Building foundation settlement detection method based on computer vision

Abstract

The invention belongs to the field of foundation settlement detection, and relates to a data analysis technology, in particular to a building foundation settlement detection method based on computer vision, which comprises a detection calibrator method and an instant decision sub-method. The calibrator detection method comprises the following steps: carrying out visual settlement detection analysis on a building foundation; analyzing the effectiveness of the visual settlement detection result of the building foundation observation point; performing visual settlement risk analysis on the observation points of the building foundation; carrying out settlement check analysis on the observation points of the building foundation; outputting a settlement detection result of the building foundation observation point; the precision and anti-interference capability of foundation settlement detection can be effectively improved, accurate measurement of millimeter-level displacement is realized based on pixel grid image processing and a coordinate extraction technology of a two-dimensional code target through a dual verification mechanism of multi-camera redundancy arrangement and coordinate variation, and the precision and anti-interference capability of foundation settlement detection are improved. And high-precision data support is provided for safety assessment of the building structure.

Description

Technical Field

[0001] The present invention belongs to the field of foundation settlement detection and relates to data analysis technology, in particular to a building foundation settlement detection method based on computer vision. Background Art

[0002] Building foundation settlement detection is a key technology for evaluating the structural safety of a building by measuring its vertical displacement changes in real time or periodically. Traditional methods rely on manual measurement, while modern technologies combine sensors, computer vision and artificial intelligence to achieve automated, high-precision and low-cost monitoring.

[0003] The invention patent with announcement number CN118095813B discloses a visualization monitoring method and system for foundation settlement based on BIM technology. The visualization monitoring method can input the extracted indicators into the foundation settlement prediction model to perform settlement prediction and obtain foundation settlement risk indicators; however, the visualization monitoring method only relies on the prediction model of a single algorithm for settlement detection and cannot be verified through multiple algorithms. When the collection terminal fails or the collected data is interfered with by the environment, the detection results have large errors; in addition, when the detection results of multiple algorithms are inconsistent, the existing technology cannot combine environmental parameters with historical verification data to generate immediate decision results, resulting in a lag in the result output.

[0004] In response to the above technical problems, this application proposes a solution. Summary of the Invention

[0005] The purpose of the present invention is to provide a building foundation settlement detection method based on computer vision, which is used to solve the problem that the existing technology cannot be verified by multiple algorithms; The technical problem to be solved by the present invention is: how to provide a building foundation settlement detection method based on computer vision that can be verified by multiple algorithms.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A building foundation settlement detection method based on computer vision, comprising a detection verification sub-method and an instant decision sub-method; The method for detecting syndromes comprises the following steps: Step S1: Perform visual settlement detection and analysis on the building foundation; Step S2: Analyze the validity of the visual settlement detection results of the building foundation observation points; Step S3: Conduct visual settlement risk analysis on building foundation observation points; Step S4: performing settlement verification analysis on the building foundation observation points; Step S5: Outputting the settlement detection results of the building foundation observation points; The instant decision-making sub-method comprises the following steps: Step P1: Statistical analysis of manual verification results; Step P2: Conduct environmental correlation analysis on the manual verification results; Step P3: Generate immediate decision results.

[0007] Furthermore, in step S1, the specific process of visual settlement detection and analysis includes: arranging a camera array at the observation point of the building, regularly capturing images of the observation point through the camera and marking the captured images as detection images, enlarging the detection images into pixel grid images and extracting the coordinates of the QR code targets in the detection images, and marking the distance between the two most recently extracted QR code target coordinate points of the same camera as the settlement detection value of the camera.

[0008] Furthermore, in step S2, the specific process of analyzing the effectiveness of the visual settlement detection results includes: performing variance calculation on the settlement detection values ​​collected by all cameras in the camera array of the same observation point to obtain the effectiveness coefficient of the observation point, and comparing the effectiveness coefficient with the preset effectiveness threshold: if the effectiveness coefficient is less than the effectiveness threshold, then it is determined that the effectiveness of the visual settlement detection result of the observation point meets the requirements, and the settlement detection values ​​collected by all cameras in the camera array of the observation point are summed and averaged to obtain the visual settlement value; if the effectiveness coefficient is greater than or equal to the effectiveness threshold, then it is determined that the effectiveness of the visual settlement detection result of the observation point does not meet the requirements, and a camera maintenance signal is generated and sent to the mobile phone terminal of the administrator.

[0009] Furthermore, in step S3, the specific process of visual subsidence risk analysis includes: comparing the visual subsidence value of the observation point with a preset subsidence threshold: if the visual subsidence value is less than the subsidence threshold, it is determined that the observation point does not have a visual subsidence risk; if the visual subsidence value is greater than or equal to the subsidence threshold, it is determined that the observation point has a visual subsidence risk.

[0010] Furthermore, in step S4, the specific process of settlement verification analysis includes: the detection period is composed of the moments when the camera performs the two most recent image acquisitions, the horizontal angle change of the observation point within the detection period is obtained through the inclinometer and marked as the settlement verification value, and the settlement verification value is compared with the preset settlement verification threshold: if the settlement verification value is less than the settlement verification threshold, it is determined that the observation point does not have the verification settlement risk; if the settlement verification value is greater than or equal to the settlement verification threshold, it is determined that the observation point has the verification settlement risk.

[0011] Furthermore, in step S5, the specific process of outputting the settlement detection result includes: if the observation point has both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is judged to be unqualified, a settlement warning signal is generated and the settlement warning signal is sent to the server; if the observation point does not have both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is judged to be qualified, a qualified detection signal is generated and the qualified detection signal is sent to the server; otherwise, it is judged that there is a dispute over the settlement detection result of the observation point, the corresponding observation point is marked as a verification point, a manual verification signal is generated and the manual verification signal is sent to the server; after receiving the manual verification signal, the server sends the manual verification signal to the mobile phone terminal of the administrator, and the administrator manually verifies the observation point and marks the visual settlement detection process or the settlement verification process as the correct process.

[0012] Furthermore, in step P1, a statistical period is generated, and the detection value JCi of the environmental parameter i when the verification point is marked within the statistical period is obtained, where i=1, 2,…, n, and n is a positive integer. The environmental parameter i includes the air temperature value, air humidity value, fog concentration value, rainfall, and wind level at the verification point.

[0013] Furthermore, in step P2, the specific process of environmental association analysis includes: when the visual settlement detection process is marked as a correct process, the corresponding verification point is marked as a visual association point; when the settlement verification process is marked as a correct process, the corresponding verification point is marked as a verification association point; data cleaning processing is performed on the visual association point to obtain the visual priority range SYi of the environmental parameter i; and data cleaning processing is performed on the verification association point in the same way to obtain the verification priority range HYi of the environmental parameter i.

[0014] Furthermore, in step P2, the specific process of data cleaning processing of visual association points includes: forming a visual set i of environmental parameter i from the detection values ​​JCi of environmental parameter i of all visual association points, performing variance calculation on all elements of the visual set i to obtain the visual concentration value SJi of environmental parameter i, and comparing the visual concentration value SJi with the preset concentration threshold JZi: if the visual concentration value SJi is greater than or equal to the concentration threshold JZi, then the maximum element and the minimum element in the visual set i are eliminated, and then the visual concentration value SJi is recalculated, and so on, until the visual concentration value SJi is less than the concentration threshold JZi; if the visual concentration value SJi is less than the concentration threshold JZi, then the maximum element and the minimum element retained in the visual set i constitute the visual priority range SYi of environmental parameter i.

[0015] Furthermore, in step P3, the process of generating an instant decision result includes: when there is a dispute over the settlement detection result of the observation point in the future, obtaining the detection value JCi of the environmental parameter i of the observation point, and determining whether the detection value JCi of the environmental parameter i is all within the corresponding visual priority range SYi: if so, sending the visual settlement detection result in step S3 to the server; if not, determining whether the detection value JCi of the environmental parameter i is all within the corresponding verification priority range HYi: if so, sending the settlement verification result in step S4 to the server; if not, generating a manual verification signal and sending the manual verification signal to the server.

[0016] The present invention has the following beneficial effects: 1. This invention can effectively improve the accuracy and anti-interference ability of foundation settlement detection. Through the redundant arrangement of multiple cameras and the dual verification mechanism of coordinate changes, it solves the problem of insufficient data reliability in traditional methods under equipment failure or environmental interference. At the same time, based on pixel grid image processing and coordinate extraction technology of QR code targets, it achieves accurate measurement of millimeter-level displacement, providing high-precision data support for building structure safety assessment. 2. This invention addresses the issue of insufficient reliability of single detection methods in existing technologies. By cross-validating data from visual and physical sensors, it significantly improves the accuracy of settlement detection results. Furthermore, when there are disputes over test results, a manual verification process can be triggered to further confirm the root cause of the problem, such as distinguishing between equipment failure and environmental interference, thereby improving the overall robustness of the system. 3. This invention establishes an environment-related priority range database, integrating real-time environmental parameters to make intelligent decisions when test results are disputed. This effectively addresses the issue of multi-algorithm verification failures caused by sensor failure or environmental interference. Furthermore, through continuous optimization of historical verification data, the accuracy of priority range determination gradually improves over time. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a flow chart of a method according to embodiment 1 of the present invention; Figure 2 This is a flow chart of the method of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0019] The technical solutions of the present invention will be described clearly and completely below with reference to the embodiments. It is obvious that the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0020] In the existing technology, building foundation settlement detection mainly relies on manual measurement or a single algorithm model for predictive analysis. Traditional manual measurement has the problems of low efficiency and high cost. Although the automated monitoring system based on sensors or computer vision has improved efficiency, it generally lacks a multi-algorithm cross-validation mechanism. When the acquisition equipment fails or the monitoring data is interfered with by the environment, a single detection result is prone to errors. In addition, when there are contradictions in multi-source data, the existing technology usually requires manual intervention to judge, and lacks an intelligent decision-making mechanism based on historical verification data and environmental parameters, resulting in a lag in the output of the results. For example, under complex environmental conditions such as rainy and foggy weather or sudden temperature changes, there may be significant deviations between visual detection data and sensor data. At this time, the system cannot quickly determine the source of the credible data, affecting the timeliness of monitoring.

[0021] In order to solve the above problems, the inventors found that the existing settlement detection system has two core defects: one is the insufficient reliability of a single detection method, and the other is the lack of an intelligent decision-making mechanism for disputed data. In response to the first point, it is considered to introduce a dual verification mechanism of visual detection and physical sensors to improve the credibility of the results through cross-validation of multi-source data. For the second point, it was found that the correlation between environmental factors and the applicability of detection methods is hidden in the manual verification records, and thus it is proposed to establish a dynamic mapping relationship between environmental parameters and the priority of detection methods. Based on this, a technical route combining phased detection and verification with intelligent and immediate decision-making is gradually formed: first, the accuracy of basic data is ensured through dual analysis of visual detection and inclinometer verification. When there is a dispute over the results, the optimal detection method is automatically selected based on the environmental correlation characteristics in the historical verification data, thereby reducing the frequency of manual intervention.

[0022] A computer vision-based building foundation settlement detection method includes a detection verification sub-method and an instant decision sub-method.

[0023] Example 1: Figure 1 As shown, the syndrome detection method includes the following steps: Step S1: Perform visual settlement detection and analysis on the building foundation: Arrange a camera array at the observation point of the building, use the camera to regularly capture images of the observation point and mark the captured images as detection images, enlarge the detection images into pixel grid images and extract the coordinates of the QR code target in the detection images, and mark the distance between the two most recently extracted QR code target coordinate points by the same camera as the camera's settlement detection value; Among them, the camera array refers to a shooting system composed of multiple cameras. Specifically, it can be implemented by industrial cameras fixedly installed at different angles around the observation point, ensuring the integrity of image acquisition through multi-angle coverage. The pixel grid image refers to the gridded image formed by magnifying the original image. Specifically, it can be implemented by image interpolation algorithms, such as bilinear interpolation or cubic convolution interpolation, to facilitate the precise positioning of coordinate points by improving the resolution. QR code target coordinate extraction refers to obtaining the position information of the QR code in the image through image recognition algorithms. Specifically, it can be implemented by edge detection combined with feature matching algorithms, such as feature point detection based on SIFT or ORB, to determine the coordinates of its center point by identifying the geometric features of the QR code.

[0024] Specifically, multiple cameras are arranged around the observation point to form an array, and each camera collects ground images at preset time intervals. The collected images are magnified and converted into pixel grid images. The image processing algorithm is used to identify the QR code target and extract its center coordinates. By comparing the distance change between the coordinate points collected twice by the same camera, the settlement detection value of the camera is calculated. For example, when the center coordinates of the QR code are (x1, y1) in the first shot of a camera and (x2, y2) in the second shot, the distance between the two coordinate points is calculated as the settlement amount using the Euclidean distance formula. This method based on redundant acquisition and coordinate comparison of multiple cameras can effectively eliminate single-point measurement errors and provide a reliable data basis for subsequent analysis.

[0025] Step S2: Analyze the validity of the visual settlement detection results of the building foundation observation point: perform variance calculation on the settlement detection values ​​collected by all cameras in the camera array of the same observation point to obtain the effective coefficient of the observation point, and compare the effective coefficient with the preset effective threshold: if the effective coefficient is less than the effective threshold, it is determined that the validity of the visual settlement detection result of the observation point meets the requirements, and the settlement detection values ​​collected by all cameras in the camera array of the observation point are summed and averaged to obtain the visual settlement value; if the effective coefficient is greater than or equal to the effective threshold, it is determined that the validity of the visual settlement detection result of the observation point does not meet the requirements, and a camera maintenance signal is generated and sent to the mobile phone terminal of the manager; The validity coefficient is a quantitative indicator reflecting the consistency of camera array data through variance calculation. Specifically, the variance formula in statistics can be used to calculate the degree of dispersion of multiple camera detection values ​​to determine data reliability. The validity threshold is a pre-set critical value for determining data validity. It can be determined through experiments or historical data, for example, a variance threshold of 0.5 mm². The camera maintenance signal is a command that triggers equipment maintenance. Specifically, abnormal status information can be pushed to the terminal through the IoT communication module, thereby quickly locating the faulty equipment.

[0026] Specifically, when the degree of discreteness of the settlement detection values ​​collected by multiple devices in the camera array is low, it indicates that the data consistency of each camera is high. At this time, the visual settlement value is obtained by taking the average value to improve the detection accuracy; when the degree of discreteness exceeds the threshold, it indicates that some cameras may have installation offset, lens contamination or hardware failure, and the maintenance process is automatically triggered.

[0027] Step S3: Perform visual settlement risk analysis on the building foundation observation point: compare the visual settlement value of the observation point with a preset settlement threshold: if the visual settlement value is less than the settlement threshold, the observation point is determined to have no visual settlement risk; if the visual settlement value is greater than or equal to the settlement threshold, the observation point is determined to have a visual settlement risk; Among them, the visual settlement value refers to the average value of the settlement detection values ​​collected by the camera array after validity verification. Specifically, it can be achieved by using an image processing algorithm to perform weighted average calculation on the change in the coordinates of the QR code target. Its function is to eliminate the measurement error of a single camera and improve the reliability of the settlement data.

[0028] Among them, the settlement threshold refers to the maximum allowable settlement set according to the building structure safety standard. Specifically, the percentage value of the foundation deformation limit in the building code can be used as the benchmark value, for example, it can be set to 5 mm. Its function is to provide a quantitative basis for automated risk judgment.

[0029] Step S4: Perform settlement verification analysis on the building foundation observation point: The detection period is composed of the moments of the two most recent image acquisitions by the camera. The horizontal angle change of the observation point during the detection period is obtained by the inclinometer and marked as the settlement verification value. The settlement verification value is compared with a preset settlement verification threshold: if the settlement verification value is less than the settlement verification threshold, it is determined that the observation point does not have a settlement risk; if the settlement verification value is greater than or equal to the settlement verification threshold, it is determined that the observation point has a settlement risk. The detection period refers to the time interval between two adjacent image acquisitions, which can be determined by the timestamp data of the camera. It is used to limit the calculation period of the settlement verification value and ensure the timeliness of the data.

[0030] Among them, the inclinometer refers to a sensor used to measure the horizontal angle change of the observation point. It can be implemented by a high-precision electronic inclinometer. By continuously recording the angle change, it provides physical parameters independent of visual detection for settlement verification.

[0031] Among them, the settlement verification value refers to the cumulative amount of horizontal angle changes during the detection period, which can be calculated by the difference in angle data output by the inclinometer. It is used to reflect the degree of foundation inclination and serves as an auxiliary verification indicator for visual settlement detection.

[0032] Among them, the settlement verification threshold refers to a pre-set critical value of angle change, which can be determined based on foundation structure safety standards or historical data, and is used to determine whether the foundation inclination exceeds the allowable range.

[0033] Specifically, in the settlement verification analysis process, the detection period is first determined based on the time point when the camera captures the image, for example, the interval between two adjacent shots is 24 hours. Subsequently, the horizontal angle change within the period is obtained through the inclinometer installed at the observation point, for example, the cumulative tilt angle is 0.5 degrees. This change is compared with the preset settlement verification threshold, for example, the threshold is 0.3 degrees. If the detection value exceeds the threshold, it is determined that there is a verification settlement risk, indicating that the foundation may tilt; if it does not exceed the threshold, it is determined that the risk is controllable. This process introduces inclinometer data to form a double verification with the visual inspection results, which can effectively identify visual inspection errors caused by camera failure or environmental interference.

[0034] Compared with existing technologies, existing methods rely on a single visual inspection algorithm, which is prone to misjudgment due to equipment anomalies or environmental interference. However, this solution combines physical inclinometer measurement data to build a multi-source verification mechanism, which can distinguish between true settlement and equipment errors. For example, even if camera lens contamination causes visual inspection anomalies, inclinometer data can still provide reliable judgment, thereby reducing the false alarm rate.

[0035] Step S5: Output the settlement detection result of the building foundation observation point: if the observation point has both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is determined to be unqualified, a settlement warning signal is generated, and the settlement warning signal is sent to the server; if the observation point does not have both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is determined to be qualified, a qualified detection signal is generated, and the qualified detection signal is sent to the server; otherwise, it is determined that there is a dispute over the settlement detection result of the observation point, the corresponding observation point is marked as a verification point, a manual verification signal is generated, and the manual verification signal is sent to the server; after receiving the manual verification signal, the server sends the manual verification signal to the mobile phone terminal of the administrator, and the administrator manually verifies the observation point and marks the visual settlement detection process or the settlement verification process as a correct process; Among them, visual settlement risk refers to the situation where the settlement value calculated by the change in the coordinates of the QR code target collected by the camera array exceeds the preset threshold. This can be achieved by using an image processing algorithm to perform pixel-level analysis on the detection image to reflect the degree of structural deformation based on visual data. Verification settlement risk refers to the situation where the change in the horizontal angle measured by the inclinometer exceeds the verification threshold. This can be achieved by using a high-precision inclinometer to collect data in real time to verify the physical accuracy of the visual detection results. The manual verification signal refers to the review instruction triggered when there is a disagreement between the two detection results. This can be achieved through the communication interface between the server and the mobile terminal to initiate the manual intervention mechanism to ensure the reliability of the detection.

[0036] Specifically, when both the visual settlement detection and inclination verification results exceed a threshold, the foundation is directly judged to be at risk of settlement and an early warning is triggered. When neither exceeds the threshold, the structural safety status is confirmed. When conflicting detection results are found, the system automatically marks the disputed observation point and sends it to management personnel for on-site review. Management personnel use on-site measurements to confirm which process, visual detection or inclination verification, contains the error and feed the correct detection method back to the system for recording. This multimodal detection result cross-validation mechanism effectively avoids misjudgments caused by single sensor failures or environmental interference.

[0037] Example 2: Figure 2 As shown, the instant decision sub-method includes the following steps: Step P1: Statistical analysis of manual verification results: Generate a statistical period and obtain the detection value JCi of the environmental parameter i at the time of the verification point marking within the statistical period, where i=1, 2, ..., n, where n is a positive integer. The environmental parameter i includes the air temperature value, air humidity value, fog concentration value, rainfall, wind force level, etc. at the verification point; The statistical period refers to the time interval used for data collection. It can be fixed or dynamically adjusted, for example, set to 24 hours or automatically adjusted based on the number of verification points. Environmental parameters i refer to the set of external factors that may affect the operating status of the detection equipment. Data collection can be achieved through devices such as temperature sensors, humidity sensors, optical sensors, rain gauges, and anemometers. The detection value JCi refers to the actual measured value of each environmental parameter at the moment the verification point is marked. It can be transmitted to the data processing system in real time via IoT devices.

[0038] Specifically, during the manual verification results statistics phase, a time-based statistical period is first established as a data collection window. During this period, when an observation point is marked as a verification point, multi-dimensional environmental parameter data corresponding to that moment is simultaneously collected. For example, when an observation point is marked due to a disputed test result, the system automatically records the temperature, humidity, and other environmental conditions at that time. By temporally associating environmental parameters with verification events, a data set encompassing environmental factors is formed, providing fundamental data support for subsequent analysis of the extent to which different detection methods are affected by the environment.

[0039] Step P2: Perform environmental correlation analysis on the manual verification results: When the visual settlement detection process is marked as a correct process, the corresponding verification point is marked as a visual correlation point; when the settlement verification process is marked as a correct process, the corresponding verification point is marked as a verification correlation point; perform data cleaning on the visual correlation point: the detection value JCi of the environmental parameter i of all visual correlation points constitutes the visual set i of the environmental parameter i, and the variance calculation of all elements of the visual set i is performed to obtain the visual concentration value SJi of the environmental parameter i, and the visual concentration value SJi is compared with the preset concentration threshold. The visual concentration value SJi is compared with the value JZi: if the visual concentration value SJi is greater than or equal to the concentration threshold JZi, the maximum element and the minimum element in the visual set i are eliminated, and then the visual concentration value SJi is recalculated, and so on, until the visual concentration value SJi is less than the concentration threshold JZi; if the visual concentration value SJi is less than the concentration threshold JZi, the maximum element and the minimum element retained in the visual set i constitute the visual priority range SYi of the environmental parameter i; the same method is used to perform data cleaning on the verification association points to obtain the verification priority range HYi of the environmental parameter i; The visual association point refers to the verification point corresponding to when the visual settlement detection process is marked as a correct process. Specifically, it can be achieved by associating the correct detection results with the coordinates of the verification points, and is used for subsequent analysis of the environmental parameter range applicable to the visual detection method. The verification association point refers to the verification point corresponding to when the settlement verification process is marked as a correct process. Specifically, it can be achieved by associating the correct verification results with the coordinates of the verification points, and is used to analyze the environmental parameter range applicable to the verification method. Data cleaning processing refers to the screening of detection data through variance calculation and outlier removal. Specifically, it can be achieved by using an iterative variance calculation method to eliminate noise data in environmental parameters. The visual priority range SYi refers to the reasonable range of environmental parameters corresponding to the visual association points retained after data cleaning. Specifically, it can be achieved by retaining the maximum and minimum detection values ​​after data cleaning to form a range interval, and is used to characterize the environmental conditions applicable to the visual detection method. The verification priority range HYi refers to the reasonable range of environmental parameters corresponding to the verification association points retained after data cleaning. Specifically, it can be achieved by using the same construction method as the visual priority range, and is used to characterize the environmental conditions applicable to the settlement verification method.

[0040] Specifically, when the visual settlement detection process is verified to be correct, the corresponding verification point is classified as a visual association point, and its environmental parameter data is included in the visual set for cleaning. By calculating the variance of the environmental parameters in the visual set, if the variance exceeds the preset threshold, the maximum and minimum detection values ​​are eliminated, and this process is repeated until the variance meets the requirements. The final retained data interval is defined as the visual priority range SYi. For the verification association point, the same data cleaning logic is used to generate the verification priority range HYi. Therefore, in the subsequent decision-making process, the corresponding detection result can be automatically selected as the effective basis based on the priority range of the environmental parameters.

[0041] Step P3: Generate an immediate decision result: When there is a dispute over the settlement detection result of the observation point later, obtain the detection value JCi of the environmental parameter i of the observation point, and determine whether the detection value JCi of the environmental parameter i is all within the corresponding visual priority range SYi: If so, send the visual settlement detection result in step S3 to the server; if not, determine whether the detection value JCi of the environmental parameter i is all within the corresponding verification priority range HYi: If so, send the settlement verification result in step S4 to the server; if not, generate a manual verification signal and send the manual verification signal to the server; When a discrepancy occurs between the visual inspection and verification results at an observation point, the system automatically extracts the current environmental parameters and compares them with the priority ranges established by historical verification. If all environmental parameters fall within the visual priority range, the current environment meets the applicable conditions for visual inspection, and the visual inspection result is prioritized. If the visual conditions are not met but the verification priority range is met, the verification result is used as the final judgment. If environmental parameters fall outside both priority ranges, the system automatically triggers a manual verification process to avoid misjudgments that may result from using a single inspection result in abnormal environments.

[0042] A computer vision-based building foundation settlement detection method. During operation, the detection and verification sub-method performs visual settlement detection and analysis on the building foundation, evaluates the validity of the detection results, performs risk assessment and verification analysis, and ultimately outputs a comprehensive detection result. The immediate decision-making sub-method statistically analyzes environmental parameters from manual verification records to establish a correlation model between detection methods and environmental conditions. In subsequent dispute scenarios, the optimal detection result output method is automatically selected.

[0043] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

[0044] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0045] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A building foundation settlement detection method based on computer vision, characterized in that: Including detection check sub-method and instant decision sub-method; The method for detecting syndromes comprises the following steps: Step S1: Perform visual settlement detection and analysis on the building foundation; Step S2: Analyze the validity of the visual settlement detection results of the building foundation observation points; Step S3: Conduct visual settlement risk analysis on building foundation observation points; Step S4: performing settlement verification analysis on the building foundation observation points; Step S5: Outputting the settlement detection results of the building foundation observation points; The instant decision-making sub-method comprises the following steps: Step P1: Statistical analysis of manual verification results; Step P2: Conduct environmental correlation analysis on the manual verification results; Step P3: Generate immediate decision results.

2. A building foundation settlement detection method based on computer vision according to claim 1, characterized in that: In step S1, the specific process of visual settlement detection and analysis includes: arranging a camera array at the observation point of the building, regularly capturing images of the observation point through the camera and marking the captured images as detection images, amplifying the detection images into pixel grid images and extracting the coordinates of the QR code targets in the detection images, and marking the distance between the two most recently extracted QR code target coordinate points of the same camera as the settlement detection value of the camera.

3. A computer vision-based building foundation settlement detection method according to claim 2, characterized in that: In step S2, the specific process of analyzing the effectiveness of the visual settlement detection results includes: performing variance calculation on the settlement detection values ​​collected by all cameras in the camera array of the same observation point to obtain the effective coefficient of the observation point, and comparing the effective coefficient with the preset effective threshold: if the effective coefficient is less than the effective threshold, it is determined that the effectiveness of the visual settlement detection result of the observation point meets the requirements, and the settlement detection values ​​collected by all cameras in the camera array of the observation point are summed and averaged to obtain the visual settlement value; if the effective coefficient is greater than or equal to the effective threshold, it is determined that the effectiveness of the visual settlement detection result of the observation point does not meet the requirements, and a camera maintenance signal is generated and sent to the mobile phone terminal of the administrator.

4. A computer vision-based building foundation settlement detection method according to claim 3, characterized in that: In step S3, the specific process of visual subsidence risk analysis includes: comparing the visual subsidence value of the observation point with a preset subsidence threshold: if the visual subsidence value is less than the subsidence threshold, it is determined that the observation point does not have a visual subsidence risk; if the visual subsidence value is greater than or equal to the subsidence threshold, it is determined that the observation point has a visual subsidence risk.

5. The computer vision-based building foundation settlement detection method according to claim 4, characterized in that: In step S4, the specific process of settlement verification analysis includes: the detection period is composed of the moments when the camera performs the two most recent image acquisitions, the horizontal angle change of the observation point during the detection period is obtained through the inclinometer and marked as the settlement verification value, and the settlement verification value is compared with the preset settlement verification threshold: if the settlement verification value is less than the settlement verification threshold, it is determined that the observation point does not have the verification settlement risk; if the settlement verification value is greater than or equal to the settlement verification threshold, it is determined that the observation point has the verification settlement risk.

6. The computer vision-based building foundation settlement detection method according to claim 5, characterized in that: In step S5, the specific process of outputting the settlement detection result includes: if the observation point has both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is judged to be unqualified, a settlement warning signal is generated and the settlement warning signal is sent to the server; if the observation point does not have both visual settlement risk and verification settlement risk, the settlement detection result of the observation point is judged to be qualified, a qualified detection signal is generated and the qualified detection signal is sent to the server; otherwise, it is judged that there is a dispute over the settlement detection result of the observation point, the corresponding observation point is marked as a verification point, a manual verification signal is generated and the manual verification signal is sent to the server; after receiving the manual verification signal, the server sends the manual verification signal to the mobile phone terminal of the administrator, and the administrator manually verifies the observation point and marks the visual settlement detection process or the settlement verification process as the correct process.

7. The computer vision-based building foundation settlement detection method according to claim 6, characterized in that: In step P1, a statistical period is generated, and the detection value JCi of the environmental parameter i when the verification point is marked within the statistical period is obtained, where i=1, 2,…, n, and n is a positive integer. The environmental parameter i includes the air temperature value, air humidity value, fog concentration value, rainfall, and wind level of the verification point.

8. The computer vision-based building foundation settlement detection method according to claim 7, characterized in that: In step P2, the specific process of environmental association analysis includes: when the visual settlement detection process is marked as a correct process, the corresponding verification point is marked as a visual association point; when the settlement verification process is marked as a correct process, the corresponding verification point is marked as a verification association point; data cleaning processing is performed on the visual association point to obtain the visual priority range SYi of the environmental parameter i; and data cleaning processing is performed on the verification association point in the same way to obtain the verification priority range HYi of the environmental parameter i.

9. The computer vision-based building foundation settlement detection method according to claim 8, characterized in that: In step P2, the specific process of data cleaning processing of visual association points includes: forming a visual set i of environmental parameter i from the detection values ​​JCi of environmental parameter i of all visual association points, performing variance calculation on all elements of visual set i to obtain a visual concentration value SJi of environmental parameter i, and comparing the visual concentration value SJi with a preset concentration threshold JZi: if the visual concentration value SJi is greater than or equal to the concentration threshold JZi, then the maximum element and the minimum element in the visual set i are eliminated, and then the visual concentration value SJi is recalculated, and so on, until the visual concentration value SJi is less than the concentration threshold JZi; if the visual concentration value SJi is less than the concentration threshold JZi, then the maximum element and the minimum element retained in the visual set i constitute the visual priority range SYi of environmental parameter i.

10. The computer vision-based building foundation settlement detection method according to claim 9, characterized in that: In step P3, the process of generating an instant decision result includes: when there is a dispute over the settlement detection result of the observation point in the future, obtaining the detection value JCi of the environmental parameter i of the observation point, and determining whether the detection value JCi of the environmental parameter i is all within the corresponding visual priority range SYi: if so, sending the visual settlement detection result in step S3 to the server; if not, determining whether the detection value JCi of the environmental parameter i is all within the corresponding verification priority range HYi: if so, sending the settlement verification result in step S4 to the server; if not, generating a manual verification signal and sending the manual verification signal to the server.

Images