CN115471691A - Method for monitoring and managing settlement of peripheral buildings in subway station construction - Google Patents
Method for monitoring and managing settlement of peripheral buildings in subway station construction Download PDFInfo
- Publication number
- CN115471691A CN115471691A CN202210910426.5A CN202210910426A CN115471691A CN 115471691 A CN115471691 A CN 115471691A CN 202210910426 A CN202210910426 A CN 202210910426A CN 115471691 A CN115471691 A CN 115471691A
- Authority
- CN
- China
- Prior art keywords
- building
- settlement
- monitoring
- image
- subway station
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000012544 monitoring process Methods 0.000 title claims abstract description 45
- 238000010276 construction Methods 0.000 title claims abstract description 34
- 230000002093 peripheral effect Effects 0.000 title claims description 14
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims abstract description 18
- 230000009466 transformation Effects 0.000 claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 7
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 230000008859 change Effects 0.000 claims description 14
- 238000005259 measurement Methods 0.000 claims description 11
- 238000007726 management method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 claims description 3
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims 1
- 238000006073 displacement reaction Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 4
- 238000013139 quantization Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000009412 basement excavation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 238000009435 building construction Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004062 sedimentation Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a settlement monitoring and managing method for surrounding buildings in subway station construction, which is characterized in that a camera is installed at a stable position of a deep foundation pit, and the camera is used for acquiring image data of building groups close to two sides of the deep foundation pit in real time; transmitting the building group image data acquired in real time to a monitoring management center, and analyzing and processing the building group image data through a visual processing module; dividing and classifying different building areas in the building group image data through a specified algorithm to obtain masks of the different building areas; carrying out Hough transformation on the building image in the appointed mask to obtain a characteristic straight line of the building image in the pair of masks; and calculating the real settlement value of the building corresponding to the building image through the three-dimensional transformation relation. The method is suitable for being used in the construction process of subway stations with dense buildings, the buildings are synchronously monitored in real time by the aid of the multi-azimuth camera, targets do not need to be arranged, and the settlement of the dense buildings around the deep foundation pit is monitored in real time and at high precision.
Description
Technical Field
The invention relates to the field of building construction, in particular to a method for monitoring and managing settlement of surrounding buildings in subway station construction.
Background
Along with the soaring of the economic level, the population density of a plurality of cities in recent countries is sharply increased, more serious traffic jam occurs in a plurality of cities, and the construction and development of urban underground tracks are led to exceed the historical development period of other countries in the world in order to relieve the traffic travel pressure of the cities in China. The problems of surrounding complex buildings (structures), foundation pit supporting systems, surrounding soil deformation and the like under the influence of deep foundation pit excavation in urban underground rail construction become important problems to be solved urgently in subway deep foundation pit engineering.
Nowadays, the demand of underground rail transit is increasing, which brings a series of technical difficulties in monitoring, such as how to perform real-time and periodic settlement monitoring when excavation of foundation pits is performed in a building dense area. The traditional building settlement observation methods are many, and the main representative methods comprise leveling measurement, full-automatic measurement and the like. Leveling is a common method for observing building settlement, which is a method for measuring the elevation of a reference point and a settlement monitoring point by using a leveling instrument and analyzing the settlement deformation condition of a building according to the elevation change of each period of the settlement monitoring point. The method is suitable for monitoring the settlement of buildings with different types, different precision requirements and different measuring conditions, is a traditional and reliable method, and has the defects of difficult operation and easy interference. With the continuous improvement of measuring instruments, the total station is widely applied to settlement monitoring, especially the popularization and application of full-automatic tracking measuring instruments, and a wide development space is provided for all-weather, all-around and high-precision full-automatic monitoring. The full-automatic measuring method is widely applied to settlement monitoring of buildings such as dams and bridges, and compared with a leveling measuring method, the full-automatic measuring method has the advantages of high efficiency, convenience, practicability, small influence of human factors on measuring results, and relatively low precision. The fully automatic measurement method is the same as the leveling method in the principle of settlement observation. Before observation, an observation point needs to be buried in a measuring area, and a reference point needs to be buried in a stable area or a settlement area outside a settlement range. In order to verify the stability of the system through the joint measurement, at least three reference points are buried in one area, and one to two levels of elevation control nets are distributed from the reference points. The primary net is observed by a precise level gauge, the secondary net is measured by a level gauge one lower than the primary net, the observation is carried out according to a certain period, and the elevation of each observation point is obtained by a strict adjustment method. The elevation difference of a certain observation point obtained from the observation results of different dates is the settlement of the observation point in the period.
