CN113610782B - Building deformation monitoring method, equipment and storage medium - Google Patents
Building deformation monitoring method, equipment and storage medium Download PDFInfo
- Publication number
- CN113610782B CN113610782B CN202110826876.1A CN202110826876A CN113610782B CN 113610782 B CN113610782 B CN 113610782B CN 202110826876 A CN202110826876 A CN 202110826876A CN 113610782 B CN113610782 B CN 113610782B
- Authority
- CN
- China
- Prior art keywords
- marker
- image
- sub
- pixel edge
- building
- 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.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000003550 marker Substances 0.000 claims abstract description 140
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000012216 screening Methods 0.000 claims description 21
- 238000007670 refining Methods 0.000 claims 1
- 238000005259 measurement Methods 0.000 description 13
- 238000012806 monitoring device Methods 0.000 description 12
- 238000012545 processing Methods 0.000 description 9
- 238000002474 experimental method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000011514 reflex Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000004642 transportation engineering Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a building deformation monitoring method, equipment and a storage medium, wherein the method comprises the following steps: acquiring a plurality of building images containing markers, which are shot by a camera at the current moment, preprocessing the building images to extract the marker images, carrying out maximum gradient projection calculation on the marker images, and acquiring the marker maximum gradient projection images and then obtaining the whole pixel edge images of the markers according to the marker maximum gradient projection images; obtaining a sub-pixel edge image of the marker according to the whole pixel edge image; positioning the center of the marker according to the sub-pixel edge images of the markers to obtain the center coordinates of the marker; and calculating the deformation of the building according to the center coordinates of the markers at the current moment and the center coordinates of the markers at the previous moment. The invention solves the problems of low working means efficiency, difficult guarantee of precision and poor timeliness when the deformation monitoring is carried out on the building at present.
Description
Technical Field
The invention relates to the technical field of building safety monitoring, in particular to a building deformation monitoring method, equipment and a storage medium.
Background
The engineering building is easy to deform due to the reasons of dead weight, load, lifting of underground water level, insufficient geological exploration, design errors, construction quality and the like. Therefore, regular deformation monitoring of buildings is a necessary engineering task to ensure building stability and safety.
At present, deformation monitoring in practical application mostly adopts a mode of manual work and measuring instrument, and the operation means has low efficiency, difficult precision assurance and poor timeliness. In the prior art, although a deformation monitoring method based on an image processing technology appears, the deformation monitoring method is mostly limited to deformation calculation of a computer simulation image or an indoor ultra-close shooting target, and a small amount of practical image measurement technology has short shooting distance and low precision.
Disclosure of Invention
The invention aims to overcome the technical defects, and provides a method, equipment and a storage medium for monitoring deformation of a building, which solve the technical problems of low working means efficiency, difficult guarantee of precision and poor timeliness when the deformation of the building is monitored in the prior art.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a method for monitoring deformation of a building, comprising the steps of:
acquiring a plurality of building images containing markers, which are shot by a camera at the current moment, and preprocessing the building images to extract the marker images, wherein the markers are fixedly arranged on a building;
carrying out maximum gradient projection calculation on the marker image, and obtaining a whole pixel edge image of the marker according to the maximum gradient projection image of the marker after obtaining the maximum gradient projection image of the marker;
obtaining a sub-pixel edge image of the marker according to the whole pixel edge image;
positioning the center of the marker according to the sub-pixel edge images of the markers to obtain the center coordinates of the marker;
and calculating the deformation of the building according to the center coordinates of the markers at the current moment and the center coordinates of the markers at the previous moment.
Preferably, in the building deformation monitoring method, the preprocessing the building image to extract the marker image specifically includes:
binarizing the building image to extract all suspected mark areas, wherein the suspected mark areas are areas combined by black and white;
and selecting a region of interest containing the marker from the suspected marker region according to the geometric characteristics of the marker, and obtaining a marker image according to the region of interest containing the marker.
Preferably, in the building deformation monitoring method, the calculating the maximum gradient projection of the marker image, and obtaining the whole pixel edge image of the marker according to the maximum gradient projection image of the marker after obtaining the maximum gradient projection image of the marker specifically includes:
obtaining the maximum gradient projection direction of each pixel point of the marker image, and respectively calculating the maximum gradient projection value of each pixel point of the marker image in the maximum gradient projection direction to obtain the maximum gradient projection image of the marker;
preprocessing the maximum gradient projection image of the marker to obtain an initial edge point of the marker;
and carrying out refinement treatment on the initial edge points, screening out smooth integral pixel edge points, and obtaining an integral pixel edge image of the marker.
Preferably, in the building deformation monitoring method, a formula for calculating a maximum gradient projection value of each pixel point of the marker image in a maximum gradient projection direction specifically includes:
wherein,
wherein (x, y) is one of the marker imagesCoordinates of pixel point, G x And G y And respectively obtaining gradient projection values of the marker image along the horizontal direction and the vertical direction, wherein alpha is the maximum gradient projection direction.
