A frozen wall closure state detection method and system based on temperature cloud image analysis

Abstract

The application relates to the technical field of image processing, in particular to a frozen wall closure state detection method and system based on temperature cloud atlas analysis. The method comprises the following steps: acquiring an anti-permeability strength image of a horizontal section of a detection depth; acquiring a binary image of the anti-permeability strength image; extracting a heat conduction skeleton of a frozen wall from the binary image by using a center axis transformation algorithm; performing multiple sampling on the heat conduction skeleton to obtain a discrete sampling point sequence; acquiring effective thermal resistance thickness at each discrete sampling point; acquiring the geometric curvature of each discrete sampling point; taking the product of the geometric curvature and a water flow scouring intensity factor as an adjustment coefficient; taking the ratio of the effective thermal resistance thickness and the adjustment coefficient as a buckling stability index; and determining the closure state of the frozen wall according to the comparison result of the global minimum value of the buckling stability index and a preset safety threshold. The application has the effect of improving the detection accuracy of the frozen wall state.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method and system for detecting the closure state of a frozen wall based on temperature cloud map analysis. Background Technology

[0002] In geological engineering fields such as underground engineering and coal mine shaft construction, artificial freezing is a commonly used special well-drilling technique. This technique involves drilling holes around the area to be excavated and circulating a low-temperature refrigerant, causing the pore water in the strata to freeze. The frozen cylinders formed by adjacent freezing pipes fuse together, ultimately creating a closed, annular water-retaining curtain, or "frozen wall." The "looping" (effective closure) of the frozen wall is a prerequisite for ensuring safe subsequent excavation.

[0003] As geological engineering projects expand into complex strata such as water-rich soft rock in the west, the freezing method faces severe groundwater challenges. In these real-world scenarios, underground currents are highly active, and continuous groundwater flow exerts significant thermal erosion and lateral mechanical thrust on the forming frozen wall. Therefore, accurately assessing the true closure state and actual water-blocking capacity of the frozen wall in complex fluid-structure interaction environments has become an urgent safety requirement on-site.

[0004] Currently, on-site engineering methods typically rely on temperature cloud maps generated from temperature measurement wells, inferring the state of the frozen wall solely by determining whether the "zero-degree isotherm" is continuous and the geometric width of the connection point. However, using this conventional technique to address monitoring issues in water-rich scour scenarios can easily lead to serious anomalies in on-site safety assessments: because simply achieving "temperature compliance" cannot reflect the actual thermal capacity and physical strength of the structure, and it ignores the asymmetric distortion (such as bending) of the frozen wall morphology caused by water flow impact, it often results in a surface that appears to have formed a ring, while the core area is actually extremely thin and fragile (i.e., a false ring).

[0005] Therefore, in the environment of groundwater scouring, existing detection methods that rely solely on temperature thresholds and geometric distances cannot effectively identify the hidden danger of "false loops" with low impermeability inside the frozen wall, nor can they perceive and assess the risk of buckling instability caused by the bending of the shape due to water flow thrust. As a result, it is difficult to accurately assess the closed state in the fluid-structure interaction environment, which increases the risk of well flooding accidents during construction. Summary of the Invention

[0006] To improve the accuracy of assessing the closure state of frozen walls in fluid-structure interaction environments, this application provides a method and system for detecting the closure state of frozen walls based on temperature contour map analysis.

[0007] Firstly, this application provides a method for detecting the closure state of a frozen wall based on temperature contour map analysis, employing the following technical solution:

[0008] A method for detecting the closure state of a frozen wall based on temperature cloud map analysis includes: acquiring a horizontal cross-sectional temperature field image of the depth to be detected; using a nonlinear mapping function to convert the temperature data in the temperature field image into an impermeability strength coefficient characterizing the freezing strength; and obtaining an impermeability strength image.

[0009] A binary image of the impermeability strength is obtained. The thermal conductivity skeleton of the frozen wall is extracted from the binary image using the median transformation algorithm. Multiple samplings are performed on the thermal conductivity skeleton to obtain a discrete sampling point sequence.

[0010] For any sampling point, obtain the normal vector of the sampling point, and perform discrete integration along the direction of the normal vector to obtain the effective thermal resistance thickness at each discrete sampling point.

