CN111964604B - Plane deformation monitoring and measuring method based on image recognition - Google Patents
Plane deformation monitoring and measuring method based on image recognition Download PDFInfo
- Publication number
- CN111964604B CN111964604B CN202011121052.6A CN202011121052A CN111964604B CN 111964604 B CN111964604 B CN 111964604B CN 202011121052 A CN202011121052 A CN 202011121052A CN 111964604 B CN111964604 B CN 111964604B
- Authority
- CN
- China
- Prior art keywords
- target monitoring
- point
- target
- image
- monitoring points
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/16—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
本发明涉及一种基于影像识别的平面变形监测测量方法，包括通过摄像头进行图像的采集，获得多张图像，然后上传；对上传的图像进行处理、计算；计算单张图像上目标监测点之间的实际相对位置，并计算一测回的水平闭合差和竖向闭合差；采用高程传递和位移传递，结合闭合差进行平差，计算目标监测点之间的实际相对位置并存储；循环前述步骤一次；将前后两次的计算结果进行对比，获取各目标监测点在监测平面内相对于起始目标监测点的实际变形量。本发明使用固定位置的可旋转变焦摄像头对目标变形进行拍摄，采用算法计算各监测点在监测平面内的变形，可实现一观测点对多目标的变形测量，从而可以实现低成本的无接触建构筑物的变形监测。
The invention relates to a plane deformation monitoring and measurement method based on image recognition, which includes collecting images through a camera, obtaining multiple images, and uploading them; processing and calculating the uploaded images; calculating the distance between target monitoring points on a single image The actual relative position of the target monitoring points is calculated, and the horizontal and vertical closure differences of a measurement round are calculated; the elevation transfer and displacement transfer are used, combined with the closure difference for adjustment, and the actual relative positions between the target monitoring points are calculated and stored; the above steps are repeated. Once; compare the calculation results before and after to obtain the actual deformation of each target monitoring point relative to the initial target monitoring point in the monitoring plane. The invention uses a fixed-position rotatable zoom camera to shoot the deformation of the target, and uses an algorithm to calculate the deformation of each monitoring point in the monitoring plane, so that the deformation measurement of one observation point to multiple targets can be realized, thereby realizing a low-cost non-contact construction. Deformation monitoring of structures.
Description
Technical Field
The invention relates to the technical field of engineering plane deformation monitoring and measurement, in particular to a plane deformation monitoring and measuring method based on image recognition.
Background
The monitoring of engineering deformation such as municipal administration, bridges, water conservancy, civil engineering and the like is an important content of structural health monitoring and an important index for evaluating structural stability. In conventional measurement methods represented by levels, total stations and the like, the monitoring means have large workload and are greatly influenced by the operation mode of the instrument. The new monitoring means represented by a measuring robot, a three-dimensional laser scanning technology, a static leveling measurement and the like are influenced by factors such as fields, installation difficulty, price and the like, and are difficult to popularize.
The visual deformation monitoring and measuring technology integrates photogrammetry, image processing and computer technology, a target image is obtained by non-measuring digital camera equipment, the image is processed by a computer, and the change of the target image on an image sequence is compared, so that the two-dimensional displacement deformation monitoring technology of displacement is calculated.
However, the existing image processing method adopts a fixed-point orientation mode, one device observes one point at a fixed visual angle, the efficiency is low, and for monitoring a measurement scene in a large area and a long distance, a plurality of devices are needed, so that the installation workload is increased while the method is not economical.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the plane deformation monitoring and measuring method based on image recognition is provided, and non-contact measurement of deformation of a target object is realized by adopting a graph splicing and displacement data transmission method.
The technical scheme adopted by the invention for solving the technical problems is as follows: a plane deformation monitoring and measuring method based on image recognition comprises the following steps,
1) acquiring images of all target monitoring points through a camera to obtain a plurality of images, wherein each image at least comprises two target monitoring points, and then transmitting the images to an image processing platform of a server;
2) processing and calculating the uploaded image to obtain the image relative position between the first target monitoring point and the last target monitoring point on a single image;
3) calculating the actual relative position between the two target monitoring points on the single image according to the relative position of the image between the two target monitoring points in the step 2), and calculating the horizontal closure difference and the vertical closure difference measured back according to the actual relative position between the two target monitoring points;
4) adopting elevation transmission and displacement transmission, carrying out adjustment by combining the horizontal closure difference and the vertical closure difference in the step 3), calculating and storing the actual relative positions of the rest target monitoring points relative to the initial target monitoring point;
5) cycling steps 1) -4) once;
6) and comparing the currently obtained calculation results of the actual relative positions of the other target monitoring points relative to the initial target monitoring point with the previously obtained calculation results of the actual relative positions of the other target monitoring points relative to the initial target monitoring point to obtain the actual deformation of each target monitoring point relative to the initial target monitoring point in the monitoring plane.
Further, in step 1), the camera is a rotary camera lens capable of adjusting focal length parameters; the target monitoring points are multiple, and all the target monitoring points are located in the same plane.
Still further, the target monitoring point is provided with a circular target with the diameter of M, and M is 10-15 cm.
Furthermore, in the step 1), n target monitoring points are provided, wherein n is a natural number greater than 1, the length of a measuring section shot each time is set as two adjacent target monitoring points during image acquisition, and the focal length of the camera is adjusted once every time a measuring section is shot; the images are shot between two points, and n target monitoring points shoot 2 (n-1) images in a reciprocating mode.
Still further, in step 2) of the present invention, the calculation method of the relative position of the image is as follows:
respectively identifying targets of two adjacent target monitoring points on the image, calculating the central point coordinates of the two targets and the pixel length of the horizontal and vertical axes of the two targets, wherein the central point coordinate of the point i target on the ith image is (x)2i-1，y2i-1) The coordinate of the central point of the (i +1) point target is (x)2i，y2i) (ii) a The pixel length of the horizontal and vertical axes of the i point target is (a)2i-1，b2i-1) The pixel length of the horizontal and vertical axes of the (i +1) point target is (a)2i，b2i）。
Still further, in step 3), the method for calculating the actual relative position between the two target monitoring points on the single image according to the relative position between the two target monitoring points in step 2) is as follows:
calculating the pixel difference value between the two adjacent target monitoring points according to the central point coordinates of the targets of the two adjacent target monitoring points
(ii) a Then the actual relative position coordinates of the (i +1) point are,
wherein
、
Respectively as average conversion ratios in the horizontal and vertical axis directions of the intermediate image,
still further, in step 3) of the present invention, a horizontal closure difference is measured
And vertical closure difference
Comprises the following steps:
still further, in step 4) of the present invention, the closure difference is determined according to the horizontal
And vertical closure difference
Calculating the horizontal correction number of each measurement section
And vertical correction number
Respectively as follows:
the actual relative position coordinate of the (i +1) point after the adjustment correction is
，
Then, using elevation and displacement transfer, the actual relative position of the (i +1) point with respect to the initial target monitoring point is
。
The method provided by the invention has the beneficial effects that the defects in the background technology are overcome, the rotatable zoom camera with a fixed position is used for shooting target monitoring points in sequence, the deformation of each monitoring point is calculated by utilizing the transmission of elevation and displacement and adopting a special graphic algorithm, the deformation measurement of a plurality of targets at one observation point can be realized, and the deformation monitoring of a low-cost non-contact building can be realized.
Drawings
FIG. 1 is a block diagram of the hardware connections of the measurement system of the present invention.
Fig. 2 is a schematic view of the state of the present invention at the time of initial measurement.
FIG. 3 is a vector plot of the distance between two points.
Fig. 4-5 are schematic diagrams of the state of the measurement process of the present invention.
Detailed Description
The invention will now be described in further detail with reference to the drawings and preferred embodiments. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution related to the present invention.
As shown in fig. 1, the system for monitoring and measuring planar deformation based on image recognition is divided into three-level structures, namely, an acquisition end, a server end and a client end. The acquisition end mainly comprises a fixed rotatable camera, a network transmission module, a power management module and an image storage module. The camera is used for collecting images to obtain images of a series of target monitoring points, and the images are transmitted to an image processing platform of the server through the network equipment module. The server side comprises an image processing platform, a data storage platform and an online display platform. The image processing platform calculates and analyzes relative position data of each monitoring point by using a special image processing algorithm and software, and further realizes the monitoring of the deformation of each point; the data storage stores the calculated deformation data into a database; the data display platform displays the deformation data in a high-readability mode through a webpage form.
The fixed rotatable camera used by the deformation monitoring system is customized hardware, debugging can be carried out according to the distribution of site reference points and monitoring points when the monitoring system is installed, and focal length parameters and the rotation angle of the camera each time are preset in matched driving software. In addition, anti-interference and fault-tolerant settings are respectively carried out on software and hardware, original image data are stored in a data storage module according to time and a shooting sequence, and manual remote acquisition and rechecking can be carried out when network abnormality and data abnormality occur.
And the acquired original image is subjected to fidelity transmission through a network transmission module and uploaded to a server. And the server side is configured with a special image processing algorithm and software, processes and calculates the image uploaded by the acquisition equipment, and finally obtains the relative position information of each monitoring point. And comparing the calculation result with the last time result to obtain the relative deformation data of each monitoring point in the horizontal and vertical directions in the monitoring plane. Particularly, when 2 or more cameras are adopted to simultaneously acquire images of the monitoring points at two different angles, the deformation of the monitoring points in the plane normal direction of the monitoring points can be analyzed according to the installation angle and the image parameters of the same monitoring point.
After the deformed data are analyzed, the server stores the data into the database, the data are visually displayed by matching with a webpage platform and a front end on the server, and a user can obtain the data by accessing the address of the display platform.
During measurement, a high-resolution camera lens capable of adjusting focal length parameters and a rotation angle, and a matched picture processing algorithm and software are preset.
For calculation convenience, the measuring method is characterized in that a circular target with the diameter M of 10cm is arranged on a target monitoring point.
The specific measurement method comprises the following steps:
as shown in fig. 1, the number of target monitoring points in this embodiment is 4, and all the target monitoring points are located in the same plane;
1) when the measurement is started, the camera is started first, and the focal length of the camera is adjusted to be f1The first shot is taken as in fig. 2. The method comprises the steps of collecting images of 4 target monitoring points through a camera, setting the length of a measuring section shot at each time as two adjacent target monitoring points, if the length from a point 1 to a point 2 is a measuring section, adjusting the rotation angle of a lens after the shooting of the measuring section is finished, and adjusting the focal length of the camera to be f2Carrying out second shooting, and so on, wherein each shooting is carried out in sequence, the camera rotates once, and the focal length of the camera is adjusted once; shooting between two points, and shooting by 4 target monitoring points back and forth
The images, namely the 1 st image is the point No. 1 and the point No. 2, the 2 nd image is the point No. 2 and the point No. 3, and the 3 rd image is the point No. 3 and the point No. 4; then the camera rotates reversely according to the original angle to shoot, namely the 4 th image is the No. 3 point and the No. 4 point, the 5 th image is the No. 2 point and the No. 3 point, and the 6 th image is the No. 1 point and the No. 2 point; as in fig. 4 and 5, thisOne round trip shooting process is a survey; then 6 images are transmitted to an image processing platform of a server; according to another embodiment, the images may be uploaded every time one image is captured, and processing methods such as calculation and conversion of the uploaded images may be the same as those of the present embodiment.
2) Processing and calculating the uploaded image to obtain the image relative position between the point i and the point (i +1) on the ith image;
respectively identifying targets of two adjacent target monitoring points on the image, and calculating the coordinates of the central points of the two targets and the pixel lengths of the horizontal and vertical axes of the two targets, as shown in fig. 3; on the ith image, the coordinate of the center point of the point i target is (x)2i-1，y2i-1) The coordinate of the central point of the (i +1) point target is (x)2i，y2i) (ii) a Since the image may be distorted when photographed, although the target is circular, the finally presented figure is not necessarily circular, and therefore, the pixel length of the horizontal and vertical axes (horizontal and vertical directions) of the i-point target is recorded as (a)2i-1，b2i-1) The pixel length of the horizontal and vertical axes of the (i +1) point target is (a)2i，b2i）。
3) Calculating the actual relative position between the two target monitoring points on a single image according to the relative position of the image between the two target monitoring points, and calculating a measured horizontal closure difference and a measured vertical closure difference according to the actual relative position between the two target monitoring points;
calculating the pixel difference value between the two adjacent target monitoring points according to the central point coordinates of the targets of the two adjacent target monitoring points as follows:
(ii) a Since the visual angle of the camera is not necessarily perpendicular to the plane of the monitoring point, angular distortion may exist, and therefore, when the actual relative position is converted, the conversion ratio of the intermediate image is estimated by taking the average conversion ratio of the horizontal axis and the vertical axis near the two target points so as to eliminate the influence of the angular distortion.
Then the actual relative position coordinates of the (i +1) point are,
wherein
、
Respectively as average conversion ratios in the horizontal and vertical axis directions of the intermediate image,
horizontal closure error of one measurement
And vertical closure difference
Comprises the following steps:
4) adopting elevation transmission and displacement transmission, carrying out adjustment by combining the horizontal closure difference and the vertical closure difference in the step 3), calculating and storing the actual relative positions of the rest target monitoring points relative to the initial target monitoring point;
according to horizontal closure difference
And verticalDifference of direction of closure
Calculating the horizontal correction number of each measurement section
And vertical correction number
Respectively as follows:
the actual relative position coordinate of the (i +1) point after the adjustment correction is
，
Then, using elevation and displacement transfer, the actual relative position of the (i +1) point with respect to the initial target monitoring point is
。
5) Cycling steps 1) -4) once, and starting the second measurement;
6) and comparing the calculation result of the second measurement with the calculation result of the second measurement, and calculating the difference between the relative positions of the monitoring points of the two adjacent measurements to obtain the relative displacement (horizontal displacement and vertical displacement) of the monitoring points of the two measurements in the plane, thereby monitoring the deformation of the building in the monitoring plane.
In the present embodiment, point No. 1 is set as a reference point with little or no deformation, and according to another embodiment, a reference point without deformation may be provided at a non-deformed position outside the monitoring plane of the target monitoring point. The reference point and the initial measurement point of the target monitor may be simultaneously present on the first image.
The plane deformation monitoring and measuring method based on image recognition can realize low-cost non-contact plane deformation monitoring. After the primary installation, manual intervention is not needed, and the defect that the deformation monitoring is carried out by the conventional geodetic surveying method is effectively overcome.
The invention relates to a lens rotation focusing design, which can realize the measurement of horizontal displacement and vertical displacement of a plurality of monitoring points by a single lens through the transmission of elevation and displacement of image marking points, and reduce the cost of hardware arrangement.
While particular embodiments of the present invention have been described in the foregoing specification, various modifications and alterations to the previously described embodiments will become apparent to those skilled in the art from this description without departing from the spirit and scope of the invention.
Claims (5)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011121052.6A CN111964604B (en) | 2020-10-20 | 2020-10-20 | Plane deformation monitoring and measuring method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011121052.6A CN111964604B (en) | 2020-10-20 | 2020-10-20 | Plane deformation monitoring and measuring method based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111964604A CN111964604A (en) | 2020-11-20 |
| CN111964604B true CN111964604B (en) | 2021-01-29 |
Family
ID=73387656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011121052.6A Active CN111964604B (en) | 2020-10-20 | 2020-10-20 | Plane deformation monitoring and measuring method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111964604B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112504137B (en) * | 2020-12-07 | 2022-07-26 | 北京智博联科技股份有限公司 | Multi-target digital image detection method based on cloud computing |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5899703A (en) * | 1981-12-09 | 1983-06-14 | Nippon Sheet Glass Co Ltd | Method for measuring degree of deformation of transparent plate |
| CN101566465A (en) * | 2009-05-18 | 2009-10-28 | 西安交通大学 | Method for measuring object deformation in real time |
| CN103927784A (en) * | 2014-04-17 | 2014-07-16 | 中国科学院深圳先进技术研究院 | Three-dimensional scanning method |
| CN108759699A (en) * | 2018-03-27 | 2018-11-06 | 西安交通大学 | A kind of measurement method and system of big visual field masonry structure material three-dimensional whole field deformation |
| CN108931230A (en) * | 2018-06-04 | 2018-12-04 | 广州建设工程质量安全检测中心有限公司 | A kind of sleeve configuration tunnel deformation monitoring method |
| CN110645901A (en) * | 2019-10-11 | 2020-01-03 | 靳鸣 | Building engineering dynamic monitoring system and using method |
-
2020
- 2020-10-20 CN CN202011121052.6A patent/CN111964604B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5899703A (en) * | 1981-12-09 | 1983-06-14 | Nippon Sheet Glass Co Ltd | Method for measuring degree of deformation of transparent plate |
| CN101566465A (en) * | 2009-05-18 | 2009-10-28 | 西安交通大学 | Method for measuring object deformation in real time |
| CN103927784A (en) * | 2014-04-17 | 2014-07-16 | 中国科学院深圳先进技术研究院 | Three-dimensional scanning method |
| CN108759699A (en) * | 2018-03-27 | 2018-11-06 | 西安交通大学 | A kind of measurement method and system of big visual field masonry structure material three-dimensional whole field deformation |
| CN108931230A (en) * | 2018-06-04 | 2018-12-04 | 广州建设工程质量安全检测中心有限公司 | A kind of sleeve configuration tunnel deformation monitoring method |
| CN110645901A (en) * | 2019-10-11 | 2020-01-03 | 靳鸣 | Building engineering dynamic monitoring system and using method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111964604A (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112254663B (en) | A method and system for monitoring and measuring plane deformation based on image recognition | |
| CN103826103B (en) | Cruise control method for tripod head video camera | |
| CN109118545A (en) | 3-D imaging system scaling method and system based on rotary shaft and binocular camera | |
| EP2847741B1 (en) | Camera scene fitting of real world scenes for camera pose determination | |
| CN104395692B (en) | Three-dimensional measurement method, device, system and image processing apparatus | |
| CN110345921B (en) | Stereo visual field vision measurement and vertical axis aberration and axial aberration correction method and system | |
| CN108288291A (en) | Polyphaser calibration based on single-point calibration object | |
| CN110243283A (en) | A variable boresight visual measurement system and method | |
| CN109632103A (en) | High vacant building Temperature Distribution and surface crack remote supervision system and monitoring method | |
| CN106840011A (en) | Steel tower deformation measuring device and its method | |
| CN101900552B (en) | Longitude-latitude camera videogrammetric method and system | |
| CN107101623A (en) | Measuring method, system and device based on coordinate transformation | |
| CN111288891B (en) | Non-contact three-dimensional measurement positioning system, method and storage medium | |
| WO2020063058A1 (en) | Calibration method for multi-degree-of-freedom movable vision system | |
| CN104089628A (en) | Self-adaption geometric calibration method of light field camera | |
| CN111076698A (en) | Observation evaluation device and calculation method for calibrating installation deviation of photoelectric product | |
| CN106709912A (en) | Automatic testing system and testing method for preset position precision | |
| CN112419425A (en) | Anti-disturbance high-precision camera group measuring method for structural deformation measurement | |
| CN115937326B (en) | A fast on-site calibration method for large-field-of-view cameras suitable for complex working conditions | |
| CN108955642B (en) | A seamless stitching method for large-format equivalent central projection images | |
| CN112907647B (en) | A three-dimensional spatial dimension measurement method based on a fixed monocular camera | |
| CN111964604B (en) | Plane deformation monitoring and measuring method based on image recognition | |
| CN115810041A (en) | Cable size detection method and system based on multi-camera vision | |
| CN115143887A (en) | Method for correcting measurement result of visual monitoring equipment and visual monitoring system | |
| CN114440769B (en) | Multi-measuring-point three-dimensional displacement measuring method and measuring system |
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 | ||
| CP03 | Change of name, title or address |
Address after: 213000 No.10, muchu Road, Zhonglou District, Changzhou City, Jiangsu Province Patentee after: Joint Inspection (Jiangsu) Technology Co.,Ltd. Country or region after: China Address before: No.10 Mucha Road, Zhonglou District, Changzhou City, Jiangsu Province Patentee before: CHANGZHOU ARCHITECTUAL RESEARCH INSTITUTE GROUP Co.,Ltd. Country or region before: China |
|
| CP03 | Change of name, title or address |