CN-121994161-A - Visual identification-based steel bar truss floor carrier plate deformation monitoring method and system

CN121994161ACN 121994161 ACN121994161 ACN 121994161ACN-121994161-A

Abstract

The invention relates to the technical field of deformation detection, in particular to a method and a system for monitoring deformation of a steel bar truss floor support plate based on visual identification, which comprise the following steps: according to the invention, the industrial camera shooting reference image is arranged at the top of the steel bar truss building support plate, the attitude angle sensor is arranged on the camera support frame to collect the reference attitude angle, the current image and the synchronous attitude are taken when the deformation is detected, the attitude deviation is calculated, the pixel coordinates of the camera attitude are corrected, the offset is obtained by comparing the reference coordinates, ten nodes before the deviation are summarized and screened, and the plane deviation state of the building support plate is output.

Inventors

WANG GUANGYU
WANG WENQIANG
SUN GUIYUAN
ZHAO YUJIAN

Assignees

山东高速德建集团有限公司

Dates

Publication Date: 20260508
Application Date: 20260331

Claims (10)

1. The method for monitoring the deformation of the steel bar truss floor support plate based on visual identification is characterized by comprising the following steps of: S1, installing an industrial camera on the top of a steel bar truss floor support plate, shooting an initial moment steel bar truss floor support plate image as a reference image, arranging an attitude angle sensor on an industrial camera support, and collecting a pitch angle and a roll angle of the industrial camera at the initial moment as a reference attitude angle; s2, when the deformation of the steel bar truss floor support plate needs to be detected, acquiring an image of the current steel bar truss floor support plate, and synchronously acquiring the attitude angle of the current industrial camera through an attitude angle sensor; S3, calculating the difference value of the current industrial camera attitude angle and the reference attitude angle to obtain a camera attitude deviation value, and according to a correction formula And Correcting node pixel coordinates in the current steel bar truss floor support plate image to obtain node correction pixel coordinates; Wherein k 1 、k 2 is a roll angle and pitch angle correction coefficient calibrated on site, delta gamma is a roll angle deviation value, The pitch angle deviation value W, H is the image width and high pixel value; S4, selecting node correction pixel coordinates one by one, simultaneously calling reference pixel coordinates of nodes in the reference image, and calculating a difference value between the node correction pixel coordinates and the reference pixel coordinates to obtain a node offset; and S5, summarizing the node offset, screening the node with the absolute value of the offset being ten, and outputting the plane offset state of the whole steel bar truss floor support plate.
2. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 1, wherein the specific steps of S1 are as follows: S101, installing an industrial camera on the top of a steel bar truss floor support plate, firstly shooting an image of the steel bar truss floor support plate at the initial moment of construction, repeatedly shooting three times to select an image with definition meeting the requirement, and generating a reference image; s102, fixing an attitude angle sensor on an industrial camera support, starting an acquisition function, capturing signals of an industrial camera pitch angle and a roll angle at an initial moment, and converting the signals into a digital format to obtain initial attitude angle acquisition data.
3. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 2, wherein the specific steps of S1 are as follows: S103, calling initial attitude angle acquisition data, and extracting pitch angle and roll angle values to obtain a reference attitude angle.
4. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 3, wherein the specific steps of S2 are as follows: S201, when the deformation of the steel bar truss floor support plate needs to be detected, an industrial camera is started, the current steel bar truss floor support plate image is shot, shooting is repeated three times, and the non-shielding and non-blurring image is screened to generate current floor support plate image data; s202, synchronously acquiring pitch angle and roll angle analog signals of a current industrial camera through an attitude angle sensor to obtain the attitude angle of the current industrial camera.
5. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 4, wherein the specific steps of S3 are as follows: S301, respectively extracting corresponding values of a pitch angle and a roll angle according to the current industrial camera attitude angle and the reference attitude angle, correspondingly calculating the difference value of the two groups of values, marking the direction attribute of each difference value, and generating a camera attitude deviation value; S302, sorting according to pitch angle roll angles based on camera attitude deviation values to obtain a camera attitude deviation parameter set.
6. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 5, wherein the specific steps of S3 are as follows: S303, according to the camera attitude deviation parameter set, the original pixel coordinates of the nodes in the current steel bar truss floor support plate image are called, and a correction formula is adopted And Correcting the pixel ordinate by using pitch angle deviation, and correcting the pixel abscissa by using roll angle deviation to obtain a node correction pixel coordinate; Wherein k 1 、k 2 is a roll angle and pitch angle correction coefficient calibrated on site, delta gamma is a roll angle deviation value, For pitch angle offset, W, H is the image wide, high pixel value.
7. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 6, wherein the specific steps of S4 are as follows: S401, selecting node correction pixel coordinates one by one, extracting coordinate values, checking the integrity of coordinate data, and generating a correction pixel coordinate sequence; S402, based on the corrected pixel coordinate sequence, invoking the reference pixel coordinates of the nodes in the reference image, matching the two coordinates according to the same node sequence, and comparing the corresponding relation of the coordinates to obtain a node coordinate matching group.
8. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 7, wherein the specific steps of S4 are as follows: S403, calling a node coordinate matching group, respectively calculating a horizontal coordinate difference value and a vertical coordinate difference value of the correction pixel coordinates and the reference pixels in each group, integrating the two difference values, and marking the positive sign and the negative sign of the difference value to obtain the node offset.
9. The visual identification-based steel bar truss floor carrier plate deformation monitoring method according to claim 8, wherein the specific steps of S5 are as follows: s501, summarizing node offset, sorting into ordered data sets, and generating a node offset summary set; S502, calculating the absolute value of each node offset based on a node offset summary set, sorting the absolute values from large to small, selecting offset data corresponding to ten nodes before sorting, and marking node numbers and offset directions to obtain a high-offset node data set; S503, calling a high-offset node data set, associating the distribution positions of nodes on the floor support plate, analyzing the offset direction concentration, integrating the node offset distribution characteristics, and outputting the plane offset state of the whole steel bar truss floor support plate.
10. The system is used for realizing the visual recognition-based steel bar truss floor carrier plate deformation monitoring method, and comprises a reference data acquisition module, a real-time data acquisition module, an attitude deviation correction module, a node deviation calculation module and a plane deviation output module; The standard data acquisition module is used for installing an industrial camera on the top of the steel bar truss floor support plate, shooting an initial moment steel bar truss floor support plate image as a reference image, arranging an attitude angle sensor on an industrial camera support, and acquiring a pitch angle and a roll angle of the industrial camera at the initial moment as a reference attitude angle; The real-time data acquisition module is used for acquiring an image of the current steel bar truss floor support plate when the deformation of the steel bar truss floor support plate is detected, and synchronously acquiring the attitude angle of the current industrial camera through the attitude angle sensor; The attitude deviation correction module is used for executing calculation of the difference value between the current industrial camera attitude angle and the reference attitude angle to obtain camera attitude deviation, correcting node pixel coordinates in the current steel bar truss floor support plate image to obtain node correction pixel coordinates; the node offset calculation module is used for executing the steps of selecting node correction pixel coordinates one by one, simultaneously calling reference pixel coordinates of nodes in the reference image, and calculating the difference value between the node correction pixel coordinates and the reference pixel coordinates to obtain node offset; the plane offset output module is used for executing the summarized node offset, screening nodes with the absolute value of the offset being ten, and outputting the plane offset state of the whole steel bar truss floor support plate.

Description

Visual identification-based steel bar truss floor carrier plate deformation monitoring method and system Technical Field The invention relates to the technical field of deformation detection, in particular to a method and a system for monitoring deformation of a steel bar truss floor support plate based on visual identification. Background The deformation detection involves the identification and quantitative analysis of the displacement, deflection and geometric form change of the structural member under the action of load, the core matters include the selection of deformation objects, the establishment of deformation references, the acquisition mode of deformation variables and the comparison and judgment method of deformation data, generally, by arranging sensing elements such as displacement meters, dial indicators or strain gauges on the detected member, the displacement values and strain values of the member in the construction or use stage are collected, and the structural configuration change process is monitored and recorded by combining preset reference point coordinates and elevation data, wherein the traditional steel bar truss floor support plate deformation monitoring refers to a technical mode of observing deflection changes of the floor support plate under the action of dead weight and construction load in the installation and concrete pouring processes of the steel bar truss floor support plate, and the technical matters are measurement of the span-center deflection quantity and the overall flatness change of the floor support plate in the construction stage. The single-point contact type measurement mode adopted by the existing deformation monitoring technology depends on a fixed support frame and a preset datum point to complete data acquisition, the support frame is affected by construction disturbance, ground subsidence and external force collision to generate position or posture offset, so that datum point elevation and coordinate offset are driven, the measurement process has no synchronous posture monitoring and coordinate correction mechanism, and errors generated by the datum point offset can be directly counted into deformation data to cause deviation between a measurement result and an actual value. Disclosure of Invention In order to solve the technical problems in the prior art, the embodiment of the invention provides a deformation monitoring method for a steel bar truss floor support plate based on visual identification, which comprises the following steps: S1, installing an industrial camera on the top of a steel bar truss floor support plate, shooting an initial moment steel bar truss floor support plate image as a reference image, arranging an attitude angle sensor on an industrial camera support, and collecting a pitch angle and a roll angle of the industrial camera at the initial moment as a reference attitude angle; s2, when the deformation of the steel bar truss floor support plate needs to be detected, acquiring an image of the current steel bar truss floor support plate, and synchronously acquiring the attitude angle of the current industrial camera through an attitude angle sensor; S3, calculating the difference value of the current industrial camera attitude angle and the reference attitude angle to obtain a camera attitude deviation value, and according to a correction formula AndCorrecting node pixel coordinates in the current steel bar truss floor support plate image to obtain node correction pixel coordinates; Wherein k 1、k2 is a roll angle and pitch angle correction coefficient calibrated on site, delta gamma is a roll angle deviation value, The pitch angle deviation value W, H is the image width and high pixel value; S4, selecting node correction pixel coordinates one by one, simultaneously calling reference pixel coordinates of nodes in the reference image, and calculating a difference value between the node correction pixel coordinates and the reference pixel coordinates to obtain a node offset; and S5, summarizing the node offset, screening the node with the absolute value of the offset being ten, and outputting the plane offset state of the whole steel bar truss floor support plate. As a further scheme of the invention, the specific steps of S1 are as follows: S101, installing an industrial camera on the top of a steel bar truss floor support plate, firstly shooting an image of the steel bar truss floor support plate at the initial moment of construction, repeatedly shooting three times to select an image with definition meeting the requirement, and generating a reference image; s102, fixing an attitude angle sensor on an industrial camera support, starting an acquisition function, capturing signals of an industrial camera pitch angle and a roll angle at an initial moment, and converting the signals into a digital format to obtain initial attitude angle acquisition data. As a further scheme of the invention, the specific steps of S1 are as follows: S103,