Preferably, in the building deformation monitoring method, the obtaining the sub-pixel edge image of the marker according to the whole pixel edge image specifically includes:
calculating first-order differences in the row, column and two diagonal directions respectively in a preset neighborhood range of the center point by taking each edge point of the whole pixel edge image as the center point, and calculating sub-pixel positions of the corresponding edge point in the corresponding directions according to the first-order differences in the four directions;
and obtaining the sub-pixel edge point of the marker according to the calculated sub-pixel position of the edge point and the position of the edge point in the whole pixel edge image, and obtaining the sub-pixel edge image of the marker according to the sub-pixel edge point.
Preferably, in the building deformation monitoring method, the positioning the center of the marker according to the sub-pixel edge images of the plurality of markers to obtain the center coordinates of the marker specifically includes:
carrying out first ellipse fitting on all sub-pixel edge points in the sub-pixel edge image by adopting a preset ellipse fitting model so as to solve the ellipse fitting model;
fitting residual calculation is carried out on all sub-pixel edge points in the sub-pixel edge image according to the solved ellipse fitting model;
calculating residual standard deviation according to fitting residual errors of all sub-pixel edge points, screening all sub-pixel edge points according to the residual standard deviation, and performing ellipse fitting on the reserved sub-pixel edge points for a plurality of times to obtain final reserved sub-pixel edge points;
taking the center of the finally reserved sub-pixel edge points for elliptical fitting as the center of the sub-pixel edge image;
and obtaining the center coordinates of the markers according to the center coordinates of the plurality of sub-pixel edge images.
Preferably, in the building deformation monitoring method, the obtaining the center coordinates of the markers according to the center coordinates of the plurality of sub-pixel edge images specifically includes:
calculating the coordinate mean value and standard deviation of the center coordinates of the plurality of sub-pixel edge images;
calculating the difference value between the central coordinate of each sub-pixel edge image and the coordinate, namely the mean value, and screening the central coordinate of each sub-pixel edge image according to the difference value;
and (3) screening the reserved central coordinates for a plurality of times, and taking the coordinate mean value of a plurality of central coordinates obtained after the screening for a plurality of times as the central coordinates of the marker.
Preferably, in the building deformation monitoring method, a formula for calculating the deformation of the building is specifically:
Δx＝x 2 -x 1 ，
Δy＝y 2 -y 1 ，
wherein, (x) 1 ，y 1 ) At t 1 Obtaining the center coordinates of the markers at the moment, (x) 2 ，y 2 ) At t 2 The center coordinates of the markers obtained at the moment, Δx is the horizontal deformation amount, Δy is the vertical deformation amount, and Δd is the deformed point distance.
In a second aspect, the present invention also provides a building deformation monitoring apparatus comprising: a processor and a memory;
the memory has stored thereon a computer readable program executable by the processor;
the processor, when executing the computer readable program, implements the steps in a building deformation monitoring method as described above.
In a third aspect, the present invention also provides a computer readable storage medium storing one or more programs executable by one or more processors to implement the steps in a building deformation monitoring method as described above.
Compared with the prior art, the building deformation monitoring method, the device and the storage medium provided by the invention have the advantages that firstly, a plurality of building images containing the markers are obtained by using a camera, then, the maximum gradient projection calculation is carried out on the marker images after the marker images are extracted, the maximum gradient projection images of the markers are obtained, then, the whole pixel edge images of the markers are lifted according to the maximum gradient projection images, then, the sub-pixel edge images are extracted according to the whole pixel edge images, finally, the central coordinates of the markers are obtained according to the plurality of sub-pixel edge images, and finally, the deformation of the building is calculated according to the central coordinates of the markers at the current moment and the central coordinates of the markers at the last moment, so that the deformation monitoring of the building is realized, the accuracy is higher, and the timeliness is strong.
Drawings
FIG. 1 is a flow chart of a method for monitoring deformation of a building according to a preferred embodiment of the present invention;
FIG. 2 is a schematic diagram of a preferred embodiment of a marker of the present invention;
FIG. 3 is a schematic view of a preferred embodiment of the scale of the present invention;
FIG. 4 is a schematic view of a preferred embodiment of a marker maximum gradient projection image according to the present invention;
FIG. 5 is a schematic diagram of a preferred embodiment of a marker whole pixel edge image according to the present invention;
FIG. 6 is a schematic diagram of a preferred embodiment of the subpixel edge and the integral pixel edge of the markers of the present invention;
FIG. 7 is a schematic diagram of a preferred embodiment of the fitted marker centers;
FIG. 8 is a schematic representation of the markers under experiment 1 of the present invention;
FIG. 9 is a schematic view of the operating environment of a preferred embodiment of the building deformation monitoring program of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, the method for monitoring deformation of a building provided by the embodiment of the invention includes the following steps;
s100, acquiring a plurality of building images containing markers, which are shot by a camera at the current moment, and preprocessing the building images to extract the marker images, wherein the markers are fixedly arranged on a building;
s200, carrying out maximum gradient projection calculation on the marker image, and obtaining an integral pixel edge image of the marker according to the maximum gradient projection image of the marker after obtaining the maximum gradient projection image of the marker;
s300, obtaining a sub-pixel edge image of the marker according to the whole pixel edge image;
s400, positioning the center of the marker according to a plurality of sub-pixel edge images of the marker so as to obtain the center coordinates of the marker;
s500, calculating the deformation of the building according to the center coordinates of the markers at the current moment and the center coordinates of the markers at the previous moment.
In this embodiment, a plurality of building images including a marker are obtained by using a camera, then the marker images are extracted, the maximum gradient projection calculation is performed on the marker images to obtain the maximum gradient projection images of the marker, then the whole pixel edge images of the marker are lifted according to the maximum gradient projection images, then the sub-pixel edge images are extracted according to the whole pixel edge images, finally the central coordinates of the marker are obtained according to the sub-pixel edge images, and then the deformation of the building is calculated according to the central coordinates of the marker at the current moment and the central coordinates of the marker at the last moment, so that the deformation monitoring of the building is realized, the accuracy is high, and the timeliness is strong.
The embodiment of the invention does not need complicated camera calibration steps, can completely separate from expensive measuring instruments such as total stations and the like, and can realize deformation monitoring of long-distance imaging by only needing a single-lens reflex with a common length and a simple manual mark; the deformation monitoring of the photographing distance of 200 meters can be completed by using the single lens reflex with the optimal photographing distance of 100 meters, so that the implementation cost is low; the method avoids the expensive cost caused by lengthening the lens, is convenient to install, and avoids the extra monitoring error caused by camera shake when photographing; the deformation is acquired based on the image processing technology, after the image acquisition is completed, the subsequent image measurement and deformation calculation can be fully automatically performed without manual intervention, so that the labor cost is saved, the measurement precision is ensured, the three-dimensional precision requirements on a standard or manual can be met, and the timeliness is strong.
In a specific embodiment, before the step S100, first, a mounting marker is fabricated on a building, as shown in fig. 2, where the marker is a square 20cm x 20cm, and the white and black are alternate, and the center is a white solid circle with a diameter of 10cm, and the periphery is a black border with a width of 5 cm. If this mark is used for image measurement, no retroreflective properties are required. If the total station is needed to be combined, the white circle is replaced by a reflecting patch with the same size and a cross at the center. After the manufacturing of the marker is completed, the marker can be installed on the surface of a building needing deformation monitoring.
After the marker is installed, the camera is installed and the proportion is determined, and when the camera is installed, in order to reduce errors caused by shaking of the camera and a lens due to dead weight during photographing as much as possible, the invention preferably pours cement piers or makes a camera tripod fixing device to install the camera, and ensures the stability of the camera in the photographing process as much as possible. When the tripod fixing device is used for fixing, three holes are marked on the fixing bottom plate according to the maximum opening interval of the tripod feet, wherein the diameter of each hole is 1cm, and holes are drilled according to the hole depth of 1cm, so that the tripod can be stably fixed on the bottom plate.
After the camera fixing equipment is installed, the camera can be installed. Specifically, the single lens reflex is fixed on a tripod, and camera parameters are set to black and white, mute, fixed focus (maximum focus) and continuous shooting modes. And installing a wireless shutter release, and shooting 20-25 images of the same marker by using a camera under the wireless control of the shutter release. According to statistical theory and experimental data, the number of pictures taken at one time is not recommended to be too small, and the marker centers of 20-25 images need to be screened in the later processing period. If the number of pictures is small and the effective data is too small after screening, the reliability of the calculated mark center is insufficient. On the contrary, the number of pictures taken at one time is not recommended to be too much, and the camera shake can be caused by too many continuous shooting times, which is not beneficial to the positioning accuracy. Thus, in a preferred embodiment, the number of images taken at one time by the camera of the present invention is 20. The image shot by the camera can be transmitted to a designated server through a single-phase camera partner and matched software (wherein the single-phase camera partner software can be installed on a (personal) computer or an intermediate transmission device, the invention is not limited to the single-phase camera partner software), the SD card of the camera can be manually removed, and the image in the SD card can be copied into the computer.
In order to facilitate calculation of the subsequent deformation, it is necessary to determine the actual distance ratio=mm/pixel represented by each pixel of the camera at the shooting range of the installation position, this step is only needed at the time of the first shooting, and since the subsequent shooting is still at the same position, once the ratio is obtained, it is a constant known quantity. In the specific implementation, firstly, a scale is manufactured, as shown in fig. 3, the scale is a ruler with two ends fixed with black patches, the known length of the middle part of the black patches is L, the purpose of the scale is to calculate the actual distance represented by each pixel in an image shot by a camera under a certain shooting range, and the length L of the scale is preferably not less than 1 meter in order to ensure the accuracy of the parameter calculation; to ensure convenient attachment of the ruler to the building surface, the scale length L is preferably no greater than 2 meters. After the manufacturing is finished, the scale is fixed on the surface of the building in the horizontal or vertical direction, and the scale is close to the fixed position of the artificial mark. The actual distance determining step may then be started, and in particular the step of determining the actual distance represented by each pixel of the image captured by the camera comprises in particular:
(1) Reading and opening a first image, finding a scale, and amplifying the image as much as possible;
(2) If the ruler is fixed in the horizontal direction, measuring the pixel coordinate x at the rightmost position of the black paste at the left end of the scale by using a mouse Left side And leftmost pixel coordinate x of right-end black paste Right side ；
(3) If the ruler is fixed in the vertical direction, measuring the pixel coordinate y at the lowest position of the black paste at the upper end of the scale by using a mouse Upper part And the uppermost pixel coordinate y of the black paste at the lower end Lower part(s) ；
(4) Calculating the pixel length l of the scale in the image p ＝x Right side -x Left side +1 or l p ＝y Lower part(s) -y Upper part +1；
(5) Calculation ofThe actual distance represented by each pixel in the image under this range can be obtained.
It should be noted that: step (2) and step (3) need to manually measure and acquire the pixel length of the scale on the image, the calculated pixel length of different operators can not exceed the difference of 2 pixels, and the obtained ratio value is different. This difference can be compensated for by using a longer scale (> =1 meter). As can be obtained from experimental data, the length of the pixels in the image of the ruler with the length of 1 meter at the farthest distance of 200 meters is about 100 pixels, and the ratio values finally calculated by different operators are respectively as follows according to the maximum difference of 2 pixelsOr->By analogy, this difference is even smaller (< 0.2 mm/pixel) if a longer ruler or camera is used less than 200 meters. Taking the coordinate positioning precision of a certain point as an example, which is 0.3 pixel, the maximum difference brought by manually calculating the ratio value is only +.> Therefore, the requirement for the three-dimensional deformation monitoring accuracy is negligible.
After the installation of the camera and the determination of the proportion are completed, a plurality of building images containing the markers can be shot by the camera, then the images are preprocessed, the marker images are extracted, specifically, in the step S100, the preprocessing is performed on the building images, so as to extract the marker images specifically includes:
binarizing the building image to extract all suspected mark areas, wherein the suspected mark areas are areas combined by black and white;
and selecting a region of interest containing the marker from the suspected marker region according to the geometric characteristics of the marker, and obtaining a marker image according to the region of interest containing the marker.
In this embodiment, the image is binarized according to a combination mode of black and white circles of the markers, all the suspicious marker regions combined in black and white are extracted, then the regions meeting the conditions are selected from the suspicious marker regions according to the geometric features of the markers and the positions of the markers in the image, the regions are used as ROIs (Region on lnterest, interested regions) containing the markers, and the interested regions containing the markers are used as marker images.
Further, in step S200, an image processing technique is adopted to complete a high-precision measurement task, first, integral pixel positioning is required, the maximum gradient projection must be calculated to realize integral pixel positioning, and then an integral pixel edge image of the marker is obtained through the maximum gradient projection image. Since images acquired in a real scene are inevitably interfered by various uncertain factors, the conventional gradient calculation method cannot meet the requirements and needs to eliminate the interference of the uncertain factors. Specifically, the step S200 specifically includes:
obtaining the maximum gradient projection direction of each pixel point of the marker image, and respectively calculating the maximum gradient projection value of each pixel point of the marker image in the maximum gradient projection direction to obtain the maximum gradient projection image of the marker;
preprocessing the maximum gradient projection image of the marker to obtain an initial edge point of the marker;
and carrying out refinement treatment on the initial edge points, screening out smooth integral pixel edge points, and obtaining an integral pixel edge image of the marker.
In the present embodiment, it is assumed that the gradient projection of one gray-scale image f (x, y) in the horizontal and vertical directions is G x And G y Based on the error ellipse theory, the maximum gradient projection value of each pixel point of the marker image along the alpha direction can be calculated, and the formula for calculating the maximum gradient projection value of each pixel point of the marker image along the maximum gradient projection direction is specifically as follows:
wherein,
wherein (x, y) is the coordinate of a pixel point in the marker image, G x And G y And respectively obtaining gradient projection values of the marker image along the horizontal direction and the vertical direction, wherein alpha is the maximum gradient projection direction.
According to the formula, the maximum gradient projection image of the marker can be calculated, as shown in fig. 4, the method for calculating the maximum gradient projection image not only completely reserves the edge of the marker, but also effectively avoids other interferences, and the accurate whole pixel edge can be obtained by using the maximum gradient projection image.
Further, after the maximum gradient projection image is obtained, preprocessing is carried out on the maximum gradient projection image, specifically, binarizing processing is carried out on the maximum gradient projection image, then inversion is carried out on the maximum gradient projection image, the inner circle of the marker is obtained, and then the boundary tracking is carried out on the inner circle of the marker, so that the initial edge point of the marker is obtained.
Further, when the initial edge is obtained, it is assumed that tan (α) is a tangent value of a maximum gradient projection direction of a certain initial edge point, if:
searching the maximum value in the maximum gradient projection of the edge point and the left and right 2 neighborhood pixels;
tan (alpha) epsilon (- ≡2) or tan (alpha) epsilon (2, ≡), searching the maximum value in the maximum gradient projection of the edge point and the upper and lower 2 neighborhood pixels;
searching the maximum value in the maximum gradient projection of the edge point and 4 neighborhood pixels on the left, right, lower left and upper right of the edge point;
tan (alpha) ∈h,2], then find the maximum value in the maximum gradient projection of the edge point and its upper, lower left, upper right 4 neighborhood pixels;
tan (alpha) ∈ [ -2, -1], then find the maximum value in the maximum gradient projection of the edge point and its upper, lower, upper left, lower right 4 neighborhood pixels;
the maximum is found in the maximum gradient projection of the edge point and its 4 neighborhood pixels left, right, top left, bottom right.
According to the edge refinement processing method, after the initial edge point is replaced by the maximum point in the maximum gradient projection direction, a new edge point is generated, the integrity of the edge is higher, and the subsequent sub-pixel positioning accuracy is ensured. After the new edge points are obtained, the curvature of the edge points is calculated, the points with overlarge curvature are deleted, the smoothness of the outline of the outer edge of the marker is ensured, and then the whole pixel edge image of the marker is obtained, as shown in fig. 5.
In a preferred embodiment, the step S300 specifically includes:
calculating first-order differences in the row, column and two diagonal directions respectively in a preset neighborhood range of the center point by taking each edge point of the whole pixel edge image as the center point, and calculating sub-pixel positions of the corresponding edge point in the corresponding directions according to the first-order differences in the four directions;
and obtaining the sub-pixel edge points of the marker according to the sub-pixel positions of the edge points in the corresponding directions and the positions of the edge points in the whole pixel edge image, and obtaining the sub-pixel edge image of the marker according to the sub-pixel edge points.
In this embodiment, the first order difference between four directions including the row, the column, and the two diagonal directions is calculated in the 3*3 neighborhood, with each edge point as the center point. The first-order difference in the row and column directions is the average value of the front-back difference or the average value of the up-down difference, and the difference in the two diagonal directions is the difference in the two diagonal directions. The sub-pixel locations in these four directions are the mean of the first order differences in that direction. Then comparing the sub-pixel position of the edge with the original whole pixel position to obtain the position variation in the row, column and two opposite angles; the one with the largest variation in the four directions is taken as the new edge point sub-pixel position. In the case of diagonal orientation, the new subpixel positions need to be located after the variance is projected in the row and column directions, respectively, as shown in fig. 6, which shows both the full pixel edge and the subpixel edge. Representing the whole pixel edge, representing the new sub-pixel edge.
In a preferred embodiment, the step S400 specifically includes:
carrying out first ellipse fitting on all sub-pixel edge points in the sub-pixel edge image by adopting a preset ellipse fitting model so as to solve the ellipse fitting model;
fitting residual calculation is carried out on all sub-pixel edge points in the sub-pixel edge image according to the solved ellipse fitting model;
calculating residual standard deviation according to fitting residual errors of all sub-pixel edge points, screening all sub-pixel edge points according to the residual standard deviation, and performing ellipse fitting on the reserved sub-pixel edge points for a plurality of times to obtain final reserved sub-pixel edge points;
taking the center of the finally reserved sub-pixel edge points for elliptical fitting as the center of the sub-pixel edge image;
and obtaining the center coordinates of the markers according to the center coordinates of the plurality of sub-pixel edge images.
Specifically, ax is used first 2 +Bxy+Cy 2 +dx+ey+f=0 as an ellipse fitting model for all edge points { (x) i ，y i ) First ellipse fitting is performed on i=1, 2, …, n } to obtain six coefficients of a, B, C, D, E, F, arbitrary edge points (x i ，y i ) The corresponding fitting residual is V i ＝Ax i 2 +Bx i y i +Cy i 2 +Dx i +Ey i +F, the fitting residual error { V of all edge points can be calculated in turn i I=1, 2, …, n }; then calculate the residual standard deviation sigma V In 2σ V All edge points are screened for limit values as followsAnd carrying out secondary ellipse fitting on the reserved edge points, carrying out fitting residual calculation and edge point screening again, and cycling for a plurality of times until all the reserved edge points meet the limit requirement of residual errors, wherein the circle center of the ellipse is regarded as the mark center of the sub-pixel edge image, and finally the mark center of the positioned sub-pixel edge image is shown in figure 7.
In this embodiment, the concept of residual error in mathematical statistics is utilized to screen edge points after sub-pixel positioning, so that edges entering fitting calculation are guaranteed to be effective points meeting statistical limit values, and the accuracy and reliability of mark center positioning are further improved.
Further, after the center of each sub-pixel edge image is obtained, the center coordinates of the marker can be obtained according to the center coordinates of the plurality of sub-pixel edge images, specifically, the center coordinates of the marker are obtained according to the center coordinates of the plurality of sub-pixel edge images:
calculating the coordinate mean value and standard deviation of the center coordinates of the plurality of sub-pixel edge images;
calculating the difference value between the central coordinate of each sub-pixel edge image and the coordinate, namely the mean value, and screening the central coordinate of each sub-pixel edge image according to the difference value;
and (3) screening the reserved central coordinates for a plurality of times, and taking the coordinate mean value of a plurality of central coordinates obtained after the screening for a plurality of times as the central coordinates of the marker.
Specifically, the camera takes 20 images of each marker, and each marker can obtain 20 central coordinates { (x) oi ，y oi ) I=1, 2,..20 }, and then calculating the coordinate mean valueAnd standard deviation->Then calculate the difference (Deltax) between each center point coordinate and the coordinate mean oi ，Δy oi ) To->And->For limit, 20 marker centers were screened: />And then repeatedly screening the reserved M (M is less than or equal to 20) mark center coordinates, wherein the repeated screening process is the same as that of the screening process, namely, the coordinate mean value and standard deviation of the reserved center coordinates are calculated again, then the difference value between the reserved center coordinates and the coordinate mean value of the reserved center coordinates is calculated again, the screening is carried out according to the difference value until all mark center coordinates meet the limit value requirement, the coordinate mean value at the moment is the sub-pixel center corresponding to the mark, and the standard deviation of the coordinates is the positioning precision.
