Construction site precast beam steel bar production management system
By introducing BIM technology and neural convolutional networks into the steel bar production management system, the three-dimensional steel bar model is automatically constructed and image processing is performed, which solves the problem of low intelligence in the existing technology and realizes more efficient steel bar production management and detection.
Patent Information
- Application Number
- CN202411136797.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-12
AI Technical Summary
The existing steel bar production management system is relatively low in intelligence, and workers need to manually establish production plans and conduct inspections, which is time-consuming and labor-intensive.
A production management system for prefabricated beam steel bars on construction site was designed, including a center console, production module and detection module. BIM technology is used to build a three-dimensional steel bar model, combined with neural convolutional network for image processing, automatically generate production schedules and perform dimension detection. .
The intelligent degree of steel bar production management has been improved, the operation is more convenient and fast, the degree of automation is improved, and the inspection results are accurate and reliable.
Smart Images

Figure CN118941246A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of steel bar production, and in particular to a production management system for prefabricated beam steel bars at a construction site. Background Art
[0002] Steel bar production management is an integral part of on-site manufacturing of prefabricated beams, but the existing steel bar production management system still has shortcomings. Specifically, the existing steel bar production management requires workers to manually establish production schedules and test the produced steel bars. The operation is time-consuming and labor-intensive, and the degree of intelligence is low.
[0003] Therefore, a construction site prefabricated beam reinforcement production management system is needed to solve the problems raised in the above background technology. Summary of the invention
[0004] The purpose of the present invention is to provide a construction site prefabricated beam reinforcement production management system to solve the problems raised in the above background technology.
[0005] To achieve the above object, the present invention provides the following technical solutions:
[0006] A construction site prefabricated beam steel bar production management system comprises a central control console, a production module and a detection module. The central control console is internally installed with production management software, and the production management software comprises a model building unit, a production planning unit and a detection unit.
[0007] As a preferred solution of the present invention, the production module and the detection module are connected to the central control console via wires, and multiple groups of the production modules and the detection modules are provided.
[0008] As a preferred solution of the present invention, the production module includes various steel bar bending and cutting machines, and the detection module includes various color cameras.
[0009] As a preferred solution of the present invention, the specific analysis steps of the model building unit are: the model building unit uses BIM technology to extract and build a three-dimensional model of the precast beam segment to be produced from the precast beam information library, removes the concrete model block in the three-dimensional model, obtains the three-dimensional model of the steel bar at the end of the precast beam, decomposes the three-dimensional model of the steel bar, obtains several groups of steel bar model blocks, and the model building unit inputs the obtained steel bar model set into the production planning unit.
[0010] As a preferred solution of the present invention, the specific analysis steps of the production planning unit are: after receiving the steel bar model block, the production planning unit extracts key points and their corresponding three-dimensional coordinates from the current steel bar model block, determines the coordinate vertices corresponding to each key point in the steel bar model block based on the three-dimensional coordinates corresponding to each key point, connects each coordinate vertex in sequence, generates a basic grid model corresponding to the current steel bar, brings the basic grid model into a preset distance and model generation parameter relationship model, obtains the steel bar model block parameters corresponding to the basic grid model, the production planning unit establishes a production plan table according to the obtained parameters, and inputs the production plan table into the production module, and the production module produces steel bars according to the production plan table.
[0011] As a preferred solution of the present invention, the specific analysis steps of the detection unit are as follows: the detection module collects the color image of the steel bars produced by the production module, and inputs the collected image into the central control console; the detection unit constructs a neural convolution network, and the neural convolution network performs redundant wavelet transform on the input image data to obtain low-frequency approximate components and high-frequency detail components in three directions, and then multiplies the high-frequency detail components in three directions with the adjacent scale wavelet coefficients to obtain a wavelet coefficient map after multiplying the wavelet coefficients in three different directions, and then linearly processes the wavelet coefficient map. After the processing is completed, the wavelet coefficient map is threshold processed, and the low-frequency approximate component is edge processed to obtain a low-frequency sub-band, and the denoised wavelet coefficient map and the low-frequency sub-band are subjected to maximum value processing, and the image at each position is processed. The pixel point with the largest pixel value at the corresponding position of the four images is taken as the edge pixel point of the image, so as to obtain the edge image, enhance the extracted edge image, expand the grayscale range, obtain the steel bar feature image, calculate the first contour vertex coordinates of the steel bar feature image, and obtain the actual vertex coordinates of the steel bar, establish a perspective transformation matrix based on the first contour vertex coordinates and the actual vertex coordinates, use the perspective transformation matrix to map the prefabricated box girder steel mesh feature image to obtain a restored image, extract the second contour of the steel bar in the restored image, calculate the second contour vertex coordinates of the second contour, use the distance formula to calculate the distance between the two second contour vertex coordinates on the diagonal line and obtain the difference, verify the difference based on a preset deviation threshold, and output the size detection result.
[0012] As a preferred solution of the present invention, the method for constructing the relationship model between the preset distance and the model generation parameter is as follows: obtaining historical steel bar model samples, the historical steel bar model samples are steel bar models without field of view loss, the historical steel bar model samples include the distance between the steel bar model and the camera, the model generation parameters of the steel bar model and its corresponding model drawing efficiency, taking the model drawing efficiency as the training goal, learning the relationship between the distance between the steel bar model and the camera and the model generation parameters in each historical steel bar model sample, generating a relationship model between the preset distance and the model generation parameter, wherein the steel bar model block parameters include the length, shape, inclination and other dimensional data of the steel bar model block.
[0013] As a preferred solution of the present invention, the specific steps of the redundant wavelet transform are: using a wavelet filter to process each pixel of the image data first by row and then by column, so as to obtain a low-frequency approximate component of the same size as the original image and high-frequency detail components in three directions.
[0014] As a preferred solution of the present invention, the high-frequency detail components in the three directions correspond to a high-frequency sub-image retaining horizontal detail information, a high-frequency sub-image containing vertical detail information, and a high-frequency sub-image reflecting diagonal details. The enhancement processing is to perform grayscale enhancement processing on the image through IMAQCastImage2 and IMAQThreshold functions, and the neural convolution network reads and stores the collected image through IMAQCreate and IMAQReadFile2 functions of LABVIEW.
[0015] Compared with the prior art, the present invention has the following beneficial effects:
[0016] 1. In the present invention, a model building unit uses BIM technology to extract and build a three-dimensional model of a precast beam segment to be produced from a precast beam information library, removes the concrete model block in the three-dimensional model, obtains a three-dimensional model of a steel bar at the end of the precast beam, decomposes the three-dimensional model of the steel bar, and obtains a plurality of groups of steel bar model blocks. The model building unit inputs the obtained steel bar model set into the production planning unit. After receiving the steel bar model block, the production planning unit generates a basic grid model corresponding to the current steel bar, substitutes the basic grid model into a preset distance and model generation parameter relationship model, obtains steel bar model block parameters corresponding to the basic grid model, and the production planning unit establishes a production plan table according to the obtained parameters, and inputs the production plan table into the production module. The production module executes the production plan according to the production plan. The table is used for steel bar production. The detection module collects the color image of the steel bars produced by the production module and inputs the collected image into the central console. The detection unit constructs a neural convolution network. The neural convolution network performs redundant wavelet transform on the input image data to obtain a wavelet coefficient map after multiplying the wavelet coefficients in three different directions. Then, the wavelet coefficient map is subjected to linear processing, threshold processing, edge processing, maximum processing, and enhancement processing to obtain the steel bar feature image. The distance formula and the steel bar feature image are used to process the size detection result. The BIM technology is used to construct the steel bar model and establish a production schedule. During the detection, the image is processed by the neural convolution network to obtain the size detection structure. The operation is relatively convenient and fast, and the degree of intelligence is relatively high. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a schematic diagram of the overall hardware structure of the present invention;
[0018] Figure 2 It is a schematic diagram of the system block structure of the present invention.
[0019] In the figure: 1. Central control console; 2. Production module; 3. Inspection module; 4. Production management software; 5. Model building unit; 6. Production planning unit; 7. Inspection unit. DETAILED DESCRIPTION
[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0021] To facilitate understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.
[0022] It should be noted that when an element is referred to as being "fixed on" another element, it may be directly on the other element or there may also be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may also be a central element. The terms "vertical", "horizontal", "left", "right" and similar expressions used in this article are for illustrative purposes only.
[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.
[0024] For examples, see Figure 1-2 , the present invention provides a technical solution:
[0025] A construction site prefabricated beam reinforcement production management system includes a central control console 1, a production module 2 and a detection module 3. The central control console 1 is internally installed with production management software 4, and the production management software 4 includes a model building unit 5, a production planning unit 6 and a detection unit 7.
[0026] Furthermore, the production module 2 and the detection module 3 are connected to the central control console 1 through wires, and the production module 2 and the detection module 3 are provided in multiple groups.
[0027] Furthermore, the production module 2 includes various steel bar bending and cutting machines, and the detection module 3 includes various color cameras.
[0028] Furthermore, the specific analysis steps of the model construction unit 5 are: the model construction unit 5 uses BIM technology to extract and construct a three-dimensional model of the precast beam segment to be produced from the precast beam information library, removes the concrete model blocks in the three-dimensional model, obtains the three-dimensional model of the steel bar at the end of the precast beam, decomposes the three-dimensional model of the steel bar, obtains several groups of steel bar model blocks, and the model construction unit 5 inputs the obtained steel bar model set into the production planning unit 6.
[0029] Furthermore, the specific analysis steps of the production planning unit 6 are as follows: after receiving the steel bar model block, the production planning unit 6 extracts key points and their corresponding three-dimensional coordinates from the current steel bar model block, determines the coordinate vertices corresponding to each key point in the steel bar model block based on the three-dimensional coordinates corresponding to each key point, connects each coordinate vertex in sequence, generates a basic grid model corresponding to the current steel bar, substitutes the basic grid model into the preset distance and model generation parameter relationship model, obtains the steel bar model block parameters corresponding to the basic grid model, and the production planning unit 6 establishes a production plan table according to the obtained parameters, and inputs the production plan table into the production module 2, and the production module 2 produces steel bars according to the production plan table.
[0030] Furthermore, the specific analysis steps of the detection unit 7 are as follows: the detection module 3 collects the color image of the steel bars produced by the production module 2, and inputs the collected image into the central control console 1; the detection unit 7 constructs a neural convolution network, and the neural convolution network performs redundant wavelet transform on the input image data to obtain low-frequency approximate components and high-frequency detail components in three directions, and then multiplies the high-frequency detail components in three directions with the adjacent scale wavelet coefficients to obtain a wavelet coefficient map after multiplying the wavelet coefficients in three different directions, and then linearly processes the wavelet coefficient map. After the processing is completed, the wavelet coefficient map is threshold processed, and the low-frequency approximate component is edge processed to obtain a low-frequency sub-band, and the denoised wavelet coefficient map and the low-frequency sub-band are subjected to maximum value processing, and the pixels at each position are The points with the largest pixel values at the corresponding positions of the four images are selected as the edge pixel points of the image to obtain the edge image, and the extracted edge image is enhanced to expand the grayscale range to obtain the steel bar feature image, calculate the first contour vertex coordinates of the steel bar feature image, and obtain the actual vertex coordinates of the steel bar, establish a perspective transformation matrix based on the first contour vertex coordinates and the actual vertex coordinates, use the perspective transformation matrix to map the prefabricated box girder steel mesh feature image to obtain a restored image, extract the second contour of the steel bar in the restored image, calculate the second contour vertex coordinates of the second contour, use the distance formula to calculate the distance between the two second contour vertex coordinates on the diagonal line and obtain the difference, verify the difference based on the preset deviation threshold, and output the size detection result.
[0031] Furthermore, a method for constructing a model of the relationship between a preset distance and model generation parameters is as follows: obtaining historical steel bar model samples, the historical steel bar model samples are steel bar models without field of view loss, the historical steel bar model samples include the distance between the steel bar model and the camera, the model generation parameters of the steel bar model and its corresponding model drawing efficiency, taking the model drawing efficiency as the training target, learning the relationship between the distance between the steel bar model and the camera and the model generation parameters in each historical steel bar model sample, generating a model of the relationship between the preset distance and the model generation parameters, and the steel bar model block parameters include the length, shape, inclination and other dimensional data of the steel bar model block.
[0032] Furthermore, the specific steps of redundant wavelet transform are: using wavelet filter to process each pixel of image data first by row and then by column, so as to obtain low-frequency approximate components of the same size as the original image and high-frequency detail components in three directions.
[0033] Furthermore, the high-frequency detail components in the three directions correspond to the high-frequency sub-images that retain the horizontal detail information, the high-frequency sub-images that contain the vertical detail information, and the high-frequency sub-images that reflect the diagonal details. The enhancement processing is to perform grayscale enhancement on the image through the IMAQCastImage2 and IMAQThreshold functions, and the neural convolution network reads and stores the collected images through the IMAQCreate and IMAQReadFile2 functions of LABVIEW.
[0034] The workflow of the present invention is as follows: when a prefabricated beam steel bar production management system designed by the present invention is in operation, the model building unit 5 uses BIM technology to extract and build a three-dimensional model of the prefabricated beam segment to be produced from the prefabricated beam information library, removes the concrete model block in the three-dimensional model, obtains the three-dimensional model of the steel bar at the end of the prefabricated beam, decomposes the three-dimensional model of the steel bar, obtains several groups of steel bar model blocks, and the model building unit 5 inputs the obtained steel bar model set into the production planning unit 6;
[0035] The production planning unit 6 obtains a historical steel bar model sample, which is a steel bar model without field of view loss. The historical steel bar model sample includes the distance between the steel bar model and the camera, the model generation parameters of the steel bar model and its corresponding model drawing efficiency. The model drawing efficiency is used as a training target, and the relationship between the distance between the steel bar model and the camera and the model generation parameters in each historical steel bar model sample is learned to generate a preset distance and model generation parameter relationship model. The steel bar model block parameters include the length, shape, inclination and other dimensional data of the steel bar model block;
[0036] After receiving the steel bar model block, the production planning unit 6 extracts key points and their corresponding three-dimensional coordinates from the current steel bar model block, determines the coordinate vertices corresponding to each key point in the steel bar model block based on the three-dimensional coordinates corresponding to each key point, connects each coordinate vertex in sequence, generates a basic grid model corresponding to the current steel bar, substitutes the basic grid model into the preset distance and model generation parameter relationship model, obtains the steel bar model block parameters corresponding to the basic grid model, and the production planning unit 6 establishes a production schedule according to the obtained parameters, and inputs the production schedule into the production module 2, and the production module 2 performs steel bar production according to the production schedule;
[0037] The detection module 3 collects the color image of the steel bars produced by the production module 2, and inputs the collected image into the central control console 1. The detection unit 7 constructs a neural convolution network. The neural convolution network uses a wavelet filter to process each pixel of the image data by row and then by column, thereby obtaining a low-frequency approximate component of the same size as the original image and high-frequency detail components in three directions. The high-frequency detail components in the three directions correspond to a high-frequency sub-image that retains horizontal detail information, a high-frequency sub-image that contains vertical detail information, and a high-frequency sub-image that reflects diagonal details. Then, the high-frequency detail components in the three directions are multiplied by the adjacent scale wavelet coefficients to obtain three wavelet coefficient images after the wavelet coefficients in different directions are multiplied. Then, the wavelet coefficient image is linearly processed. After the processing is completed, the wavelet coefficient image is threshold processed, and the low-frequency approximate component is edge processed to obtain a low-frequency sub-band. The denoised wavelet coefficient image and the low-frequency sub-band are subjected to maximum value processing. The pixel point at each position takes the point with the largest pixel value at the corresponding position of the four images as the edge pixel point of the image, thereby obtaining an edge image.
[0038] The extracted edge image is enhanced by IMAQCastImage2 and IMAQThreshold functions to expand the grayscale range and obtain the steel bar feature image. The first contour vertex coordinates of the steel bar feature image are calculated, and the actual vertex coordinates of the steel bar are obtained. A perspective transformation matrix is established based on the first contour vertex coordinates and the actual vertex coordinates. The perspective transformation matrix is used to map the prefabricated box girder steel mesh feature image to obtain a restored image. The second contour of the steel bar in the restored image is extracted, and the second contour vertex coordinates of the second contour are calculated. The distance formula is used to calculate the distance between the two second contour vertex coordinates on the diagonal line and the difference is obtained. The difference is verified based on the preset deviation threshold, and the size detection result is output.
[0039] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A construction site prefabricated beam steel bar production management system, comprising a central control console (1), a production module (2) and a detection module (3), characterized in that: The central control console (1) is internally installed with production management software (4), and the production management software (4) includes a model building unit (5), a production planning unit (6) and a detection unit (7).
2. A construction site prefabricated beam reinforcement production management system according to claim 1, characterized in that: The production module (2) and the detection module (3) are both connected to the central control console (1) via wires, and a plurality of groups of the production module (2) and the detection module (3) are provided.
3. A construction site prefabricated beam reinforcement production management system according to claim 1, characterized in that: The production module (2) includes various steel bar bending and cutting machines, and the detection module (3) includes various color cameras.
4. A construction site prefabricated beam reinforcement production management system according to claim 1, characterized in that: The specific analysis steps of the model building unit (5) are as follows: the model building unit (5) uses BIM technology to extract and build a three-dimensional model of the precast beam segment to be produced from the precast beam information library, removes the concrete model block in the three-dimensional model, obtains the three-dimensional model of the steel bar at the end of the precast beam, decomposes the three-dimensional model of the steel bar to obtain a plurality of groups of steel bar model blocks, and the model building unit (5) inputs the obtained steel bar model set into the production planning unit (6).
5. The construction site prefabricated beam reinforcement production management system according to claim 1, characterized in that: The specific analysis steps of the production planning unit (6) are as follows: after receiving the steel bar model block, the production planning unit (6) extracts key points and their corresponding three-dimensional coordinates from the current steel bar model block, determines the coordinate vertices corresponding to each key point in the steel bar model block based on the three-dimensional coordinates corresponding to each key point, sequentially connects each coordinate vertex to generate a basic grid model corresponding to the current steel bar, substitutes the basic grid model into a preset distance and model generation parameter relationship model to obtain the steel bar model block parameters corresponding to the basic grid model, the production planning unit (6) establishes a production plan table according to the obtained parameters, and inputs the production plan table into the production module (2), and the production module (2) produces steel bars according to the production plan table.
6. A construction site prefabricated beam reinforcement production management system according to claim 1, characterized in that: The specific analysis steps of the detection unit (7) are as follows: the detection module (3) collects the color image of the steel bars produced by the production module (2), and inputs the collected image into the central control console (1); the detection unit (7) constructs a neural convolution network, and the neural convolution network performs redundant wavelet transform on the input image data to obtain low-frequency approximate components and high-frequency detail components in three directions, then multiplies the high-frequency detail components in the three directions with the adjacent scale wavelet coefficients to obtain a wavelet coefficient map after the wavelet coefficients in three different directions are multiplied, and then linear processing is performed on the wavelet coefficient map. After the processing is completed, threshold processing is performed on the wavelet coefficient map, and edge processing is performed on the low-frequency approximate component to obtain a low-frequency sub-band. The denoised wavelet coefficient map and the low-frequency sub-band are subjected to maximum value processing. Each position The pixel points with the largest pixel value at the corresponding positions of the four images are taken as the edge pixel points of the image, so as to obtain the edge image, enhance the extracted edge image, expand the grayscale range, obtain the steel bar feature image, calculate the first contour vertex coordinates of the steel bar feature image, and obtain the actual vertex coordinates of the steel bar, establish a perspective transformation matrix based on the first contour vertex coordinates and the actual vertex coordinates, use the perspective transformation matrix to map the prefabricated box girder steel mesh feature image to obtain a restored image, extract the second contour of the steel bar in the restored image, calculate the second contour vertex coordinates of the second contour, use the distance formula to calculate the distance between the two second contour vertex coordinates on the diagonal line and obtain the difference, verify the difference based on a preset deviation threshold, and output the size detection result.
7. A construction site prefabricated beam reinforcement production management system according to claim 5, characterized in that: The method for constructing the relationship model between the preset distance and the model generation parameter is as follows: obtaining historical steel bar model samples, the historical steel bar model samples are steel bar models without field of view loss, the historical steel bar model samples include the distance between the steel bar model and the camera, the model generation parameters of the steel bar model and its corresponding model drawing efficiency, taking the model drawing efficiency as the training target, learning the relationship between the distance between the steel bar model and the camera and the model generation parameters in each historical steel bar model sample, and generating a relationship model between the preset distance and the model generation parameters, wherein the steel bar model block parameters include the length, shape, inclination and other dimensional data of the steel bar model block.
8. The construction site prefabricated beam reinforcement production management system according to claim 6, characterized in that: The specific steps of the redundant wavelet transform are: using a wavelet filter to process each pixel of the image data first by row and then by column, so as to obtain a low-frequency approximate component of the same size as the original image and high-frequency detail components in three directions.
9. A construction site prefabricated beam reinforcement production management system according to claim 6, characterized in that: The high-frequency detail components in the three directions correspond to the high-frequency sub-image retaining the horizontal detail information, the high-frequency sub-image containing the vertical detail information and the high-frequency sub-image reflecting the diagonal details. The enhancement processing is to perform grayscale enhancement processing on the image through the IMAQCastImage2 and IMAQThreshold functions, and the neural convolution network reads and stores the collected image through the IMAQCreate and IMAQReadFile2 functions of LABVIEW.
Citation Information
Cited By
Construction site precast beam steel bar production management system
CN120355178A