In conclusion, the existing building settlement observation method is complex in operation process, requires observation points and reference points to be preset, and is difficult to be used in building dense areas with sheltering and limited visual angles; in addition, the labor cost of periodic monitoring is high, and the requirement of real-time measurement, monitoring and management cannot be met.
Disclosure of Invention
In order to solve the problems, the invention aims to provide an economical, efficient and high-precision method for monitoring and managing the settlement of surrounding buildings in the construction of subway stations.
In order to realize the technical purpose, the scheme of the invention is as follows: the method for monitoring and managing the settlement of the peripheral building in the subway station construction comprises the following specific steps:
s1, installing at least two cameras at the stable position of a deep foundation pit, and acquiring building group image data close to two sides of the deep foundation pit in real time through the cameras;
s2, in the construction process of the deep foundation pit, transmitting the building group image data acquired in real time to a monitoring management center, and analyzing and processing the building group image data through a vision processing module in the monitoring management center;
s3, during analysis processing, different building areas in the building group image data are divided and classified through a specified algorithm to obtain masks of the different building areas;
s4, carrying out Hough transformation on the building image in the specified mask to obtain a characteristic straight line of the building image in the pair of masks, and comparing the change of the characteristic straight line of the building image before and after the change;
and S5, calculating the real settlement value of the building corresponding to the building image through the three-dimensional transformation relation.
Preferably, in step S3, the designated algorithm is set as a multitasking working network, and the loss function is selected as a weighted value of each branch network:
Loss＝L cls +L bbox +L mask
wherein L is cls For the classification loss function, for any ROI, let its prediction probability distribution of correct class be p u The classification loss is defined as: l is cls ＝-log p u ；
L bbox For the regression frame loss function, let the reference values of the lower left corner coordinate and length and width of any ROI be t = (x, y, w, h) T Predicted value is
Each component being t i ，v i (i = 1.. 4), the regression loss is defined as:
L mask is the loss function of a mask, for the kth mask, its loss function is defined as the average binary cross entropy over an ROI of size m × m:
the masks corresponding to the regions of the building are obtained by the above method, and the analysis of steps S4 to S5 is performed on the building images within the corresponding masks.
Preferably, in step S4, first, the building image in the mask is optimized, and a (ρ, θ) value corresponding to a characteristic straight line in the building image is obtained through Hough transformation;
when the (rho, theta) is not changed, the corresponding building is considered not to be settled; when the change of (rho, theta) is more remarkable, the corresponding building is considered to be settled.
Preferably, a line segment identified by Hough transformation is used for placing the building image in a coordinate system, the origin is positioned at the upper left corner of the building image, the horizontal left side is the positive direction of an x axis, the vertical downward side is the positive direction of a y axis, and the coordinates of two end points are assumed to be (x is x) 1 ，y 1 )、(x 2 ，y 2 ) (ii) a Obviously, when x 1 ＝x 2 When, there is ρ = x 1 ＝x 2 θ =0; when y is 1 ＝y 2 When, there is ρ = y 1 ＝y 2 θ = π/2; when x is 1 ≠x 2 、y 1 ≠y 2 When it comes to
From this, the (ρ, θ) value of the identified building contour in the building image can be established.
Preferably, the relationship between the coordinates X in the actual three-dimensional space and the coordinates X of the architectural image in step S5 can be expressed as follows:
wherein r is 11 ，...，r 23 Are the coordinates of the actual three-dimensional space building and the camera coordinates,
is an external reference matrix, c 1 ，...，c 3 Is a constant determined by the camera parameters, c is an internal reference matrix, tau is a scale factor, and tau is linearly related to the distance from the actual position X to the imaging plane.