[0011] The geometric curvature of each discrete sampling point is obtained, and the product of the geometric curvature and the water flow scour intensity factor is used as the adjustment coefficient. The ratio of the effective thermal resistance thickness to the adjustment coefficient is used as the buckling stability index. The closure state of the frozen wall is determined based on the comparison between the global minimum value of the buckling stability index and the preset safety threshold.

[0012] By acquiring a horizontal cross-sectional temperature field image of the depth to be detected and using a nonlinear mapping function to convert the temperature data into a permeability resistance coefficient, the original temperature information is transformed from a single thermal parameter into a permeability resistance coefficient with physical significance. This nonlinear mapping amplifies the sensitive changes in the phase transition range, enabling accurate identification of strength degradation in areas approaching the freezing critical temperature, thus improving the authenticity of strength characterization from the source. Secondly, by binarizing the permeability resistance image and using a mid-axis transformation algorithm to extract the thermal conduction skeleton, the extracted skeleton represents the main channels for cold energy transfer and the core load-bearing structure of the frozen wall. Key load-bearing paths are extracted geometrically, providing a foundation for subsequent mechanical risk analysis. Furthermore, discrete sampling is performed on the skeleton, and the permeability resistance coefficient is discretely integrated along the normal direction to obtain the effective thermal resistance thickness. This ensures that the thickness parameter simultaneously contains both "geometric width" and "internal strength" information, enabling the identification of weak sections with geometrical intersections but insufficient internal heat capacity.

[0013] Finally, a buckling stability index is constructed by introducing geometric curvature and water flow scour intensity factor. The evaluation is made by comparing the global minimum value with the safety threshold, so that the evaluation results take into account both structural strength and the morphological bending risk under fluid-structure interaction environment. This effectively solves the problem that the existing technology cannot identify false loops and buckling instability risks, and achieves accurate determination of the closure state of the frozen wall.

[0014] Optionally, the nonlinear mapping function is a Sigmoid function, and the independent variable of the Sigmoid function is the difference between the temperature value in the temperature field image and the preset freezing critical temperature constant.

[0015] The nonlinear mapping function is defined as a Sigmoid function, with the independent variable being the difference between the temperature value and the freezing critical temperature constant. This causes the impermeability coefficient to exhibit a steep change near the freezing critical temperature, accurately reflecting the abrupt changes in the physical and mechanical properties of frozen soil within the phase transition range.

[0016] Optionally, the heat conduction skeleton of the frozen wall is extracted from the binary image using the median transformation algorithm, including: extracting an initial skeleton from the binary image using the median transformation algorithm, and smoothing the initial skeleton to obtain the heat conduction skeleton.

[0017] The initial skeleton typically contains pixel-level jagged noise. If used directly for curvature and normal vector calculations, it can easily lead to abrupt changes in local orientation, thus affecting the accuracy of subsequent integration. Smoothing processes eliminate high-frequency noise, enhancing the continuity of the skeleton curve, making curvature calculations more stable, and ensuring more accurate normal vector directions.

[0018] Optionally, the normal vector of the sampling point can be obtained by using the central difference method to obtain the normal vector at each sampling point.

[0019] Optionally, the geometric curvature of each discrete sampling point is obtained, including: obtaining the coordinates of each sampling point and its adjacent sampling points, and calculating the geometric curvature at each sampling point based on the first-order difference and second-order difference of the coordinates.

[0020] By introducing the coordinates of adjacent sampling points for differential calculation, the curvature of the curve can be accurately obtained under discrete data conditions. Geometric curvature, as a morphological distortion index, can reflect the bending trend of the frozen wall under the thrust of water flow. By quantifying the curvature change, the subsequent buckling stability index can reflect morphological risk factors.

[0021] Optionally, obtaining a binary image of the impermeability strength image includes: setting an intensity threshold, setting regions in the impermeability strength image that are greater than or equal to the intensity threshold to 1, and regions that are less than the intensity threshold to 0.

[0022] Setting an intensity threshold allows for binarization of the anti-permeability image, distinguishing between effective and ineffective frozen areas.

[0023] Optionally, the water scour intensity factor is determined by the ratio of groundwater flow velocity to a reference flow velocity.

[0024] The scour intensity factor is defined as the ratio of groundwater flow velocity to the reference flow velocity. By introducing the velocity ratio, the hydrodynamic environment is quantified into a parameter that can be used in calculations, allowing the buckling stability index to reflect the differences in scour intensity under different hydrological conditions.

[0025] Optionally, the closure state of the frozen wall is determined based on the comparison result between the global minimum value of the buckling stability index and the preset safety threshold, including: if the global minimum value is greater than or equal to the preset safety threshold, the frozen wall is determined to be in an effective closure state.

[0026] If the global minimum value is less than the preset safety threshold, the frozen wall is determined to be in a high-risk closed state.

[0027] Using the global minimum as the criterion allows us to focus on the weakest point in the structure.

[0028] Optionally, the initial skeleton can be smoothed using a B-spline curve fitting method.

[0029] B-splines possess local support and high-order continuity, enabling them to eliminate noise interference while maintaining the overall shape of the skeleton.

[0030] Secondly, this application provides a frozen wall closure state detection system based on temperature cloud map analysis, employing the following technical solution:

[0031] A frozen wall closure state detection system based on temperature cloud map analysis includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the frozen wall closure state detection method based on temperature cloud map analysis described above.

[0032] The above-mentioned method for detecting the closure state of a frozen wall based on temperature cloud map analysis is generated into a computer program and stored in a memory for loading and execution by a processor. Thus, a system is created based on the memory and processor for convenient use.

[0033] This application has the following technical effects:

[0034] By converting temperature data into physical impermeability strength through nonlinear mapping, and calculating the effective thermal resistance thickness using central axis transformation and normal integral, the hidden danger of "false interlocking" due to insufficient internal heat capacity can be accurately identified. At the same time, the buckling stability index is constructed by integrating the geometric curvature of the skeleton and the water flow scouring factor, realizing a comprehensive quantitative assessment of the impermeability and morphological instability risk of the frozen wall. Attached Figure Description

[0035] Figure 1 This is a flowchart of a method for detecting the closure state of a frozen wall based on temperature cloud map analysis, according to an embodiment of this application.

[0036] Figure 2 This is a schematic diagram of the detection results of the existing detection method for the frozen wall.

[0037] Figure 3 This is a schematic diagram of the test results of the detection method in this application on the frozen wall. Detailed Implementation

[0038] This application discloses a method for detecting the closure state of a frozen wall based on temperature cloud map analysis. First, the original temperature field is transformed into a seepage resistance strength image with physical seepage resistance significance using a nonlinear mapping function, eliminating interference from invalid data and quantifying the strength differences in low-temperature regions. Then, a central axis transformation algorithm is used to extract a sequence of discrete sampling points representing the main channels for cold energy transmission and the core pressure-bearing structure of the frozen wall. By discrete integration along the normal vector direction, the effective thermal resistance thickness, coupled with both "thickness" and "strength" information, is calculated, thereby identifying vulnerable areas with insufficient heat capacity. Finally, the geometric curvature of the framework and the water flow scour intensity factor characterizing the impact of the groundwater environment are introduced to construct a buckling stability index that quantifies morphological risk. By comparing the global minimum value of this index with a safety threshold, the accurate determination of the frozen wall's closure state is achieved.

[0039] Reference Figure 1 A method for detecting the closure state of a frozen wall based on temperature cloud map analysis includes steps S1-S4.

[0040] S1: Obtain the horizontal cross-sectional temperature field image of the depth to be detected, and use a nonlinear mapping function to convert the temperature data in the temperature field image into the impermeability strength coefficient that characterizes the freezing strength, thereby obtaining the impermeability strength image.

[0041] In this embodiment, specifically, the temperature field image of the horizontal cross-section at the depth to be detected needs to be acquired first, denoted as... Its unit is Celsius (°C) This horizontal cross-sectional temperature field image reflects a continuous temperature distribution over a plane at a specific depth.

[0042] The temperature field image is generated by spatial interpolation of discrete temperature data collected from multiple temperature measuring holes deployed within the frozen area. For example, Kriging (…) can be used. Kriging interpolation is a linear discrete-point interpolation method based on variogram theory and structural analysis. It can consider the correlation of spatial attributes in spatial location, thus making the generated horizontal cross-sectional temperature field image more geologically significant when simulating the temperature distribution of underground strata.

[0043] Understandably, traditional detection methods only focus on Isotherms are observed, but the physical strength of frozen soil increases non-linearly with decreasing temperature. To quantify this difference, this step constructs a permeability strength image using a non-linear mapping function. The formula for calculating the permeability strength coefficient can be expressed as:

[0044] ;

[0045] In the formula: Representing coordinates The impermeability strength coefficient at the location is a dimensionless parameter, and its value ranges from 0 to 10. between; Representing coordinates The actual temperature value at the location, in units of ; This represents the critical freezing temperature constant of pore water in soil, expressed in units of... ; This represents the phase transition sensitivity coefficient, with units of . ; Represented by natural constant An exponential function with base 0.

[0046] use The nonlinear characteristics of the function simulate the abrupt change in the physical and mechanical properties of frozen soil near the phase transition point. When the actual temperature value is much higher than the critical freezing temperature constant, the difference within the parentheses is positive. After being amplified by the phase transition sensitivity coefficient, the exponential term in the denominator tends to infinity, causing the impermeability strength coefficient to tend towards... This indicates that the area is not frozen and lacks impermeability; when the actual temperature is much lower than the critical freezing temperature constant, the value in parentheses is negative, and the exponent term tends to... This leads to the impermeability strength coefficient tending to This indicates that the area has been completely frozen and has reached its maximum impermeability.

[0047] Freezing critical temperature constant The value range is set based on staff experience. to If the setting is too high, weakly frozen soil in the phase transition range may be misclassified as hard frozen soil; if the setting is too low, the actual water-sealing capacity of the frozen wall may be underestimated. Phase transition sensitivity coefficient. The preferred range is to .when Below At that time, the intensity changes too gradually with temperature, failing to reflect the critical characteristics of the phase transition point; when Higher than At times, the model is too sensitive, and even small temperature fluctuations can cause drastic changes in the intensity assessment.

[0048] For example, suppose we set , .

[0049] If a certain sampling point temperature Substitute into the formula for calculating the impermeability strength coefficient: This reflects a highly frozen state.

[0050] If a certain sampling point temperature Substitute into the formula for calculating the impermeability strength coefficient: This reflects a fragile state where, although frozen, the strength has not yet fully developed.

[0051] S2: Obtain a binary image of the impermeability strength image, extract the thermal conductivity skeleton of the frozen wall from the binary image using the median transformation algorithm, and obtain a discrete sampling point sequence by sampling multiple times on the thermal conductivity skeleton.

[0052] To extract the core load-bearing structure of the frozen wall, the continuous impermeability strength image needs to be converted into a discrete binary form. First, a strength threshold is set. (Example, desirable) ). (The image of impermeability strength) Set the area to (Effective frozen area) will Set the area to (Invalid region), generate a binary image .

[0053] Subsequently, axis transformations were applied to binary images. The median transformation algorithm is a classic algorithm for extracting the skeleton of a graphic. Its specific implementation involves finding the set of all points within a region that are at least two points closest to the region's boundary. This algorithm then extracts the refined connected curves of the region, i.e., the initial skeleton.

[0054] To eliminate pixel-level jagged noise on the initial skeleton, as a preferred approach, this embodiment uses the B-spline curve fitting method to parametrically smooth the initial skeleton to obtain the heat conduction skeleton. Spline fitting defines a curve using control points and basis functions, exhibiting local support and good smoothness. Equal-interval sampling is then performed on the smoothed curve to generate a discrete sequence of sampling points. ,in This is the index for discrete sampling points.

[0055] S3: For any sampling point, calculate the normal vector of the sampling point, and perform discrete integration along the direction of the normal vector on the impermeability strength coefficient to obtain the effective thermal resistance thickness at each discrete sampling point.

[0056] Effective thermal resistance thickness is used to measure the water-sealing capacity of a frozen wall. First, for each sampling point on the thermal conductivity framework... The normal vector at each sampling point is obtained using the central difference method. The specific formula is as follows:

[0057] ;

[0058] ;

[0059] ;

[0060] ;

[0061] in, The coordinates of the next sampling point. These are the coordinates of the previous sampling point.

[0062] After calculating the unit normal vector, probe rays are emitted to both sides along that direction, and then incremented by a step size. Cumulative impermeability strength coefficient. Effective thermal resistance thickness. The calculation formula is as follows:

[0063] ;

[0064] In the formula: Indicates the first heat conduction skeleton The effective thermal resistance thickness at each discrete sampling point, in meters ( ); This represents the spatial sampling step size along the normal vector direction, which can be set based on the experience of those skilled in the art, and the unit is 1. ; Indicates the index of the discrete sampling point along the direction of the normal vector; This indicates the maximum search range (i.e., the maximum half-length of the ray scan). The impermeability coefficient corresponding to the coordinates of the detection point; and The first The horizontal and vertical components of the unit normal vector for each discrete sampling point; Indicates the coordinates of the detection point The corresponding impermeability strength coefficient at that location.

[0065] When the impermeability coefficient along the sampling path is generally high (i.e., the temperature is low), the accumulated effective thermal resistance thickness is large; conversely, if a certain point has a large geometric width but an internal temperature close to the freezing critical temperature, resulting in a low impermeability coefficient, the integrated effective thermal resistance thickness will be significantly reduced. This effectively identifies "false loops" formed by water flow thermal erosion.

[0066] For example, setting the spatial sampling step size .

[0067] In the first case, assuming it's at the skeleton point... There are 40 points on the normal scan path. All values ​​are above 0.9. (Cumulative sum) .

[0068] but This indicates that there is an equivalent of 1.8 meters of solid permafrost at that location.

[0069] In the second scenario, assuming it's at the skeleton point... Although the geometric width is also 2 meters (covering 40 points), due to thermal erosion from the water flow, the temperature at most points is around -1.5℃. It is approximately 0.6.

[0070] Summation .

[0071] but .

[0072] Although the geometric width of the interlocking loops is the same, the effective thermal resistance thickness in the second case is significantly smaller than that in the first case, accurately reflecting the potential problem of insufficient internal heat capacity.

[0073] Thus, by introducing a weighted integral of the impermeability strength coefficient, the interlocking thickness is physically coupled with strength. Only when the physical thickness is large and the internal temperature is sufficiently low (high strength) will the calculated effective thermal resistance thickness be large, thereby avoiding the defect of traditional geometric methods that ignore the internal temperature distribution gradient.

[0074] S4: Obtain the geometric curvature of each discrete sampling point, use the product of the geometric curvature and the water flow scour intensity factor as the adjustment coefficient, use the ratio of the effective thermal resistance thickness to the adjustment coefficient as the buckling stability index, and determine the closure state of the frozen wall based on the comparison result of the global minimum value of the buckling stability index and the preset safety threshold.

[0075] The geometric curvature is calculated using the first and second differences of the coordinates; specifically, for any sampling point, the formula for calculating its geometric curvature can be expressed as:

[0076] In the formula, Indicates sampling point The geometric curvature of the skeleton; and Sampling points The first difference between the x and y coordinates, and Sampling points The second difference between the x-axis and y-axis.

[0077] Subsequently, a buckling stability index was constructed. Its calculation formula can be expressed as:

[0078] In the formula: Represents discrete sampling points The buckling stability index at the location, in units of ; Represents discrete sampling points Effective thermal resistance thickness at the location, in units of ; Represents discrete sampling points Geometric curvature at point, in units of ; This represents a dimensionless water erosion intensity factor; for example, this factor is derived from the actual groundwater flow velocity. Compared with the reference flow rate The ratio is determined; Represents the dimensional balance and sensitivity coefficient, with units of . .

[0079] When the frozen wall morphology becomes severely bent (increased geometric curvature) and the external water flow environment is harsh (increased water flow scour intensity factor), the penalty term in the denominator increases significantly, leading to a sharp decrease in the buckling stability index. This means that even if the thickness meets the standard, if the morphology is unstable, it will still be judged as a high-risk state.

[0080] Finally, by iterating through the buckling stability indices of all points on the skeleton, the global minimum value is selected. .

[0081] like If so, the frozen wall is determined to be in an effective closed state. A preset safety threshold, such as 1.2m, is set based on the experience of those skilled in the art.

[0082] like If the condition is deemed a high-risk closed state, the system will automatically trigger an alarm event and output the coordinates corresponding to the global minimum value. An alarm event can be understood as a safety warning signal issued by the system, such as displaying a flashing red marker on the monitoring interface or sending reinforcement instructions to construction personnel.

[0083] As a preferred option, dimensional balance and sensitivity coefficient The range of values ​​is to This coefficient reflects the construction party's tolerance for morphological risks; the higher the value, the more severe the penalty for bending distortion.

[0084] Combination Figure 2 and Figure 3 , Figure 2This is a schematic diagram of the detection results of the frozen wall in the existing technology. The frozen wall connects the geometric center of the two frozen pipes, and the bending and displacement of the actual frozen wall cannot be detected under the action of water flow. Figure 3 This is a schematic diagram of the detection results of the frozen wall in this application. In the detection method of this application, by extracting the heat conduction skeleton between two frozen pipes and calculating the curvature, the high curvature high-risk area in the figure can be accurately identified. Although this area is connected, due to excessive bending and insufficient thickness, the buckling stability index value is extremely low, and there is a large risk of water collapse, thus improving the accuracy of the identification of the frozen wall condition.

[0085] This application also discloses a frozen wall closure state detection system based on temperature cloud map analysis, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a frozen wall closure state detection method based on temperature cloud map analysis according to this application is implemented.

[0086] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0087] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for detecting the closure state of a frozen wall based on temperature contour map analysis, characterized in that, include: A horizontal cross-sectional temperature field image of the depth to be detected is obtained. The temperature data in the temperature field image is converted into a permeability resistance coefficient that characterizes the freezing strength using a nonlinear mapping function, thus obtaining a permeability resistance image. A binary image of the impermeability strength is obtained. The thermal conductivity skeleton of the frozen wall is extracted from the binary image using the median transformation algorithm. Multiple samplings are performed on the thermal conductivity skeleton to obtain a discrete sampling point sequence. For any sampling point, obtain the normal vector of the sampling point, and perform discrete integration along the direction of the normal vector to obtain the effective thermal resistance thickness at each discrete sampling point. The geometric curvature of each discrete sampling point is obtained, and the product of the geometric curvature and the water flow scour intensity factor is used as the adjustment coefficient. The ratio of the effective thermal resistance thickness to the adjustment coefficient is used as the buckling stability index. The closure state of the frozen wall is determined based on the comparison between the global minimum value of the buckling stability index and the preset safety threshold. The formula for calculating the impermeability strength coefficient is: ; Representing coordinates Permeability resistance coefficient at the location; Representing coordinates The actual temperature value at that location; This represents the critical freezing temperature constant of pore water in soil. This represents the phase transition sensitivity coefficient; Represented by natural constant An exponential function with base 0; The water scour intensity factor is determined by the ratio of groundwater flow velocity to the reference flow velocity.

2. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, The nonlinear mapping function is a Sigmoid function, and the independent variable of the Sigmoid function is the difference between the temperature value in the temperature field image and the preset freezing critical temperature constant; the value range of the freezing critical temperature constant is... to The range of values ​​for the phase transition sensitivity coefficient is: to .

3. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, The method of extracting the thermal conductivity skeleton of the frozen wall from a binary image using the median transformation algorithm includes: extracting the initial skeleton from the binary image using the median transformation algorithm, and smoothing the initial skeleton to obtain the thermal conductivity skeleton.

4. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, Obtaining the normal vector of the sampling point includes: using the central difference method to obtain the normal vector at each sampling point.

5. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, Obtaining the geometric curvature of each discrete sampling point includes: obtaining the coordinates of each sampling point and its adjacent sampling points, and calculating the geometric curvature at each sampling point based on the first-order and second-order differences of the coordinates.

6. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, Obtaining a binary image of the permeability resistance strength image includes: setting an intensity threshold, setting regions in the permeability resistance strength image that are greater than or equal to the intensity threshold to 1, and regions that are less than the intensity threshold to 0.

7. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 1, characterized in that, The closure state of the frozen wall is determined based on the comparison between the global minimum value of the buckling stability index and the preset safety threshold, including: if the global minimum value is greater than or equal to the preset safety threshold, the frozen wall is determined to be in an effective closure state. If the global minimum value is less than the preset safety threshold, the frozen wall is determined to be in a high-risk closed state.

8. The method for detecting the closure state of a frozen wall based on temperature cloud map analysis according to claim 3, characterized in that, The initial skeleton was smoothed using a B-spline curve fitting method.

9. A system for detecting the closure state of a frozen wall based on temperature cloud map analysis, characterized in that, include: The processor and memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a method for detecting the closure state of a frozen wall based on temperature contour map analysis according to any one of claims 1-8.

Images