending pointing point, the specification update corresponding to the specification multiple specifically refers to the width specification of the offset pointer), therefore, it can be directly determined that the offset means corresponding to the two are in different offset sub-intervals, and Figure 2 The corresponding offset mean should be greater than Figure 3 The corresponding offset mean.

[0046] In addition, in this embodiment, the method further includes: Determine any offset public announcement area as a primary public announcement area, and determine each other offset public announcement area surrounding the primary public announcement area as a secondary public announcement area; Determine the total number of secondary areas corresponding to all secondary public display areas, and multiply the total number of secondary areas by the retrieved preset offset ratio to obtain an allowable offset number; Comparing the direction of the offset pointer located in the primary public display area with the direction of the offset pointer located in each secondary public display area to obtain the direction offset degree corresponding to each secondary public display area; In response to the number of directional offsets corresponding to all directional offsets greater than the preset offset, which is greater than the allowed offset number, the main-level public area is marked as abnormal.

[0047] For example, in this embodiment, in a soil deformation test based on transparent soil, in order to quickly identify abnormal soil deformation areas and ensure the reliability of the test results, it is necessary to further analyze and judge the offset public area. The specific implementation method can be based on the following content: First, each displacement disclosure area is identified as a primary disclosure area, and each other displacement disclosure area surrounding the primary disclosure area is identified as a secondary disclosure area. This constructs a hierarchical analysis structure, with the corresponding primary disclosure area serving as the core analysis object and forming an association with the surrounding secondary disclosure areas. This facilitates comprehensive analysis of soil deformation from both local and global perspectives, improves analysis efficiency, and helps researchers grasp key information more quickly. Next, the total number of sub-levels corresponding to all sub-publicized areas is determined, and the product of the total number of sub-levels and the retrieved preset offset ratio is calculated to obtain the allowable offset number. It can be explained that the preset offset ratio is a reasonable threshold set based on a large amount of experimental data and engineering experience. By calculating the allowable offset number, a quantitative standard is provided for subsequent judgment, making the judgment of soil deformation more objective and scientific, avoiding subjective arbitrariness. In actual engineering applications, researchers can use this standard to quickly assess whether the deformation of a certain area is within a reasonable range, providing a reliable basis for engineering decision-making. Subsequently, the direction of the offset pointer located in the primary public area is compared with the direction of the offset pointer located in each secondary public area to obtain the directional offset of each secondary public area. Here, it can be explained that the offset pointer intuitively shows the direction of soil deformation. By comparing the directions of the offset pointers in the primary and secondary public areas, the differences in deformation directions between areas can be accurately captured. This difference analysis helps to discover abnormal trends in soil deformation. For example, under normal circumstances, the soil deformation directions in adjacent areas should have a certain correlation. If there is a significant directional offset, it may indicate potential geological problems. Finally, in response to the directional offset number corresponding to all directional offsets greater than the preset offset number being greater than the allowable offset number, the main-level public area is marked as abnormal. That is, when the actual directional offset number exceeds the allowable offset number, it indicates that the soil deformation direction of the main-level public area and its surrounding areas is not in line with the norm. At this time, the abnormal marking can quickly attract the attention of researchers and engineers. In the actual engineering site, this abnormal marking can help staff to promptly discover areas that may be at risk, take measures in advance to reinforce or prevent them, effectively avoid engineering accidents caused by soil deformation, and ensure the safety and stability of the project. At the same time, for soil mechanics research, these abnormal markings also provide important clues for further exploring the laws of soil deformation under special geological conditions, and promote the in-depth development of research in related fields.

[0048] It can be explained that the abnormal identification of the main-level public announcement area can be carried out, for example, based on pixel values, that is, pixel identification of the main-level public announcement area; or an abnormal symbol for abnormal identification can be generated in the main-level public announcement area. The above-mentioned implementation methods can all correspond to the implementation of abnormal identification of the main-level public announcement area. This embodiment does not specifically limit the actual method adopted.

[0049] To sum up, based on the technical content disclosed in this embodiment, first, in the process of building an initial test model based on transparent soil with a test correlation relationship with the test target and obtaining the first morphological distribution map, the optical transparency of the transparent soil is fully utilized, which can more realistically simulate the mechanical properties of the actual soil. Compared with traditional soil sample testing, it greatly reduces the result deviation caused by soil sample disturbance and test condition differences. The initial test model is carefully analyzed for distribution, and a first morphological distribution map composed of multiple initial distribution areas is obtained, which provides an accurate initial data reference for subsequent tests, making the data basis of the entire test process more reliable; secondly, based on the test item, a deformation experiment is performed on the initial test model to obtain the second morphological distribution map, which can closely simulate the soil deformation condition around the actual engineering needs. Taking advantage of transparent soil, which allows for intuitive observation of internal changes in the soil, this method can more comprehensively and accurately record soil deformation under different conditions and obtain rich deformation data, compared to in-situ testing, which is difficult to fully present the deformation process. Next, the first and second morphological distribution maps are compared and the regional offset value is determined. This quantitatively reflects the difference between the soil before and after deformation, allowing researchers and engineers to clearly grasp the specific location, extent, and trend of soil deformation, providing strong data support for in-depth research on soil deformation laws. Finally, a current result public diagram is generated based on the regional offset value, presenting complex data in a visual form. Compared with the rough data processing and visualization methods of traditional test, this public diagram can clearly and intuitively display subtle changes in soil deformation. Researchers and engineers can quickly obtain key information, facilitate accurate assessment of the impact of soil deformation on engineering structures, optimize engineering design plans, take targeted measures to prevent risks in advance, effectively reduce the probability of engineering accidents, ensure the safety and stability of the project, and improve the corresponding test accuracy.

[0050] Another embodiment of the present invention provides a transparent soil model analysis system, Figure 4 For its corresponding system block diagram, the system includes: a first analysis module configured to construct an initial test model based on transparent soil having a test association relationship with a test target, and perform distribution analysis on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas; A second analysis module is configured to perform a deformation experiment of the transparent soil corresponding to the initial test model based on any test item, and perform a distribution analysis on the obtained current test model to obtain a second morphological distribution map consisting of multiple current distribution areas; an analysis and comparison module configured to perform an image comparison between the first morphological distribution map and the second morphological distribution map, and determine, based on the comparison result, a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position; The result display module is configured to generate a current result display diagram corresponding to the test item based on all regional offset values.

[0051] Another embodiment of the present invention provides a method for performing compliance testing on the initial test model constructed above, wherein the method includes: Dividing the initial test model into arrays based on the horizontal and vertical directions to obtain comparison areas; Controlling the irradiation unit to irradiate light corresponding to the test pixel value in the region extension direction of each comparison region, and controlling the acquisition unit to acquire an image of each comparison region to obtain a region comparison map; Determine each test pixel point located in each regional comparison image and having the same pixel value as the test pixel value, and determine other pixel points except the test pixel points as comparison pixel points; Determine any regional comparison graph as a first comparison graph, and determine a regional comparison graph corresponding to a region below the first comparison graph as a second comparison graph based on the initial test model; The first comparison image and the second comparison image are compared based on the comparison pixels, and a model state corresponding to the initial test model is determined based on the comparison result.

[0052] For example, in this embodiment, based on the above content, it can be seen that the initial test model is constructed based on the test target. In order to perform corresponding compliance detection on the constructed initial test model to determine whether the test conditions are met, it can be specifically implemented based on the following technical content: First, the initial test model can be divided into a horizontal / vertical array to obtain various comparison areas. It can be shown that by dividing the initial test model, the initial test model can be orderly disassembled into multiple independent and clearly located comparison areas, allowing subsequent evaluation to be carried out one by one for different areas, avoiding omissions or misjudgments due to mixed areas, laying the foundation for a comprehensive and detailed evaluation of the model, and ensuring that the evaluation results cover all parts of the model; Next, the irradiation unit is controlled to irradiate light corresponding to the test pixel value in the direction of the area extension of each comparison area, and the acquisition unit is controlled to acquire an image of each comparison area to obtain a regional comparison map. Here, irradiating light along the direction of the area extension can fully utilize the optical properties of the transparent soil, so that the light evenly penetrates the comparison area, reducing the impact of uneven light irradiation on image quality. At the same time, the acquisition unit can accurately capture the image information after the light interacts with the transparent soil. The obtained regional comparison map can truly reflect the state of the comparison area, provide a clear and reliable image basis for subsequent evaluation, and improve the accuracy of the evaluation; Subsequently, each test pixel point located in the comparison image of each region and having the same pixel value as the test pixel value can be further determined, and the other pixel points except the test pixel points can be determined as comparison pixel points. Here, the test pixel points can be used as a benchmark reference for evaluation, while the comparison pixel points can reflect the actual characteristics of the transparent soil in the comparison area. By clearly distinguishing these two types of pixel points, irrelevant pixel interference can be eliminated, allowing the evaluation to focus on key information, making subsequent image comparison more targeted, and improving evaluation efficiency and accuracy; Afterwards, any regional comparison graph is determined as the first comparison graph, and based on the initial test model, the regional comparison graph corresponding to the region below the first comparison graph is determined as the second comparison graph. The selection of the corresponding comparison graph based on the spatial position of the initial test model allows the evaluation to focus on regions at different levels of the model, conforms to the structural characteristics of the model, facilitates the analysis of the vertical consistency of the model, avoids evaluation bias caused by unreasonable selection of comparison regions, and ensures that the evaluation can reflect the overall structural status of the model; Finally, the first and second comparison images are compared based on the comparison pixels, and the model state corresponding to the initial test model is determined based on the comparison results. It can be shown here that by comparing the differences in the comparison pixels in the two types of comparison images, it is possible to clearly determine whether the transparent soil characteristics of different hierarchical regions of the model are consistent and whether there are any anomalies, thereby determining whether the model meets the standards. For example, if the difference in the comparison pixels is small, it indicates that the characteristics of each region of the model are uniform and meet the test requirements; if the difference is large, it indicates that the model is defective and needs to be rebuilt. This allows for the rapid and accurate evaluation of the quality of the initial test model, ensuring the reliability of subsequent soil deformation test results based on the model, providing a solid foundation for engineering applications, avoiding test data errors caused by substandard models, and thus causing engineering decision-making errors, thereby ensuring the scientific nature and safety of the project.

[0053] Furthermore, in this embodiment, the above-mentioned “performing image comparison on the first comparison image and the second comparison image based on the comparison pixels, and determining the model state corresponding to the initial test model based on the comparison result” may further include the following steps: Calculating a ratio of a first number of pixels corresponding to all comparison pixel points located in the first comparison image to a second number of pixels corresponding to all comparison pixel points located in the second comparison image, and determining a quantity comparison value of a corresponding quantity dimension based on the obtained quantity ratio; Placing the first comparison image on top of the second comparison image in such a manner that a center point of a first image of the first comparison image coincides with a center point of a second image of the second comparison image, thereby obtaining an overlapping comparison image; Obtaining the number of overlapping pixels corresponding to all comparison pixel points located in the overlap comparison image, and determining the largest one of the first number of pixels and the second number of pixels as the deviation from the reference number; Calculating a ratio of the deviation reference number to the number of overlapping pixels, and determining a deviation contrast value corresponding to the deviation dimension based on the obtained deviation ratio; The quantity comparison value and the deviation comparison value are weightedly summed up and calculated, and if the real-time detection value obtained in response is greater than a preset detection value, the model state is determined to be a qualified state, otherwise it is determined to be a defective state.

[0054] For example, in this embodiment, when evaluating whether the constructed initial test model meets the standards, the first comparison image and the second comparison image are compared based on the comparison pixels, and the specific process of determining the model status based on the comparison result is as follows: First, a ratio of the number of first pixels corresponding to all comparison pixels in the first comparison image to the number of second pixels corresponding to all comparison pixels in the second comparison image can be calculated, and a quantitative comparison value of the corresponding quantitative dimension can be determined based on the obtained quantitative ratio. It can be shown that by quantifying the quantitative relationship between the comparison pixels in the two layers of comparison images, the distribution differences of transparent soil characteristics in different hierarchical areas can be intuitively reflected from a quantitative dimension. This quantification method avoids the subjectivity of judgment based solely on visual observation, making the evaluation more objective and accurate, and providing a reliable quantitative dimension basis for subsequent comprehensive judgment. Next, the first comparison image is placed on top of the second comparison image in such a way that the center point of the first image of the first comparison image coincides with the center point of the second image of the second comparison image, thereby obtaining an overlapped comparison image. This means that the two comparison images are spatially aligned by aligning their centers, ensuring the consistency of the spatial reference for subsequent calculations of the overlap of comparison pixel points, reducing comparison errors caused by positional offsets, and making the statistics of the number of overlapping pixels more accurate, thus laying the foundation for the consistency of the evaluation model in spatial distribution. Subsequently, the number of overlapping pixels corresponding to all comparison pixels located in the overlap comparison image is obtained, and the largest of the first and second pixel numbers is determined as the deviation from the baseline. Selecting the largest number of pixels as the deviation baseline provides a more rigorous measure of the degree of overlap of the comparison pixels. Even when the number of comparison pixels in one layer is large, the degree of deviation can be accurately reflected by the ratio to the baseline, avoiding underestimation or overestimation of the deviation due to improper baseline selection. Afterwards, the ratio of the deviation benchmark number to the number of overlapping pixels can be calculated, and the deviation comparison value of the corresponding deviation dimension can be determined based on the obtained deviation ratio. It can be explained that this process quantifies the overlap of the contrasting pixels in the two layers of comparison images from the deviation dimension of spatial distribution, which can effectively reflect the consistency of the spatial distribution of transparent soil characteristics in different hierarchical areas, supplement the spatial distribution information not covered by the quantitative dimension, and make the evaluation dimension more comprehensive. Finally, the quantity comparison value and the deviation comparison value are weighted and summed, and if the real-time detection value obtained in response is greater than the preset detection value, the model state is determined to be in a qualified state; otherwise, it is determined to be in a defective state. It can be explained that the weighted summation method combines the evaluation results of the quantity dimension and the deviation dimension, which is more comprehensive than the judgment of a single dimension. It can take into account both quantity differences and spatial distribution consistency, ensuring that the evaluation results are more in line with the actual state of the model. The preset detection value is an objective threshold set based on the model compliance standard. By comparing it with the real-time detection value, it can clearly and accurately define whether the model meets the standard, avoiding the ambiguity of subjective judgment, and providing a clear and reliable basis for quality control of the initial test model, ensuring that the subsequent soil deformation test results based on the qualified model are true and valid, and guaranteeing the scientific nature and accuracy of the test from the source.

[0055] For example, the calculation formula for the quantity comparison value may be specifically: quantity comparison value = 2×min(first pixel number, second pixel number) / (first pixel number + second pixel number) (wherein the first pixel number is the number of all comparison pixels in the first comparison image, the second pixel number is the number of all comparison pixels in the second comparison image, and min() represents the smaller of the two values). It can be seen that the smaller the difference between the first pixel number and the second pixel number, the closer the ratio of the min value to the sum of the two numbers is to 1, and the larger the quantity comparison value. At the same time, the calculation formula of the deviation contrast value can be specifically as follows: deviation contrast value = deviation base number / overlapping pixel number (wherein, the overlapping pixel number is the overlapping pixel number of all contrast pixel points of the overlapping contrast image, and the deviation base number is the largest one of the first pixel number and the second pixel number). It can be explained that when the difference between the deviation base number and the overlapping pixel number is smaller, that is, the closer the deviation base number is to the overlapping pixel number, the closer the ratio of the two is to 1, the greater the deviation contrast value.

[0056] In the description provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems may also be used in conjunction with the examples of the present invention. Based on the above description, it is apparent that the structure required for constructing such systems is well understood. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​may be utilized to implement the present invention described herein, and the description of specific languages ​​above is provided for the purpose of disclosing preferred embodiments of the present invention.

[0057] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0058] Similarly, it should be understood that in order to streamline the disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0059] Those skilled in the art will appreciate that the modules, units, or components of the devices in the examples disclosed herein may be arranged in the device described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or further divided into multiple submodules.

[0060] Those skilled in the art will appreciate that the modules in the devices of the embodiments can be adaptively changed and installed in one or more devices different from the embodiments. The modules, units, or components in the embodiments can be combined into one module, unit, or component, and furthermore, they can be divided into multiple submodules, subunits, or subcomponents.

[0061] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features and not other features included in other embodiments, the combination of features from different embodiments is intended to be within the scope of the invention and to form different embodiments.

[0062] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions described. Thus, a processor having the necessary instructions for implementing the method or method element forms a device for implementing the method or method element. Furthermore, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0063] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved and are not intended to imply that the objects so described must have a given order in time, space, ranking, or in any other manner.

[0064] Although the present invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of the foregoing description, will appreciate that other embodiments are contemplated within the scope of the invention thus described. Furthermore, it should be noted that the language used in this specification has been selected primarily for readability and instructional purposes and is not selected to explain or limit the subject matter of the present invention.

Claims

1. A transparent soil model analysis method, characterized in that: The following steps are involved: Building an initial test model based on transparent soil that has a test association relationship with the test target, and performing distribution analysis on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas; Based on any test item, a deformation experiment of the transparent soil is performed on the initial test model, and a distribution analysis is performed on the obtained current test model to obtain a second morphological distribution map consisting of multiple current distribution areas; Performing an image comparison between the first morphological distribution map and the second morphological distribution map, and determining a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position based on the comparison result; A current result display diagram corresponding to the test item is generated based on all regional offset values.

2. The analysis method according to claim 1, characterized in that An initial test model based on transparent soil having a test association relationship with the test target is constructed, and a distribution analysis is performed on the initial test model to obtain a first morphological distribution map consisting of multiple initial distribution areas, including: Determine the test specifications corresponding to the test target, and fill the test cavity with transparent soil corresponding to the test specifications to obtain an initial test model; The control acquisition unit acquires images of the initial test model and performs image recognition on the obtained front view of the model to determine a soil area located in the front view of the model and indicating transparent soil; forming a rectangular outline circumscribing the soil area based on the front view of the model, and dividing the initial test model into an array corresponding to the rectangular outline to obtain initial spot areas; A distribution analysis is performed on each initial spot area to obtain a first morphological distribution map consisting of a plurality of initial distribution areas.

3. The analysis method according to claim 2, characterized in that Perform distribution analysis on each initial spot area to obtain a first morphological distribution map consisting of multiple initial distribution areas, including: Controlling the irradiation unit to irradiate light corresponding to the test pixel value toward the center point of each initial spot area, and controlling the acquisition unit to acquire an image of the initial spot area based on a frontal perspective to obtain a frontal image of the area; performing pixelation processing on the regional front image, and determining each image pixel point located in the regional front image, except for the image pixel point corresponding to the test pixel value, as a spot pixel point; An image coordinate system of the front view of the corresponding region is established with the center point of the image corresponding to the front view of the region as the origin, and the initial spot region is divided into regions based on the front view angle based on the X-axis and Y-axis of the corresponding image coordinate system to obtain each divided sub-region; Based on the distribution analysis of all the spot pixels included in each divided sub-region, a first morphological distribution map consisting of a plurality of initial distribution regions is obtained.

4. The analysis method according to claim 3, characterized in that Based on the distribution analysis of all the spot pixels included in each divided sub-region, a first morphological distribution map consisting of multiple initial distribution regions is obtained, including: Performing pixel connections based on adjacent positions on all spot pixels located in the same divided sub-region based on the regional front view, and summarizing the obtained spot points into a regional division group corresponding to the divided sub-region; In response to the presence of different spot points having a connection relationship in the plurality of divided sub-regions, determining the point area corresponding to each spot point, and removing all spot points except the spot point with the largest point area from the area division group of the corresponding divided sub-region; generating a first initial distribution map, wherein the first initial distribution map includes an initial distribution area corresponding to each divided sub-area; The number of sub-regions of all the spot points in each region division group is determined, and the number of each sub-region is filled into the corresponding initial distribution region to obtain a first morphological distribution map.

5. The analysis method according to claim 4, characterized in that The method further comprises: Grouping all initial spot regions in a horizontal direction and a vertical direction based on the rectangular outline to obtain horizontal region groups and vertical region groups; Merging all initial spot regions in the same horizontal region group and vertical region group, and controlling the acquisition unit to acquire images of the obtained horizontal merged region and vertical merged region based on the side view and the vertical view to obtain a regional side view and a regional vertical view; Based on the first morphological distribution map, a lateral distribution area having a lateral relationship with each initial distribution area and a vertical distribution area having a vertical relationship with each initial distribution area are formed to convert each initial distribution area from an initial two-dimensional form to a three-dimensional form; Dividing the initial spot area into regions based on the side view based on the regional side view, and filling the number of sub-regions of all spot points included in each divided sub-region into the side distribution region corresponding to the initial spot area located in the same horizontal region group; The initial spot area is divided into regions based on the vertical view angle based on the regional vertical map, and the number of sub-regions of all spot points included in each divided sub-region is filled into the vertical distribution area corresponding to the initial spot area located in the same vertical area group.

6. The analysis method according to claim 4 or 5, characterized in that Performing image comparison on the first morphological distribution map and the second morphological distribution map, and determining a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position based on the comparison result, including: Performing transparency processing on the image portions other than the number of each sub-region in the first morphological distribution map and the second morphological distribution map, and overlapping the obtained first processed map and second processed map; In response to any number of sub-regions in the first processing image completely overlapping with any number of sub-regions in the corresponding same region position in the second processing image, the region position is determined as a coincidence attribute, otherwise it is determined as an offset attribute; In response to any region position being an offset attribute, determining the number of subregions corresponding to the region position in the first processed image as a first comparison number, determining the number of subregions corresponding to the region position in the second processed image as a second comparison number, and calculating based on a difference between the first comparison number and the second comparison number to obtain a region offset value corresponding to the region position; In response to any region position being a matching attribute, the region offset value corresponding to the region position is determined to be 0.

7. The analysis method according to claim 6, characterized in that Generate a current result display diagram corresponding to the test item based on all regional offset values, including: generating an initial result public graph, wherein the initial result public graph includes an offset public region corresponding to each initial spot region; Establishing a regional coordinate system for the offset public area with the center point of the area corresponding to the offset public area as the origin, and dividing the offset public area based on the X-axis and Y-axis of the corresponding regional coordinate system to obtain each publicized sub-area; Based on the regional coordinate system and the image coordinate system, a digital association relationship is established between each public sub-region and the regional offset value corresponding to the same regional position; The area offset values ​​with digital associations in each publicized sub-area located in the same offset publicized area are aggregated into an offset pointing group, and an offset pointer located in the corresponding offset publicized area is generated based on each offset pointing group to obtain the current result publicized map.

8. The analysis method according to claim 7, characterized in that Generating an offset pointer located in a corresponding offset disclosure area based on each offset pointer group includes: Determine the maximum regional offset value corresponding to the same offset pointing group as the ending offset value, and the minimum regional offset value as the starting offset value; and determine the regional center point of the public sub-region that has a numerical association with the ending offset value as the ending pointing point, and determine the regional center point of the public sub-region that has a numerical association with the starting offset value as the starting pointing point; Based on each offset orientation group, the largest regional offset value is determined as the maximum offset value, the corresponding minimum regional offset value is determined as the minimum offset value, and an offset interval consisting of the minimum offset value and the maximum offset value is determined; Dividing the offset interval into a number of intervals corresponding to a preset number of intervals, and configuring each obtained offset sub-interval corresponding to a different specification multiple; generating an offset pointer corresponding to a preset symbol specification and pointing from the starting pointing point to the ending pointing point, and performing an average calculation on the offset values ​​of all regions in the same offset pointing group; The offset mean obtained in response is located in any offset sub-interval, and the offset pointer corresponding to the offset pointing group is updated with a specification multiple corresponding to the offset sub-interval.

9. The analysis method according to claim 8, characterized in that The method further comprises: Determine any offset public announcement area as a primary public announcement area, and determine each other offset public announcement area surrounding the primary public announcement area as a secondary public announcement area; Determine the total number of secondary areas corresponding to all secondary public display areas, and multiply the total number of secondary areas by the retrieved preset offset ratio to obtain an allowable offset number; Comparing the direction of the offset pointer located in the primary public display area with the direction of the offset pointer located in each secondary public display area to obtain the direction offset degree corresponding to each secondary public display area; In response to the number of directional offsets corresponding to all directional offsets greater than the preset offset, which is greater than the allowed offset number, the main-level public area is marked as abnormal.

10. A transparent soil model analysis system, characterized in that: include: a first analysis module configured to construct an initial test model based on transparent soil having a test association relationship with a test target, and perform distribution analysis on the initial test model to obtain a first morphological distribution map consisting of a plurality of initial distribution areas; A second analysis module is configured to perform a deformation experiment of the transparent soil corresponding to the initial test model based on any test item, and perform a distribution analysis on the obtained current test model to obtain a second morphological distribution map consisting of multiple current distribution areas; an analysis and comparison module configured to perform an image comparison between the first morphological distribution map and the second morphological distribution map, and determine, based on the comparison result, a region offset value between an initial distribution region and a current distribution region corresponding to each identical region position; The result display module is configured to generate a current result display diagram corresponding to the test item based on all regional offset values.

11. A method for testing the initial test model constructed in claim 1 for compliance with standards, characterized in that: The following steps are involved: Dividing the initial test model into arrays based on the horizontal and vertical directions to obtain comparison areas; Controlling the irradiation unit to irradiate light corresponding to the test pixel value in the region extension direction of each comparison region, and controlling the acquisition unit to acquire an image of each comparison region to obtain a region comparison map; Determine each test pixel point located in each regional comparison image and having the same pixel value as the test pixel value, and determine other pixel points except the test pixel points as comparison pixel points; Determine any regional comparison graph as a first comparison graph, and determine a regional comparison graph corresponding to a region below the first comparison graph as a second comparison graph based on the initial test model; The first comparison image and the second comparison image are compared based on the comparison pixels, and a model state corresponding to the initial test model is determined based on the comparison result.

12. The detection method according to claim 11, characterized in that Performing image comparison based on the comparison pixels on the first comparison image and the second comparison image, and determining a model state corresponding to the initial test model based on the comparison result, includes: Calculating a ratio of a first number of pixels corresponding to all comparison pixel points located in the first comparison image to a second number of pixels corresponding to all comparison pixel points located in the second comparison image, and determining a quantity comparison value of a corresponding quantity dimension based on the obtained quantity ratio; Placing the first comparison image on top of the second comparison image in such a manner that a center point of a first image of the first comparison image coincides with a center point of a second image of the second comparison image, thereby obtaining an overlapping comparison image; Obtaining the number of overlapping pixels corresponding to all comparison pixel points located in the overlap comparison image, and determining the largest one of the first number of pixels and the second number of pixels as the deviation from the reference number; Calculating a ratio of the deviation reference number to the number of overlapping pixels, and determining a deviation contrast value corresponding to the deviation dimension based on the obtained deviation ratio; The quantity comparison value and the deviation comparison value are weightedly summed up and calculated, and if the real-time detection value obtained in response is greater than a preset detection value, the model state is determined to be a qualified state, otherwise it is determined to be a defective state.

Images