A building settlement monitoring management system comprising:
the camera is used for acquiring real-time data of the building group images close to two sides of the deep foundation pit;
the measurement management center is used for collecting and storing the building group image data collected in real time;
and the visual processing module is used for analyzing and processing the building group images so as to obtain the real settlement value of the corresponding building.
A computer readable storage medium, wherein a plurality of instructions are stored, the instructions are suitable for being loaded by a processor of a terminal device and executing the method for monitoring and managing settlement of surrounding buildings in subway station construction.
A terminal device comprising a processor and a computer readable storage medium, the processor for implementing instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method for monitoring and managing the settlement of the peripheral building in the subway station construction.
The method has the advantages that the method is suitable for being used in the construction process of subway stations with dense buildings, the multi-azimuth camera is adopted to synchronously monitor the building group in real time, a target is not required to be arranged, an image classification algorithm is adopted to automatically label and classify different buildings in an image, hough transformation straight line detection is adopted, the settlement of the dense building group constructed next to the subway deep foundation pit is obtained through a three-dimensional projection relation, and the real-time and high-precision monitoring of the settlement of the dense building group around the deep foundation pit is realized.
Drawings
FIG. 1 is a diagram illustrating a process for building group graphics according to one embodiment of the present invention;
FIG. 2 is a diagram of the Hough transform result of a corresponding building according to an embodiment of the present invention;
FIG. 3 is a graph of sedimentation data of a first measurement point JGC01-01 according to an embodiment of the present invention;
FIG. 4 is a graph of settling data for a measurement point JGC01-02 in accordance with an embodiment of the present invention;
FIG. 5 is a graph of settlement data for a building complex at 4 months and 1 day according to an embodiment of the present invention;
FIG. 6 is a 5 month and 1 day settlement data graph of a building group according to an embodiment of the present invention;
FIG. 7 is a graph of settlement data for a building complex at 6 months and 1 day according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of the present invention;
FIG. 9 is a flow chart of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific embodiments.
The invention is described in further detail below with reference to the figures and the specific embodiments, as shown in figures 1-9. The following detailed description is to be construed as exemplary and not restrictive, and the terms "including" and "having" and their conventional variations are intended to cover non-exclusive inclusions in the following detailed description. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus. Any minor modifications, equivalents and improvements made to the following embodiments in accordance with the technical spirit of the present application should be included in the scope of the technical solution of the present application.
The first embodiment is as follows:
the subway station is located in the Futian area of Shenzhen city, laid along the east-west direction of the Fuhua road, has an underground three-layer single-column two-span frame structure and comprises rail connecting sections of an incoming line and an outgoing line, the total length of the station is about 626m, and the width of the station is 17-22.2m. The buried depth of the station bottom plate is about 28.5-33.5m. And covering soil for the stations to be about 4-8m. The peripheral buildings of the station are dense, the north side of the station is mostly 6-10 layers of shallow foundation buildings, the south side of the station is mostly 20-32 layers of friction pile foundation buildings, the distance between the building buildings at the two sides is about 30-37 meters, the basement of the building is about 0.8m closest to the foundation pit, and part of the building enclosure structure invades into the range of the foundation pit. Therefore, efficient real-time monitoring means need to be adopted for surrounding building groups in the construction process, and accurate monitoring data are timely provided for construction safety.
Two cameras are arranged on two sides of one end of the deep foundation pit, building group image data close to two sides of the foundation pit are collected in multiple directions, the cameras are installed at positions close to the underground diaphragm wall, a concrete base is constructed, stability of the cameras is guaranteed, meanwhile, a camera image collection card is installed, and building group image data changing along with time are collected accurately.
In the construction process of the deep foundation pit of the subway station, a camera is adopted to collect image data of a dense building group close to the foundation pit in real time, the building group image is transmitted to a monitoring management center, and the obtained image data is processed by a vision processing module through a computer vision technology to obtain settlement data of different buildings in the construction process of the deep foundation pit.
Because the buildings are dense in the construction area, the building image data comprises a large number of areas belonging to different buildings, and the distance difference between the different buildings and the cameras is too large, the buildings in different areas need to be compared and measured respectively in the settlement monitoring process, and different correction parameters are adopted. Mask RCNN is a deep learning based object recognition and image segmentation algorithm that recognizes each object in an image and its boundary extent and creates a pixel-level Mask (Mask) for it. Therefore, mask RCNN is adopted to segment and classify different areas of the video. The main network structure of Mask RCNN consists of residual error network (ResNet), feature Pyramid Network (FPN), region-generating network (RPN), region alignment (ROI align), mask prediction (Mask) and classifier (classification).
The Mask RCNN algorithm is set as a multitask working network, and the loss function of the multitask working network is selected as the weighted value of each branch network:
Loss＝L als +L bbox +L madk
wherein L is cls For the classification loss function, for any ROI, let its prediction probability distribution of correct class be p u The classification loss is defined as: l is cls ＝-log p u ；
L bbox For the regression frame loss function, let the reference values of the lower left corner coordinate and length and width of any ROI be t = (x, y, w, h) T Predicted value is
Each component being t i ，v i (i = 1.. 4), the regression loss is defined as:
L mask is the loss function of a mask, for the kth mask, its loss function is defined as the average binary cross entropy over an ROI of size m × m:
masks of different building areas are obtained through Mask RCNN, and subsequent analysis is carried out in the sub-areas after the masks are applied. In addition, because the camera is relatively static in the measurement process, only the first frame of the video needs to be classified and segmented, and the obtained mask can be continuously applied to all video frames.
And processing the acquired image data of the building groups in the construction process by adopting a Mask RCNN method, and performing region segmentation on different building groups, wherein different buildings are segmented by masks with different color blocks as shown in the second picture of figure 1.
In the masked area, the displacement of the building is reflected in the change of the outer contour. In order to monitor the displacement condition of the building in real time, the characteristic straight lines of the building in the area are identified through Hough transformation, whether the structure generates displacement or not is judged through comparing the parameters of the characteristic straight lines before and after the structure generates displacement, and further the real settlement of the building can be obtained through calculation of a three-dimensional transformation relation.
The principle of Hough transformation is mainly to map information in an image space into a parameter space, establish the relation between a straight line in the image space and a midpoint in the parameter space, and transfer the problem which is difficult to solve in the image space into the parameter space for solving. The straight line can be expressed as:
ρ＝xcosθ+ysinθ；
where ρ is the distance from the origin to the closest point on the line and θ is the angle between the x-axis and the line connecting the origin and the closest point. Thus, each line of the image may be associated as a pair (ρ, θ). The (ρ, θ) plane is called Hough space and represents a two-dimensional set of straight lines.
Due to the influence of illumination intensity, line segment positioning is not easy to position to the same end point every time. Therefore, the value of (ρ, θ) of the straight line where the line segment is located is more robust than the coordinate change of the end point. When the (ρ, θ) change is not significant, it can be considered that the building has not settled. When the (rho, theta) change is obvious, image frames with obvious line segment end point change in the video can be removed, and coordinate values of the end points of the residual image are used as the basis of building settlement change.
For a characteristic straight line segment identified by Hough transformation in an image, considering that in an image coordinate system, an origin is positioned at the upper left corner of the image, and the horizontal left is the positive direction of an x axisThe vertical direction is the positive y-axis direction, and the coordinates of two end points are assumed to be (x) 1 ，y 1 )、(x 2 ，y 2 ). Obviously, when x 1 ＝x 2 When, there is ρ = x 1 ＝x 2 θ =0; when y is 1 ＝y 2 When, there is ρ = y 1 ＝y 2 θ = π/2; when x is 1 ≠x 2 、y 1 ≠y 2 When there is
From this, the (ρ, θ) value of the identified building outline in the image can be established.
By carrying out graying processing on the image in advance and carrying out contrast adjustment, the influence of illumination change and shadow shielding on characteristic straight line resolution can be effectively reduced.