In this embodiment, various factors such as illumination, weather, shielding, shake of a camera during continuous photographing, and minimum false identification of a mark directly affect the positioning of the center of the mark, so although continuous photographing is performed in a short time to obtain the same batch of mark images, the positioning of the center of the mark for each image is not completely the same, and the situation that the center coordinates are abnormal or have large difference with the center positioning of other images is most likely to occur, if the unreliable or even erroneous center positioning participates in the mean value calculation of the center coordinates of the mark, the final center positioning of the mark must not meet the precision requirement, and must be deleted. Therefore, the embodiment of the invention screens a group of circle centers acquired by the same batch of mark images again according to the statistical criterion, deletes the abnormal circle center coordinates or the circle centers with larger difference with other circle center coordinates, and ensures that the mark center positioning meets the precision requirement. The limit value of the coordinate difference is formulated through the statistical criterion, so that the center positioning meeting the precision requirement can be automatically screened without the reason of causing the abnormal center coordinates, and the invention can be suitable for different outdoor environments and can still meet the measurement precision requirement in response to sudden scenes.
In a preferred embodiment, in the step S500, the formula for calculating the deformation amount of the building is specifically:
Δx＝x 2 -x 1 ，
Δy＝y 2 -y 1 ，
wherein, (x) 1 ，y 1 ) At t 1 Obtaining the center coordinates of the markers at the moment, (x) 2 ，y 2 ) At t 2 The center coordinates of the markers obtained at the moment, Δx is the horizontal deformation amount, Δy is the vertical deformation amount, and Δd is the deformed point distance.
The invention performs a plurality of experiments aiming at various factors which can influence the image measurement precision, such as weather, illumination, camera setting, environment, foot rest height, shaking and the like, and the embodiment is described by two experiments.
In experiment 1, 5 circles with the diameter of 10cm are stuck on a black matrix flat plate, the distance between the centers of the circles is known, and the total station and the single lens reflex matched with the invention simultaneously carry out center positioning and center distance measurement on 5 marks at 200 meters, 150 meters, 100 meters and 50 meters, and the result shows that the positioning accuracy of the mark center of the invention is high, and the measurement accuracy of the mark center distance is equivalent to that of the total station, so that the total station can be completely replaced for deformation monitoring.
In experiment 2, the actual distance represented by one pixel in the image was calculated using a magnetic level of known length, and the total station was not used. Fixing a camera on a foot rest with the lowest height under the same shooting range (50 meters, 100 meters, 150 meters, 200 meters), and firstly collecting a batch of images; then the camera is manually rocked, and a batch of images are collected after the camera is stable. Therefore, the two batches of pictures are arranged in the same shooting range, and the result shows that the invention can accurately measure the image shot in the field within the 200 m shooting range, the measurement result is stable and reliable, and the measurement precision meets the requirements of three precision of 'water transportation engineering measurement Specification' and 'building deformation measurement Specification'.
As shown in fig. 9, based on the building deformation monitoring method, the present invention further provides a building deformation monitoring device, where the building deformation monitoring device may be a computing device such as a mobile terminal, a desktop computer, a notebook computer, a palm computer, and a server. The building deformation monitoring device includes a processor 10, a memory 20, and a display 30. Fig. 9 shows only a portion of the components of the building deformation monitoring apparatus, but it should be understood that not all of the illustrated components are required to be implemented, and more or fewer components may alternatively be implemented.
The memory 20 may in some embodiments be an internal storage unit of the building deformation monitoring device, such as a hard disk or a memory of the building deformation monitoring device. The memory 20 may in other embodiments also be an external storage device of the building deformation monitoring device, such as a plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card) or the like, which is provided on the building deformation monitoring device. Further, the memory 20 may also include both internal and external memory units of the building deformation monitoring device. The memory 20 is used for storing application software and various data installed on the building deformation monitoring device, such as program codes for installing the building deformation monitoring device. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 20 stores a building deformation monitoring program 40, and the building deformation monitoring program 40 is executable by the processor 10 to implement the building deformation monitoring method according to the embodiments of the present application.
The processor 10 may in some embodiments be a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip for executing program code or processing data stored in the memory 20, for example for performing the building deformation monitoring method or the like.
The display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. The display 30 is used for displaying information at the building deformation monitoring device and for displaying a visual user interface. The components 10-30 of the building deformation monitoring device communicate with each other via a system bus.
In one embodiment, the steps in the building deformation monitoring method described above are implemented when the processor 10 executes the building deformation monitoring program 40 in the memory 20.
In summary, the method, the device and the storage medium for monitoring the deformation of the building provided by the invention are characterized in that firstly, a plurality of building images containing the markers are obtained by using a camera, then, after the marker images are extracted, the maximum gradient projection calculation is carried out on the marker images to obtain the maximum gradient projection images of the markers, then, the whole pixel edge images of the markers are lifted according to the maximum gradient projection images, then, the sub-pixel edge images are extracted according to the whole pixel edge images, finally, after the center coordinates of the markers are obtained according to the plurality of sub-pixel edge images, the deformation of the building is calculated according to the center coordinates of the markers at the current moment and the center coordinates of the markers at the last moment, so that the deformation monitoring of the building is realized, the accuracy is high, the timeliness is high, the hardware cost and the labor cost are low, the distance is taken, and the environmental adaptability is strong.
Of course, those skilled in the art will appreciate that implementing all or part of the above-described methods may be implemented by a computer program for instructing relevant hardware (e.g., a processor, a controller, etc.), where the program may be stored in a computer-readable storage medium, and where the program may include the steps of the above-described method embodiments when executed. The storage medium may be a memory, a magnetic disk, an optical disk, or the like.
The above-described embodiments of the present invention do not limit the scope of the present invention. Any other corresponding changes and modifications made in accordance with the technical idea of the present invention shall be included in the scope of the claims of the present invention.
Claims (7)
1. A method for monitoring deformation of a building, comprising the steps of:
acquiring a plurality of building images containing markers, which are shot by a camera at the current moment, and preprocessing the building images to extract the marker images, wherein the markers are fixedly arranged on a building;
carrying out maximum gradient projection calculation on the marker image, and obtaining a whole pixel edge image of the marker according to the maximum gradient projection image of the marker after obtaining the maximum gradient projection image of the marker;
obtaining a sub-pixel edge image of the marker according to the whole pixel edge image;
positioning the center of the marker according to the sub-pixel edge images of the markers to obtain the center coordinates of the marker;
calculating the deformation of the building according to the center coordinates of the markers at the current moment and the center coordinates of the markers at the previous moment;
the calculating the maximum gradient projection of the marker image, obtaining the maximum gradient projection image of the marker, and obtaining the whole pixel edge image of the marker according to the maximum gradient projection image of the marker specifically comprises:
obtaining the maximum gradient projection direction of each pixel point of the marker image, and respectively calculating the maximum gradient projection value of each pixel point of the marker image in the maximum gradient projection direction to obtain the maximum gradient projection image of the marker;
preprocessing the maximum gradient projection image of the marker to obtain an initial edge point of the marker;
refining the initial edge points, screening out smooth integral pixel edge points, and obtaining integral pixel edge images of the markers;
the formula for calculating the maximum gradient projection value of each pixel point of the marker image in the maximum gradient projection direction is specifically as follows:
，
wherein,
，
，
wherein (x, y) is the coordinate of a pixel point in the marker image,and->Gradient projection values of the marker image along the horizontal direction and the vertical direction are respectively +.>Is the maximum gradient projection direction;
the obtaining the sub-pixel edge image of the marker according to the whole pixel edge image specifically includes:
calculating first-order differences in the row, column and two diagonal directions respectively in a preset neighborhood range of the center point by taking each edge point of the whole pixel edge image as the center point, and calculating sub-pixel positions of the corresponding edge point in the corresponding directions according to the first-order differences in the four directions;
and obtaining the sub-pixel edge point of the marker according to the calculated sub-pixel position of the edge point and the position of the edge point in the whole pixel edge image, and obtaining the sub-pixel edge image of the marker according to the sub-pixel edge point.
2. The method of claim 1, wherein preprocessing the building image to extract a marker image specifically comprises:
binarizing the building image to extract all suspected mark areas, wherein the suspected mark areas are areas combined by black and white;
and selecting a region of interest containing the marker from the suspected marker region according to the geometric characteristics of the marker, and obtaining a marker image according to the region of interest containing the marker.
3. The building deformation monitoring method according to claim 1, wherein the positioning the center of the marker according to the sub-pixel edge images of the plurality of markers to obtain the center coordinates of the marker specifically includes:
carrying out first ellipse fitting on all sub-pixel edge points in the sub-pixel edge image by adopting a preset ellipse fitting model so as to solve the ellipse fitting model;
fitting residual calculation is carried out on all sub-pixel edge points in the sub-pixel edge image according to the solved ellipse fitting model;
calculating residual standard deviation according to fitting residual errors of all sub-pixel edge points, screening all sub-pixel edge points according to the residual standard deviation, and performing ellipse fitting on the reserved sub-pixel edge points for a plurality of times to obtain final reserved sub-pixel edge points;
taking the center of the finally reserved sub-pixel edge points for elliptical fitting as the center of the sub-pixel edge image;
and obtaining the center coordinates of the markers according to the center coordinates of the plurality of sub-pixel edge images.
4. A building deformation monitoring method according to claim 3, wherein the center coordinates of the markers obtained according to the center coordinates of the plurality of sub-pixel edge images are specifically:
calculating the coordinate mean value and standard deviation of the center coordinates of the plurality of sub-pixel edge images;
calculating the difference value between the central coordinate of each sub-pixel edge image and the coordinate mean value, and screening the central coordinate of each sub-pixel edge image according to the difference value;
and (3) screening the reserved central coordinates for a plurality of times, and taking the coordinate mean value of a plurality of central coordinates obtained after the screening for a plurality of times as the central coordinates of the marker.
5. The building deformation monitoring method according to claim 1, wherein the formula for calculating the deformation amount of the building is specifically:
，
，
，
wherein,is->Obtaining the central coordinates of the markers at the moment, +.>Is->Center coordinates of the markers obtained at the moment, +.>Is horizontal deformation amount->Is vertical deformation amount->Is the deformed spot distance.
6. A building deformation monitoring apparatus, comprising: a processor and a memory;
the memory has stored thereon a computer readable program executable by the processor;
the processor, when executing the computer readable program, implements the steps of the building deformation monitoring method according to any one of claims 1-5.
7. A computer readable storage medium storing one or more programs executable by one or more processors to implement the steps in the building deformation monitoring method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110826876.1A CN113610782B (en) | 2021-07-21 | 2021-07-21 | Building deformation monitoring method, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110826876.1A CN113610782B (en) | 2021-07-21 | 2021-07-21 | Building deformation monitoring method, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113610782A CN113610782A (en) | 2021-11-05 |
| CN113610782B true CN113610782B (en) | 2024-01-02 |
Family
ID=78305075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110826876.1A Active CN113610782B (en) | 2021-07-21 | 2021-07-21 | Building deformation monitoring method, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113610782B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117606362A (en) * | 2023-11-23 | 2024-02-27 | 湖南科天健光电技术有限公司 | Detection method and detection system for slope displacement |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551046A (en) * | 2015-12-17 | 2016-05-04 | 浙江宇视科技有限公司 | Vehicle face location method and device |
| CN106441138A (en) * | 2016-10-12 | 2017-02-22 | 中南大学 | Deformation monitoring method based on vision measurement |
| CN110823116A (en) * | 2019-10-25 | 2020-02-21 | 同济大学 | Image-based building component deformation measurement method |
| CN112381847A (en) * | 2020-10-27 | 2021-02-19 | 新拓三维技术(深圳)有限公司 | Pipeline end head space pose measuring method and system |
| CN112419287A (en) * | 2020-11-27 | 2021-02-26 | 杭州鲁尔物联科技有限公司 | Building deflection determination method and device and electronic equipment |
| CN113077467A (en) * | 2021-06-08 | 2021-07-06 | 深圳市华汉伟业科技有限公司 | Edge defect detection method and device for target object and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279372B (en) * | 2015-09-29 | 2018-12-14 | 百度在线网络技术（北京）有限公司 | A kind of method and apparatus of determining depth of building |
-
2021
- 2021-07-21 CN CN202110826876.1A patent/CN113610782B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551046A (en) * | 2015-12-17 | 2016-05-04 | 浙江宇视科技有限公司 | Vehicle face location method and device |
| CN106441138A (en) * | 2016-10-12 | 2017-02-22 | 中南大学 | Deformation monitoring method based on vision measurement |
| CN110823116A (en) * | 2019-10-25 | 2020-02-21 | 同济大学 | Image-based building component deformation measurement method |
| CN112381847A (en) * | 2020-10-27 | 2021-02-19 | 新拓三维技术(深圳)有限公司 | Pipeline end head space pose measuring method and system |
| CN112419287A (en) * | 2020-11-27 | 2021-02-26 | 杭州鲁尔物联科技有限公司 | Building deflection determination method and device and electronic equipment |
| CN113077467A (en) * | 2021-06-08 | 2021-07-06 | 深圳市华汉伟业科技有限公司 | Edge defect detection method and device for target object and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 减小长度投影变形的一种地图投影新方法;施一民 等;《同济大学学报（自然科学版）》;第35卷(第3期);第418-421页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113610782A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106441138B (en) | The deformation monitoring method of view-based access control model measurement | |
| CN107917695B (en) | House inclination monitoring method based on image recognition technology | |
| US10089530B2 (en) | Systems and methods for autonomous perpendicular imaging of test squares | |
| CN102768762B (en) | Digital camera calibration method targeted to shield tunnel defect digital radiography detection and device thereof | |
| CN104657711B (en) | A kind of readings of pointer type meters automatic identifying method of robust | |
| WO2018204552A1 (en) | Gps offset calibration for uavs | |
| JP6516558B2 (en) | Position information processing method | |
| CN110619662A (en) | Monocular vision-based multi-pedestrian target space continuous positioning method and system | |
| CN112598755B (en) | Intelligent face analysis method based on drill jumbo | |
| WO2021212477A1 (en) | Point cloud data correction method, and related device | |
| CN1820282A (en) | Image processing device | |
| CN106600561B (en) | Aerial image perspective distortion automatic correction method based on projection mapping | |
| CN113610782B (en) | Building deformation monitoring method, equipment and storage medium | |
| CN111442845A (en) | Infrared temperature measurement method and device based on distance compensation and computer storage medium | |
| CN115578315A (en) | Bridge strain close-range photogrammetry method based on unmanned aerial vehicle image | |
| CN114005108A (en) | Pointer instrument degree identification method based on coordinate transformation | |
| WO2022126339A1 (en) | Method for monitoring deformation of civil structure, and related device | |
| JP3597832B2 (en) | Trajectory error measurement method and trajectory error measurement system used for the method | |
| JP6803940B2 (en) | Remote meter reading computer, its method and program | |
| CN113240635B (en) | Structural object detection image quality testing method with crack resolution as reference | |
| CN114782555A (en) | Map mapping method, apparatus, and storage medium | |
| Frangione et al. | Multi-step approach for automated scaling of photogrammetric micro-measurements | |
| WO2017107564A1 (en) | Board image acquisition method and system | |
| CN109636840B (en) | Method for detecting building shadow based on ghost image | |
| CN112964192A (en) | Engineering measurement online calibration method and system based on image video |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |