Computer vision-based intelligent monitoring and data processing system for building deformation
By using a computer vision-based monitoring system, the problems of real-time and accuracy in deformation monitoring during building construction have been solved, achieving efficient and intelligent structural deformation detection and reducing costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI CONSTRUCTION FOURTH CONSTRUCTION GROUP CO LTD
- Filing Date
- 2023-04-07
- Publication Date
- 2026-06-02
AI Technical Summary
Existing methods for monitoring deformation in building construction cannot achieve real-time, accurate, and efficient monitoring, especially in high-altitude environments where effective structural deformation detection is difficult and costly.
A computer vision-based monitoring system is adopted, including camera calibration, automatic aperture adjustment, monitoring position tilt correction and intelligent data processing modules. Combined with displacement and tilt sensors, it realizes non-contact monitoring and intelligent data processing of building structures.
It achieves high-precision and intelligent building deformation monitoring, reduces manpower and material costs, and can monitor and plot structural displacement-time curves in real time, improving monitoring accuracy and efficiency.
Smart Images
Figure CN116358436B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a computer vision-based intelligent monitoring and data processing system for building deformation. Background Technology
[0002] Construction often involves various procedures such as dismantling formwork and tensioning cables, which inevitably cause structural deformation. In some cases, structural deformation may even present potentially dangerous conditions. Currently, structural deformation monitoring during construction primarily utilizes total stations or hydrostatic levels to periodically measure deformation. However, existing monitoring hardware is limited. Total stations cannot monitor structural deformation in real time, only providing initial and final readings. Furthermore, the measurement process is cumbersome, the monitoring frequency is low, and significant manpower and material costs are required. Hydrostatic levels, being contact-based measurements, are unusable when the object being monitored is at height or difficult to access. Summary of the Invention
[0003] The purpose of this invention is to provide a computer vision-based intelligent monitoring and data processing system for building deformation.
[0004] To address the above problems, this invention provides a computer vision-based intelligent monitoring and data processing system for building deformation, comprising:
[0005] The monitoring setting module is used to calibrate the camera, adjust the aperture of the camera lens, use the calibrated camera with the adjusted aperture to acquire images, and track the position of monitoring points in the acquired images to obtain monitoring results. The monitoring points include: the measured structure or landmark of the building.
[0006] The monitoring position tilt correction module is used to correct the monitoring results based on the displacement and tilt sensors and the reference point;
[0007] The data intelligent processing module is used to determine whether there are any abnormalities at individual monitoring points based on the corrected location of the monitoring points, draw displacement and time curves for each monitoring point, and display them.
[0008] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the monitoring setting module includes:
[0009] The camera calibration unit is used to calibrate using the homography matrix method to find the projection relationship between the three-dimensional world coordinates and the two-dimensional image coordinates, so as to realize the transformation of each point in the image to the three-dimensional world.
[0010] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the monitoring setting module further includes:
[0011] The monitoring principle unit is used to acquire grayscale or color images. It directly uses the grayscale value of the acquired grayscale image or the grayscale value of the acquired color image after it has been converted into a grayscale image. Based on the grayscale value and according to the selected image features, it performs position tracking on the measured structure or its marker to determine the position of the measured structure or its marker in each frame of the video or image sequence. Finally, it calculates the motion of the measured structure or its marker in the image. Infrared lights are installed at the measured structure or its marker.
[0012] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the monitoring setting module further includes:
[0013] The automatic aperture adjustment unit is used to automatically adjust the aperture size according to the light intensity, so that the imaging brightness in a fixed area near the infrared lamp of the measured structure or its marker is always a constant value.
[0014] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the monitoring position tilt angle correction module includes:
[0015] The self-sensor correction module is used to obtain the position changes of the monitoring setting module during the monitoring process based on the displacement and angular displacement sensors built into the monitoring setting module, including: vertical displacement x and tilt displacement θ; and to correct the monitoring results according to the position changes. The corrected monitoring result = original monitoring result - θ*L1 / cosφ-x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
[0016] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the monitoring position tilt angle correction module further includes:
[0017] The reference point correction unit is used to determine the tilt displacement θ = dx * cosφ / L of the monitoring setting module when the reference point position dx of the monitoring setting module changes, where L is the distance between the monitoring setting module and the reference point. The unit corrects the monitoring result based on the position change. The corrected monitoring result = original monitoring result - θ * L1 / cosφ - x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
[0018] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the intelligent data processing module includes:
[0019] The data anomaly monitoring unit is used to obtain the structural displacement of the associated monitoring points based on the corrected location of the monitoring point and the location of the associated monitoring points, determine the linear relationship between the monitoring point and its associated monitoring points based on the structural displacement, and determine whether there is an anomaly at the monitoring point based on the linear relationship.
[0020] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the intelligent data processing module further includes:
[0021] The automatic correction unit is used to check whether there is any abnormal displacement after the correction of the monitoring point position. If so, the displacement after the abnormal displacement is set to the displacement before the abnormal displacement.
[0022] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the intelligent data processing module further includes:
[0023] The noise reduction processing unit is used to generate the time and displacement curves of the monitoring points based on the corrected position of the monitoring points, and to eliminate the spikes in the time and displacement curves by means denoising or Kalman filtering denoising.
[0024] Furthermore, in the aforementioned computer vision-based intelligent monitoring and data processing system for building deformation, the intelligent data processing module further includes:
[0025] The platform visualization unit is used to send the time and displacement curves of the monitoring points after burr removal to the mobile device.
[0026] Compared with existing technologies, the monitoring setup module in this invention is a summary process of methods before monitoring, mainly including: a camera calibration unit, a monitoring principle unit, and an automatic aperture adjustment unit. The monitoring position tilt correction module is the self-position correction process during monitoring, mainly including two parts: a self-sensor correction unit and a reference point correction unit. The intelligent data processing module is the data processing process during monitoring, mainly divided into a data anomaly monitoring unit, an automatic correction unit, a noise reduction processing unit, and a platform visualization unit.
[0027] Compared to current deformation monitoring and data processing systems, the main advantages of this invention are as follows:
[0028] (1) High monitoring accuracy. Based on computer vision methods, this invention features an automatic aperture adjustment and monitoring position tilt correction module, which can reduce the impact of its own vibration through algorithms, thus achieving high accuracy;
[0029] (2) Intelligent deformation measurement and processing. This invention combines the displacement correlation between monitoring points to automatically analyze abnormal situations in the monitoring data, and completes intelligent correction and noise reduction of the data. It establishes a platform to draw displacement-time curves, visualizes structural displacement, and realizes intelligent deformation measurement and processing. Attached Figure Description
[0030] Figure 1 This is a schematic diagram of a computer vision-based intelligent monitoring and data processing system for building deformation according to an embodiment of the present invention;
[0031] Figure 2 This is a schematic diagram of a monitoring setting module according to an embodiment of the present invention;
[0032] Figure 3 This is a schematic diagram of a monitoring position tilt correction module according to an embodiment of the present invention;
[0033] Figure 4 This is a schematic diagram of a data intelligent processing module according to an embodiment of the present invention;
[0034] Figure 5 This is a schematic diagram of a computer vision-based intelligent monitoring and data processing system for building deformation, according to an embodiment of the present invention. Detailed Implementation
[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0036] like Figures 1 to 5 As shown, this invention provides a computer vision-based intelligent monitoring and data processing system for building deformation, comprising:
[0037] The monitoring setting module is used to calibrate the camera, adjust the aperture of the camera lens, use the calibrated camera with the adjusted aperture to acquire images, and track the position of monitoring points in the acquired images to obtain monitoring results. The monitoring points include: the measured structure or landmark of the building.
[0038] The monitoring position tilt correction module is used to correct the monitoring results based on the displacement and tilt sensors and the reference point;
[0039] The data intelligent processing module is used to determine whether there are any abnormalities at individual monitoring points based on the corrected location of the monitoring points, draw displacement and time curves for each monitoring point, and display them.
[0040] This invention addresses the problems of cumbersome deformation monitoring and measurement processes, low monitoring frequency, and high manpower and material costs in existing building construction processes. It proposes a new intelligent construction deformation monitoring method based on non-contact monitoring, integrating intelligent monitoring and data processing. This method can effectively monitor structural deformation during building construction in a highly applicable, stable, and labor-saving manner. At the same time, it comprehensively and intelligently processes relevant data information, greatly improving monitoring accuracy.
[0041] like Figure 2 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the monitoring setting module includes:
[0042] The camera calibration unit is used to calibrate using the homography matrix method to find the projection relationship between the three-dimensional world coordinates and the two-dimensional image coordinates, so as to realize the transformation of each point in the image to the three-dimensional world.
[0043] Here, in the camera calibration unit, the three-dimensional world coordinates will be distorted in the projected image, including central perspective and radial distortion. The camera calibration process is also a process of correcting the distorted image. The homography matrix method is used to find the projection relationship from the three-dimensional world coordinates to the two-dimensional image coordinates, so as to realize the transformation of each point in the image to the three-dimensional world.
[0044] like Figure 2 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the monitoring setting module further includes:
[0045] The monitoring principle unit is used to acquire grayscale or color images. It directly uses the grayscale value of the acquired grayscale image or the grayscale value of the acquired color image after it has been converted into a grayscale image. Based on the grayscale value and according to the selected image features, it performs position tracking on the measured structure or its marker to determine the position of the measured structure or its marker in each frame of the video or image sequence. Finally, it calculates the motion of the measured structure or its marker in the image. Infrared lights are installed at the measured structure or its marker.
[0046] Here, in terms of image tracking and pixel calculation, the grayscale values of the acquired grayscale images or the grayscale values of the acquired black-color images converted to grayscale are directly used. Based on the grayscale values and the selected image features, the position of the measured structure or its markers is tracked to determine the position of the measured structure or its markers in each frame of the video or image sequence. Finally, the motion of the measured structure or its markers in the image is calculated. In particular, considering the complexity of the construction site environment, this method selects infrared lights as the objects to be monitored to avoid the influence of the external environment.
[0047] like Figure 2 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the monitoring setting module further includes:
[0048] The automatic aperture adjustment unit is used to automatically adjust the aperture size according to the light intensity, so that the imaging brightness in a fixed area near the infrared lamp of the measured structure or its marker is always a constant value.
[0049] Here, the automatic aperture adjustment unit takes into account the influence of sunlight on infrared lamp imaging in outdoor environments and automatically adjusts the aperture size according to the light intensity: by connecting the camera to the computer, the aperture size of the lens is controlled so that the imaging brightness in a fixed area near the infrared lamp at the measured structure or its marker is always a constant value, thereby minimizing the influence of infrared components in natural light on construction monitoring.
[0050] The monitoring setup module is a summary process of methods before monitoring, mainly including a camera calibration unit, a monitoring principle unit, and an automatic aperture adjustment unit.
[0051] like Figure 3 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the monitoring position tilt angle correction module includes:
[0052] The self-sensor correction module is used to obtain the position changes of the monitoring setting module during the monitoring process based on the displacement and angular displacement sensors built into the monitoring setting module, including: vertical displacement x and tilt displacement θ; and to correct the monitoring results according to the position changes. The corrected monitoring result = original monitoring result - θ*L1 / cosφ-x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
[0053] Here, the self-sensor correction module corrects the monitoring results based on the displacement and tilt sensors built into the monitoring setting module. The displacement and angular displacement sensors built into the monitoring setting module can obtain the positional changes of the monitoring setting module during the monitoring process, including: vertical displacement x and tilt displacement θ; the monitoring results are corrected accordingly, and the corrected monitoring result = original monitoring result - θ*L1 / cosφ-x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
[0054] In one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the monitoring position tilt angle correction module further includes:
[0055] The reference point correction unit is used to determine the tilt displacement θ = dx * cosφ / L of the monitoring setting module when the reference point position dx of the monitoring setting module changes, where L is the distance between the monitoring setting module and the reference point. The unit corrects the monitoring result based on the position change. The corrected monitoring result = original monitoring result - θ * L1 / cosφ - x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
[0056] Here, the reference point correction unit determines, based on existing knowledge, a point where the monitoring module itself does not change position during the monitoring process, as the reference point. When the position of the reference point of the monitoring module changes by dx, the tilt displacement θ of the monitoring module can be determined as dx*cosφ / L, where L and φ have the same meaning as before. Based on this, the tilt correction module function can be implemented using both the sensor correction section and the reference point correction section.
[0057] The monitoring position tilt correction module is a self-position correction process during the monitoring process, which mainly includes two parts: the self-sensor correction unit and the reference point correction unit.
[0058] like Figure 4 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the intelligent data processing module includes:
[0059] The data anomaly monitoring unit is used to obtain the structural displacement of the associated monitoring points based on the corrected location of the monitoring point and the location of the associated monitoring points, determine the linear relationship between the monitoring point and its associated monitoring points based on the structural displacement, and determine whether there is an anomaly at the monitoring point based on the linear relationship.
[0060] Here, since the method for monitoring structural displacement is based on computer vision, individual monitoring points may be affected by factors other than the construction process. Combining the characteristics before, during and after the construction process, the structural displacements of related points (such as parts in close proximity) will have a linear correlation. Therefore, the structural displacements of appropriate related monitoring points are selected, and their linear correlation coefficient R (distributed between -1 and 1) is calculated. Combined with certain empirical values, it is determined whether there are any abnormalities in individual monitoring points.
[0061] like Figure 4 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the intelligent data processing module further includes:
[0062] The automatic correction unit is used to check whether there is any abnormal displacement after the correction of the monitoring point position. If so, the displacement after the abnormal displacement is set to the displacement before the abnormal displacement.
[0063] Here, the abnormal situations of the automatic correction unit mainly include two types of data mutations: sudden overall data mutation and data mutation followed by a return to the original value. The automatic correction program is set to correct the mutated data. The displacement after the abnormal mutation is set to the state before the abnormal mutation, so as to correct the abnormal mutation. When abnormal mutations occur frequently, it can be combined with the actual situation on site to check whether it is caused by frequent occlusion, etc.
[0064] like Figure 4 As shown, in one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the intelligent data processing module further includes:
[0065] The noise reduction processing unit is used to generate the time and displacement curves of the monitoring points based on the corrected position of the monitoring points, and to eliminate the spikes in the time and displacement curves by means denoising or Kalman filtering denoising.
[0066] Here, the noise reduction unit addresses the issue of environmental noise during the monitoring process, which can cause spikes in the curve. Therefore, methods such as mean denoising and Kalman filtering are used to eliminate these spikes and smooth the displacement curve.
[0067] In one embodiment of the computer vision-based intelligent monitoring and data processing system for building deformation of the present invention, the intelligent data processing module further includes:
[0068] The platform visualization unit is used to send the time and displacement curves of the monitoring points after burr removal to the mobile device.
[0069] Here, the platform visualization unit transmits the structural displacement monitored by the monitoring settings module to the data display platform, and plots the displacement-time curves of each monitoring point. Based on this, the real-time dynamic displacement curves of the structure can be visualized and viewed on mobile devices such as mobile phones and tablets.
[0070] The data intelligent processing module is responsible for the data processing during the monitoring process. It is mainly divided into a data anomaly monitoring unit, an automatic correction unit, a noise reduction processing unit, and a platform visualization unit.
[0071] In summary, the monitoring setup module of this invention summarizes the methods before monitoring, mainly including: a camera calibration unit, a monitoring principle unit, and an automatic aperture adjustment unit. The monitoring position tilt correction module is the self-position correction process during monitoring, mainly including a self-sensor correction unit and a reference point correction unit. The intelligent data processing module handles the data processing during monitoring, mainly divided into a data anomaly monitoring unit, an automatic correction unit, a noise reduction unit, and a platform visualization unit.
[0072] Compared to current deformation monitoring and data processing systems, the main advantages of this invention are as follows:
[0073] (1) High monitoring accuracy. Based on computer vision methods, this invention features an automatic aperture adjustment and monitoring position tilt correction module, which can reduce the impact of its own vibration through algorithms, thus achieving high accuracy;
[0074] (2) Intelligent deformation measurement and processing. This invention combines the displacement correlation between monitoring points to automatically analyze abnormal situations in the monitoring data, and completes intelligent correction and noise reduction of the data. It establishes a platform to draw displacement-time curves, visualizes structural displacement, and realizes intelligent deformation measurement and processing.
[0075] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0076] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0077] Obviously, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.
Claims
1. A computer vision-based intelligent monitoring and data processing system for building deformation, characterized in that, include: The monitoring setting module is used to calibrate the camera, adjust the aperture of the camera lens, use the calibrated camera with the adjusted aperture to acquire images, and track the position of monitoring points in the acquired images to obtain monitoring results. The monitoring points include: the measured structure or landmark of the building. The monitoring position tilt correction module is used to correct the monitoring results based on the displacement and tilt sensors and the reference point; The data intelligent processing module is used to determine whether there are any abnormalities at individual monitoring points based on the corrected location of the monitoring points, draw the displacement and time curves of each monitoring point, and display them. The monitoring position tilt correction module also includes: The reference point correction unit is used to determine the tilt displacement θ=dx*cosφ / L of the monitoring setting module when the reference point position dx of the monitoring setting module changes, where L is the distance between the monitoring setting module and the reference point. The unit corrects the monitoring result based on the position change. The corrected monitoring result = original monitoring result - θ*L1 / cosφ-x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
2. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 1, characterized in that, The monitoring settings module includes: The camera calibration unit is used to calibrate using the homography matrix method to find the projection relationship between the three-dimensional world coordinates and the two-dimensional image coordinates, so as to realize the transformation of each point in the image to the three-dimensional world.
3. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 2, characterized in that, The monitoring settings module also includes: The monitoring principle unit is used to acquire grayscale or color images. It directly uses the grayscale value of the acquired grayscale image or the grayscale value of the acquired color image after it has been converted into a grayscale image. Based on the grayscale value and according to the selected image features, it performs position tracking on the measured structure or its marker to determine the position of the measured structure or its marker in each frame of the video or image sequence. Finally, it calculates the motion of the measured structure or its marker in the image. Infrared lights are installed at the measured structure or its marker.
4. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 3, characterized in that, The monitoring settings module also includes: The automatic aperture adjustment unit is used to automatically adjust the aperture size according to the light intensity, so that the imaging brightness in a fixed area near the infrared lamp of the measured structure or its marker is always a constant value.
5. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 1, characterized in that, The monitoring position tilt correction module includes: The self-sensor correction module is used to obtain the position changes of the monitoring setting module during the monitoring process based on the displacement and angular displacement sensors built into the monitoring setting module, including: vertical displacement x and tilt displacement θ; and to correct the monitoring results according to the position changes. The corrected monitoring result = original monitoring result - θ*L1 / cosφ-x, where L1 is the distance between the monitoring setting module and the monitoring point, and φ is the elevation angle of the monitoring setting module.
6. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 1, characterized in that, The data intelligent processing module includes: The data anomaly monitoring unit is used to obtain the structural displacement of the associated monitoring points based on the corrected location of the monitoring point and the location of the associated monitoring points, determine the linear relationship between the monitoring point and its associated monitoring points based on the structural displacement, and determine whether there is an anomaly at the monitoring point based on the linear relationship.
7. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 6, characterized in that, The data intelligent processing module also includes: The automatic correction unit is used to check whether there is any abnormal displacement after the correction of the monitoring point position. If so, the displacement after the abnormal displacement is set to the displacement before the abnormal displacement.
8. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 7, characterized in that, The data intelligent processing module also includes: The noise reduction processing unit is used to generate the time and displacement curves of the monitoring points based on the corrected position of the monitoring points, and to eliminate the spikes in the time and displacement curves by means denoising or Kalman filtering denoising.
9. The intelligent monitoring and data processing system for building deformation based on computer vision as described in claim 1, characterized in that, The data intelligent processing module also includes: The platform visualization unit is used to send the time and displacement curves of the monitoring points after burr removal to the mobile device.