If the building has local settlement or overall settlement, compared with the previously shot building group graph, the values of (rho, theta) corresponding to the characteristic straight lines on the building group graph are changed, and for the characteristic straight lines with obvious changes of the values of (rho, theta), the settlement condition of the building in a two-dimensional image plane can be obtained by using the information of the image frame which is less influenced by illumination, as shown in fig. 2.
Fig. 2 shows a distribution diagram of (r, θ) values of characteristic lines of a building. When actual processing is carried out, a certain constraint condition is added to r and theta, the (r, theta) value of the corresponding straight line can be obtained in each frame of image in a positioning mode, and the settlement information of the building is obtained through calculation according to the result obtained through the identification of each frame of image.
Since the building structure to be photographed is in a three-dimensional space and the image is only two-dimensional, in order to recover the three-dimensional information of the structure from the two-dimensional visual data, the position of the target point in the three-dimensional space in the two-dimensional image must be determined by camera calibration, which is the camera calibration. The relationship of the actual coordinate X to the image coordinate X may be expressed as follows:
wherein r is 11 ，...，r 23 Are parameters determined by the building space coordinates and the camera coordinates,
also called external reference matrix, c 1 ，...，c 3 Is a constant determined by the camera's own parameters, c is also called an internal reference matrix, τ is a scale factor, linearly related to the distance of the actual position X from the imaging plane, and τ is constant when the optical axis of the camera is perpendicular to the plane of motion of the object.
In practical tests, at least two sets of visual measurements from different angles are required to recover the three-dimensional displacement from the image. However, when the displacement of the structure is relatively small, it can be assumed that the displacement component along the optical axis direction is zero, and at this time, the two-dimensional displacement information of the image can be converted into the real three-dimensional displacement only by the corrected depth information of the camera and the structure. As shown in fig. 3-4, the settlement monitoring result of the dense building groups on both sides of the deep foundation pit is obtained through monitoring and pixel-actual displacement conversion calculation. 5-7, it can be seen that the method of the present application can accurately detect the settlement value of each building around the deep foundation pit. Compared with the traditional monitoring method, the method does not need to set a target, can monitor the settlement condition of each building in the building group at two sides of the deep foundation pit in a multi-azimuth synchronous mode by combining the camera with the vision processing module, can monitor the settlement condition one by one with high precision, and saves the construction cost. In the embodiment, the economic benefit of millions of yuan is saved. The settlement of surrounding buildings is avoided, and the property and personal safety of surrounding residents is guaranteed.
In the using process, when the camera shoots, the video frame rate and the resolution ratio are set in advance, the frame ratio is usually 16. The method comprises the steps of firstly manually clicking a screen to enable a camera to be focused on a structure to be measured, then clicking 'start', and freely determining the acquisition time according to the observation requirement. In the aspect of gray level, as the subjective quality of an image is improved, the more detail of the corresponding image is, and the gray level requirement is not so high. In the aspect of non-uniform digital image sampling, quantization needs to be performed at a place where gray level variation is prominent, and a fine quantization method is adopted. In the smoothly varying gray level region, a coarse quantization method may be employed. In practical applications, in order to reduce and avoid the influence of quantization roughness, it is necessary to eliminate the appearance of false contours so as to obtain a real image. After the monitoring data are obtained, the data are timely arranged, a time-varying curve graph of displacement or stress is drawn, the settlement variation rule of surrounding building groups in the construction process is analyzed, data support is timely provided for construction safety, and construction quality is guaranteed.
A building settlement monitoring management system comprising: the camera is used for acquiring real-time data of the building group images close to two sides of the deep foundation pit;
the measurement management center is used for collecting and storing the building group image data collected in real time;
and the visual processing module is used for analyzing and processing the building group images so as to obtain the real settlement value of the corresponding building.
A computer readable storage medium, wherein a plurality of instructions are stored, the instructions are suitable for being loaded by a processor of a terminal device and executing the method for monitoring and managing settlement of surrounding buildings in subway station construction.
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method for monitoring and managing the settlement of the peripheral building in the subway station construction.
In the specific embodiments of the present application, the size of the serial number of each process does not mean that the execution sequence necessarily occurs, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
The functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The functional elements described above may be implemented in software and sold or used as a stand-alone product, which may be stored in a memory accessible by a computer device. The technical solution of the present application, or portions thereof that substantially or substantially contribute to the prior art, or all or portions thereof, may be embodied in the form of a software product stored in a memory, which includes several requests for causing a computer device to perform some or all of the steps of the above-described methods of the various embodiments of the present application.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and any minor modifications, equivalent replacements and improvements made to the above embodiment according to the technical spirit of the present invention should be included in the protection scope of the technical solution of the present invention.
Claims (8)
1. The method for monitoring and managing the settlement of the peripheral buildings in the subway station construction is characterized by comprising the following specific steps of:
s1, installing at least two cameras at the stable position of a deep foundation pit, and acquiring building group image data close to two sides of the deep foundation pit in real time through the cameras;
s2, in the construction process of the deep foundation pit, transmitting the building group image data acquired in real time to a monitoring management center, and analyzing and processing the building group image data through a vision processing module in the monitoring management center;
s3, during analysis processing, different building areas in the building group image data are divided and classified through a specified algorithm to obtain masks of the different building areas;
s4, carrying out Hough transformation on the building image in the specified mask to obtain a characteristic straight line of the building image in the pair of masks, and comparing the change of the characteristic straight line of the building image before and after the change;
and S5, calculating the real settlement value of the building corresponding to the building image through the three-dimensional transformation relation.
2. The method for monitoring and managing settlement of the peripheral building in the subway station construction as claimed in claim 1, wherein: in step S3, the assigned algorithm is set as a multitask working network, and the loss function is selected as the weighted value of each branch network:
Loss＝L cls +L bbox +L mask ；
wherein L is cls For the classification loss function, for any ROI, let its prediction probability distribution of correct class be p u The classification loss is defined as: l is a radical of an alcohol cls ＝-log p u ；
L bbox For the regression frame loss function, let the reference values of the lower left corner coordinate and length and width of any ROI be t = (x, y, w, h) T Predicted value is
Each component being t i ，v i (i = 1.. 4), the regression loss is defined as:
L mask is the loss function of a mask, for the kth mask, its loss function is defined as the average binary cross entropy over an ROI of size m × m:
the masks corresponding to the building areas are obtained by the above method, and the analysis of steps S4 to S5 is performed on the building images within the corresponding masks.
3. The method for monitoring and managing the settlement of the peripheral buildings in the construction of the subway station as claimed in claim 1, wherein: in the step S4, firstly, the building image in the mask is optimized, and the (rho, theta) value corresponding to the characteristic straight line in the building image is obtained through Hough transformation;
when (rho, theta) is not changed, the corresponding building is considered not to be settled; when the change of (rho, theta) is more remarkable, the corresponding building is considered to be settled.
4. The method for monitoring and managing the settlement of the peripheral buildings in the construction of the subway station as claimed in claim 4, wherein: a characteristic straight line segment identified by Hough transformation is arranged in a coordinate system, the original point is positioned at the upper left corner of the building image, the horizontal left direction is the positive direction of an x axis, the vertical downward direction is the positive direction of a y axis, and the coordinates of two end points are assumed to be (x is the positive direction of the x axis) 1 ，y 1 )、(x 2 ，y 2 ) (ii) a Obviously, when x is 1 ＝x 2 When, there is ρ = x 1 ＝x 2 θ =0; when y is 1 ＝y 2 When, there is ρ = y 1 ＝y 2 θ = π/2; when x is 1 ≠x 2 、y 1 ≠y 2 When there is
From this, the (ρ, θ) value of the identified building contour in the building image can be established.
5. The method for monitoring and managing the settlement of the peripheral buildings in the construction of the subway station as claimed in claim 4, wherein: the relationship between the coordinates X in the actual three-dimensional space and the building image coordinates X in step S5 can be expressed as follows:
wherein r is 11 ，...，r 23 Are the coordinates of the actual three-dimensional space building and the camera coordinates,
is a reference matrix, c 1 ，...，c 3 Is a constant determined by the camera parameters, c is an internal reference matrix, tau is a scale factor, and tau is linearly related to the distance from the actual position X to the imaging plane.
6. A building settlement monitoring management system, comprising:
the camera is used for acquiring real-time data of the building group images close to two sides of the deep foundation pit;
the measurement management center is used for collecting and storing the building group image data collected in real time;
and the visual processing module is used for analyzing and processing the building group images so as to obtain the real settlement value of the corresponding building.
7. A computer-readable storage medium characterized by: the settlement monitoring management method for the peripheral buildings in the subway station construction is characterized in that a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the settlement monitoring management method for the peripheral buildings in the subway station construction, as claimed in any one of claims 1-6.
8. A terminal device characterized by: comprising a processor and a computer readable storage medium, the processor for implementing instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method for monitoring and managing the settlement of the peripheral building in the subway station construction as claimed in any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210910426.5A CN115471691A (en) | 2022-07-29 | 2022-07-29 | Method for monitoring and managing settlement of peripheral buildings in subway station construction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210910426.5A CN115471691A (en) | 2022-07-29 | 2022-07-29 | Method for monitoring and managing settlement of peripheral buildings in subway station construction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115471691A true CN115471691A (en) | 2022-12-13 |
Family
ID=84367134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210910426.5A Pending CN115471691A (en) | 2022-07-29 | 2022-07-29 | Method for monitoring and managing settlement of peripheral buildings in subway station construction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115471691A (en) |
-
2022
- 2022-07-29 CN CN202210910426.5A patent/CN115471691A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105956942B (en) | Machine vision-based quality detection method and device for electric power pipe gallery reinforcing mesh | |
| CN111260615B (en) | Laser and machine vision fusion-based method for detecting apparent diseases of unmanned aerial vehicle bridge | |
| CN112766274A (en) | Water gauge image water level automatic reading method and system based on Mask RCNN algorithm | |
| CN113819974B (en) | River water level visual measurement method without water gauge | |
| CN109297428A (en) | A kind of high-precision deformation based on unmanned plane patrols survey technology method | |
| CN114359841B (en) | Video water level identification method based on space-time average | |
| CN110146030A (en) | Side slope surface DEFORMATION MONITORING SYSTEM and method based on gridiron pattern notation | |
| CN113011283B (en) | Non-contact type rail sleeper relative displacement real-time measurement method based on video | |
| CN105354832B (en) | A kind of method on mountain area satellite image autoregistration to geographical base map | |
| WO2022206161A1 (en) | Feature point recognition-based block movement real-time detection method | |
| CN111473776A (en) | Landslide crack monitoring method based on single-image close-range photogrammetry | |
| Liang et al. | Research on concrete cracks recognition based on dual convolutional neural network | |
| CN111452840B (en) | Railway steel rail crawling displacement detection method based on monocular vision measurement technology | |
| CN112695598A (en) | Road maintenance analysis management cloud platform based on big data and cloud computing | |
| CN112967312A (en) | Real-time robust displacement monitoring method and system for field rigid body target | |
| CN115330684A (en) | Underwater structure apparent defect detection method based on binocular vision and line structured light | |
| CN115471691A (en) | Method for monitoring and managing settlement of peripheral buildings in subway station construction | |
| CN116839539A (en) | Automatic monitoring and measuring system for tunnel engineering construction | |
| CN110969576A (en) | Highway pavement image splicing method based on roadside PTZ camera | |
| CN111784785A (en) | Bridge dynamic displacement identification method | |
| CN115761265A (en) | Method and device for extracting substation equipment in laser radar point cloud | |
| CN110969103B (en) | Method for measuring length of highway pavement disease based on PTZ camera | |
| Kye et al. | Feasibility study for the fine crack width estimation of concrete structures based on fiducial markers | |
| Zhang et al. | Research on monitoring method of remote deformation and system application based on image | |
| Xu et al. | Crack width recognition of tunnel tube sheet based on yolov8 algorithm and 3d imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |