A method and device for identifying and measuring reinforcing bars
By using visual and convolutional neural network-based rebar recognition and measurement methods, the problem of low rebar correction efficiency in existing technologies has been solved, realizing intelligent rebar correction and improving construction efficiency and sleeve installation efficiency.
Patent Information
- Application Number
- CN202210732346.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-24
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2042-06-24
AI Technical Summary
Existing technologies for rebar alignment are inefficient, and worker operations are subject to randomness and subjectivity, leading to difficulties in component installation.
A rebar identification and measurement method based on vision and convolutional neural networks is adopted. Images are acquired through a depth camera, preprocessed, and multi-class segmentation models are constructed to identify rebar regions, determine three-dimensional coordinate points, and calculate rebar length.
It improved the efficiency of rebar alignment, reduced the workload of workers, promoted the intelligent development of prefabricated construction, and improved construction efficiency and sleeve installation efficiency.
Smart Images

Figure CN115170489B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of computer technology, and in particular to a method and apparatus for identifying and measuring reinforcing bars. Background Technology
[0002] Precast components are now widely used in the construction of prefabricated structures for buildings and infrastructure. Compared with cast-in-place structures, prefabricated structures have advantages such as modular design and production, short construction cycles, and energy conservation and environmental protection. In prefabricated buildings, the installation of precast components is inevitably a crucial step.
[0003] Precast wall installation involves two main steps: rebar alignment and component installation. First, the rebar positions are checked and aligned. Then, the wall is lowered according to the established control lines to complete the installation. Rebar alignment, ensuring that the protruding rebar positions correspond one-to-one with the sleeve positions of the precast wall, is the core step in precast wall installation.
[0004] Currently, the main method for correcting reinforcing bars is to use a measuring tape to measure the distance of each bar from the edge line (this position corresponds to the distance of the sleeve from the edge of the component at the same location), and then compare it with the information on the drawings to correct any bars that are misaligned. This traditional method is very time-consuming and inefficient; moreover, the randomness and subjectivity of worker operations make it difficult to avoid measurement errors that could lead to difficulties in component installation. Summary of the Invention
[0005] This embodiment provides a method and apparatus for identifying and measuring reinforcing bars, which solves the problem of low efficiency in manual measurement of reinforcing bars in the prior art.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] In a first aspect, embodiments of this application provide a method for identifying reinforcing bars, the method comprising:
[0008] Acquire images;
[0009] The image is preprocessed to obtain a depth image;
[0010] Construct an optimal multi-class segmentation model, wherein the multi-classes include the rebar region and the background region;
[0011] The depth image is processed by the optimal multi-class segmentation model to obtain the rebar region within the depth image.
[0012] Secondly, embodiments of this application also provide a method for measuring reinforcing bars, the method comprising:
[0013] Obtain valid images of areas containing reinforcing bars;
[0014] Determine the three-dimensional coordinates of the upper and lower endpoints of the reinforcing bars within the valid image;
[0015] The length of the reinforcing bar is obtained based on the three-dimensional coordinates of the upper and lower endpoints.
[0016] Thirdly, embodiments of this application also provide a rebar identification device, the device comprising:
[0017] The first acquisition module is used to acquire images;
[0018] The preprocessing module is used to preprocess the image to obtain a depth image;
[0019] The model building module is used to build an optimal multi-class segmentation model, wherein the multi-classes include the reinforcement area and the background area;
[0020] A rebar recognition module is used to obtain the rebar region within the depth image by using the optimal multi-class segmentation model.
[0021] Fourthly, embodiments of this application also provide a rebar measuring device, comprising:
[0022] The second acquisition module is used to acquire valid images of areas containing reinforcing bars;
[0023] The 3D coordinate module is used to determine the 3D coordinates of the upper and lower endpoints of the reinforcing bars within the valid image.
[0024] The calculation module is used to obtain the length of the reinforcing bar based on the three-dimensional coordinates of the upper and lower endpoints.
[0025] The above-mentioned at least one technical solution adopted in the embodiments of this application can achieve the following beneficial effects: intelligent assisted installation can effectively improve the work efficiency of workers and reduce the amount of work done by workers. At the same time, it also lays the foundation for the intelligent construction of prefabricated components. Attached Figure Description
[0026] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:
[0027] Figure 1 A flowchart illustrating a rebar identification method provided in the embodiments of this book;
[0028] Figure 2 A flowchart illustrating a rebar measurement method provided in the embodiments of this specification;
[0029] Figure 3 A structural schematic diagram of a rebar identification device provided in the embodiments described in this book;
[0030] Figure 4 A structural schematic diagram of a rebar measuring device provided in the embodiments described in this book;
[0031] Figure 5 This is a schematic diagram of the rebar identification results provided in the embodiments of this specification;
[0032] Figure 6 This is a schematic diagram of the three-dimensional coordinate system of the reinforcing bars provided in the embodiments of this specification. Detailed Implementation
[0033] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0034] Currently, the main method for correcting reinforcing bars is to use a measuring tape to measure the distance of each bar from the edge line (this position corresponds to the distance of the sleeve from the edge of the component at the same location), and then compare it with the information on the drawings to correct any bars that are misaligned. This traditional method is very time-consuming and inefficient; moreover, the randomness and subjectivity of worker operations make it difficult to avoid measurement errors that could lead to difficulties in component installation.
[0035] Based on this, this invention proposes a method and device for rebar identification and measurement, using vision and convolutional neural networks. This method identifies and locates the rebars, and uses the relative positional relationships between the rebars to assist in sleeve installation. This approach not only improves construction efficiency and saves labor costs, but also promotes the intelligent development of prefabricated construction to a certain extent, serving as an important cornerstone for the automation of prefabricated construction.
[0036] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.
[0037] This disclosure provides a method for identifying reinforcing bars, used to accurately identify reinforcing bar areas in an image, laying the foundation for obtaining the accurate location of the reinforcing bars. The method can be applied, but is not limited to, in the construction market.
[0038] Please see Figure 1 and Figure 5 , Figure 1 A schematic flowchart illustrating a rebar identification method provided in this embodiment of the present disclosure; Figure 5 This is a schematic diagram illustrating the rebar identification results provided in an embodiment of this disclosure. Figure 1 As shown, the method includes the following steps:
[0039] S101, Acquire image.
[0040] In specific implementations, image acquisition methods include, but are not limited to, acquisition devices based on depth cameras or depth camera arrays. Furthermore, image acquisition using depth cameras or depth camera arrays can be used to create a rebar dataset with multiple scenes. This embodiment uses a depth camera for image acquisition, capturing RGB images and corresponding depth images. In specific applications, the image acquisition method is selected based on the usage scenario; no limitation is made here.
[0041] To further clarify, the depth camera can be selected from, but is not limited to, a binocular camera or a ToF-based monocular depth camera, as well as other depth cameras that can meet the accuracy requirements.
[0042] It should be understood that the specific related content listed above is for illustrative purposes only and should not be construed as limiting the scope of this application.
[0043] S103. Preprocess the image to obtain a depth image;
[0044] In specific implementations, the preprocessing methods for the image include, but are not limited to, image correction, image alignment, and image filtering to obtain a depth image.
[0045] Image correction aims to reduce or eliminate image distortion caused by the camera. Taking a binocular depth camera as an example, the Zhang calibration method can be used to calibrate the left and right cameras of the binocular vision system, obtaining the intrinsic and extrinsic parameter matrices (CAMs) of the two cameras respectively. Furthermore, the images captured by the corresponding cameras are corrected using these CAM matrices to eliminate the influence of distortion. The left camera image is taken as the final RGB image, and after image correction, a color image is obtained, denoted as RGB_image.
[0046] Image alignment refers to the pixel alignment operation between the depth image obtained by a depth camera and the RGB_image. Different depth cameras employ different correction methods. Taking a stereo camera as an example, the RGB_image is the left-eye image, while the effective area of the depth image is the overlapping region of the left and right-eye images. Therefore, the depth image is first expanded to the size of the left-eye image, and depth values exceeding the overlapping region are padded with 0s. This ensures a one-to-one correspondence between the pixel coordinates of the depth image and the pixel coordinates of the left-eye RGB image, ultimately obtaining the depth value information for each pixel in the RGB image, i.e., the depth image.
[0047] Image filtering primarily removes noise and outliers from the depth image, improving the accuracy of 3D information. Common image filtering algorithms include mean filtering, median filtering, and Gaussian filtering. The appropriate algorithm can be selected based on the specific depth image characteristics of the chosen embodiment.
[0048] It should be understood that the specific related content listed above is for illustrative purposes only and should not cause any harm to this application.
[0049] S105. Construct an optimal multi-class segmentation model, wherein the multi-class includes the rebar area and the background area;
[0050] In practical implementation, constructing the optimal multi-class segmentation model includes, but is not limited to, constructing a multi-class segmentation model based on image segmentation methods. The multi-class categories include the rebar region and the background region.
[0051] A deep learning dataset is constructed, and the dataset is then filtered to obtain image data containing areas of reinforced concrete. The filtering methods include, but are not limited to, manual filtering.
[0052] The image data containing the reinforced steel bars are divided into a training set and a validation set.
[0053] The multi-class segmentation model is trained using the training set and validated using the validation set in order to obtain the optimal multi-class segmentation model.
[0054] Furthermore, data augmentation is performed on the input data of the multi-class segmentation model, including but not limited to image chroma adjustment, image rotation, image scaling, image translation, and image blurring; specific augmentation strategies are adopted considering the data conditions of specific embodiments, for example, adjusting different chroma to adapt to different lighting environments, such as bright natural light and dim evening light; adjusting different Gaussian blurs to enrich the scenes with different motion blurs during shooting; and using image scaling to adapt to the shapes of recognition objects of different sizes to adapt to complex construction site environments.
[0055] It should be understood that the specific related content listed above is for illustrative purposes only and should not cause any harm to this application.
[0056] S107. The depth image is processed by the optimal multi-class segmentation model to obtain the rebar region within the depth image.
[0057] In specific implementation, such as Figure 5 As shown, the steel reinforcement region and the black background region within the depth image are obtained through the optimal category segmentation model.
[0058] This embodiment can obtain a depth image by preprocessing the image, and then use the depth image to obtain the rebar region within the depth image through the optimal multi-class segmentation model, thereby achieving the goal of accurately identifying the rebar region in the image and laying the foundation for obtaining the accurate location of the rebar. Furthermore, specific data augmentation strategies are added during the multi-class segmentation model training process, such as using Gaussian blur to simulate image blurring and using different brightness levels to simulate different lighting environments, to reduce the impact of environmental factors. Simultaneously, this embodiment uses a depth camera as the main acquisition device, which has the advantages of low cost and high portability, and can be deployed on mobile devices, making it more suitable for prefabricated construction applications.
[0059] Furthermore, this disclosure provides a method for measuring reinforcing bars, which can be applied, but is not limited to, in the construction market.
[0060] See Figure 2 and Figure 5 , Figure 2 A schematic flowchart illustrating a rebar measurement method provided in this embodiment of the present disclosure; Figure 5 This is a schematic diagram of the three-dimensional coordinate system for reinforcing bars provided in an embodiment of this disclosure. Please refer to [link / reference]. Figure 2 As shown in the figure, this disclosure provides a method for measuring reinforcing bars, the method including the following steps:
[0061] S201. Obtain a valid image of the area containing reinforcing bars;
[0062] In practice, the methods for obtaining valid images of areas containing reinforcing bars can be, but are not limited to, referring to the content of a reinforcing bar recognition method, which will not be described again here.
[0063] S203. Determine the three-dimensional coordinates of the upper and lower endpoints of the reinforcing bars within the valid image;
[0064] For specific implementation details, please refer to [link / reference]. Figure 6 As shown, the median line of each rebar in the valid image is obtained, and the median line is extended to both ends of the rebar area to obtain the upper and lower ends of the rebar.
[0065] The depth values of the upper and lower endpoints are obtained according to expression (1), where the depth values refer to the Z values in the coordinate system;
[0066] The three-dimensional coordinates of the upper and lower endpoints are obtained according to expression (2);
[0067] The expression (1) is:
[0068] p_depth=median(depth[(xr):(x+r),(yr):(y+r)]) (1)
[0069] Where depth represents the depth image;
[0070] x and y represent the two-dimensional coordinates of the point in the image;
[0071] r represents taking the two-dimensional coordinate r neighborhood to prevent the depth value drift problem of the depth camera;
[0072] median() represents taking the median of the depth values in all neighborhoods as the depth value of that point;
[0073] p_depth represents the depth value of that point;
[0074]
[0075] Where x and y represent the two-dimensional coordinates of the point in the image;
[0076] X, Y, and Z represent the coordinate values in the camera coordinate system;
[0077] ppx and ppy are camera intrinsic parameters, representing the offset of the optical center;
[0078] fx and fy are camera intrinsic parameters, representing the camera's focal length.
[0079] S205. Obtain the length of the reinforcing bar based on the three-dimensional coordinates of the upper and lower endpoints.
[0080] In practical implementation, the length of the reinforcing bar is obtained by expression (3);
[0081] The expression (3) is
[0082]
[0083] Where X, Y, and Z represent the coordinate values in the camera coordinate system;
[0084] L represents the length of the reinforcing bar.
[0085] This implementation can acquire a valid image of an area containing reinforcing bars; determine the three-dimensional coordinates of the upper and lower endpoints of the reinforcing bars within the valid image; and obtain the length of the reinforcing bars based on the three-dimensional coordinates of the upper and lower endpoints. This achieves intelligent positioning of the reinforcing bars, reducing the workload for workers and improving work efficiency.
[0086] Furthermore, based on the three-dimensional coordinates of the upper and lower endpoints, the relative actual distance between the reinforcing bars is obtained to facilitate better installation of the sleeve.
[0087] In practice, the distance between adjacent reinforcing bars can be calculated by the difference between the three-dimensional coordinates of the upper and lower ends of the reinforcing bars, so as to improve the installation efficiency of the sleeve.
[0088] Furthermore, based on the three-dimensional coordinates of the upper and lower endpoints, the actual distribution of the reinforcing bars is reconstructed in three dimensions using BIM.
[0089] Furthermore, this disclosure also provides a rebar identification device; please refer to [link to relevant documentation]. Figure 3 As shown, Figure 3 This is a structural schematic diagram of a rebar identification device provided in an embodiment of this disclosure. The device includes:
[0090] The first acquisition module 301 is used to acquire images;
[0091] Preprocessing module 302 is used to preprocess the image to obtain a depth image;
[0092] The model construction module 303 is used to build an optimal multi-class segmentation model, wherein the multi-class includes the reinforcement area and the background area;
[0093] The rebar recognition module 304 is used to obtain the rebar region within the depth image by using the optimal multi-class segmentation model.
[0094] The filtering module is used to construct a deep learning dataset and filter the deep learning dataset to obtain image data containing areas with reinforcing steel bars.
[0095] The data partitioning module is used to divide the image data containing the steel reinforcement area into a training set and a validation set.
[0096] The training module is used to train the multi-class segmentation model using the training set and to validate the trained multi-class segmentation model using the validation set in order to obtain the optimal multi-class segmentation model.
[0097] Furthermore, this disclosure provides a rebar measuring device; please refer to [link to relevant documentation]. Figure 4 As shown, Figure 4 This is a schematic diagram of a rebar measuring device provided in an embodiment of the present disclosure. The device includes:
[0098] The second acquisition module 401 is used to acquire a valid image of the area containing reinforcing bars;
[0099] The three-dimensional coordinate module 402 is used to determine the three-dimensional coordinate points of the upper and lower ends of the reinforcing bars within the effective image;
[0100] The calculation module 403 is used to obtain the length of the reinforcing bar based on the three-dimensional coordinates of the upper and lower endpoints.
[0101] The calculation module is also used to obtain the relative actual distance between the reinforcing bars based on the three-dimensional coordinates of the upper and lower endpoints, so as to better install the sleeve.
[0102] The 3D modeling module is used to reconstruct the actual distribution of reinforcing bars in 3D based on the 3D coordinates of the upper and lower endpoints using BIM.
[0103] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0104] Therefore, this application also proposes a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the methods described in any embodiment of this application.
[0105] Furthermore, this application also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any embodiment of this application.
[0106] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0107] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0108] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0109] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0110] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0111] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0112] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0113] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.
Claims
1. A method for identifying reinforcing bars, characterized in that, The method includes: Acquire images; The image is corrected, aligned, and filtered to obtain a depth image. An RGB image and its corresponding depth image are obtained by capturing images with a depth camera. The intrinsic and extrinsic parameter matrices (cam) of the two cameras are obtained respectively. These cam matrices are used to correct the images captured by the corresponding cameras to eliminate distortion. The left-eye camera image is taken as the final RGB image, and after image correction, a color image, denoted as RGB_image, is obtained. Image alignment refers to the pixel alignment operation between the depth image obtained by the depth camera and the RGB_image. Different depth camera principles have corresponding correction methods. This includes: RGB_image is the left-eye image, while the effective area of the depth image is the overlapping area of the left and right-eye images. Therefore, the size of the depth image is first expanded to the size of the left-eye image, and depth values exceeding the overlapping area are filled with 0. This ensures a one-to-one correspondence between the pixel coordinates of the depth image and the pixel coordinates of the left-eye RGB image, ultimately obtaining the depth value information of each pixel in the RGB image, i.e., the depth image. An optimal multi-class segmentation model is constructed, wherein the multi-classes include the rebar region and the background region; the depth image is processed by the optimal multi-class segmentation model to obtain the rebar region within the depth image; Obtain a valid image of the area containing reinforcing bars; obtain the median line of each reinforcing bar in the valid image, extend the median line to both ends of the reinforcing bar area, and obtain the upper and lower endpoints of the reinforcing bars; The depth values of the upper and lower endpoints are obtained according to expression (1), where the depth values refer to the Z values in the coordinate system; The three-dimensional coordinates of the upper and lower endpoints are obtained according to expression (2); wherein, expression (1) is: ; Where depth represents the depth image; x and y represent the two-dimensional coordinates of the point in the image; r represents taking the neighborhood of the two-dimensional coordinate r, in order to prevent the depth value drift problem of the depth camera; median() represents taking the median of all depth values in the neighborhood as the depth value of that point; p_depth represents the depth value of that point; Expression (2) is: ; Where x and y represent the two-dimensional coordinates of the point in the image; X, Y, Z represent the coordinate values in the camera coordinate system; ppx, pp are camera intrinsic parameters, representing the offset of the optical center; fx and fy are camera intrinsic parameters, representing the camera's focal length; The length of the reinforcing bar is obtained based on the three-dimensional coordinates of the upper and lower endpoints.
2. The method of claim 1, wherein The construction of the optimal multi-class segmentation model includes: constructing a multi-class segmentation model based on image segmentation methods; A deep learning dataset is constructed, and the dataset is filtered to obtain image data containing areas with rebar; the image data containing areas with rebar is divided into a training set and a validation set. The multi-class segmentation model is trained using the training set, and the trained model is evaluated using the validation set. The multi-class segmentation model was validated in order to obtain the optimal multi-class segmentation model.
3. The method of claim 1, wherein The step of obtaining the length of the reinforcing bar based on the three-dimensional coordinates of the upper and lower endpoints includes: the length of the reinforcing bar is obtained by expression (3); the expression (3) is: ; Where X, Y, and Z represent the coordinate values in the camera coordinate system; L represents the length of the reinforcing bar.
4. The method of claim 1, wherein The process of determining the three-dimensional coordinates of the upper and lower endpoints of the reinforcing bars within the valid image further includes: obtaining the relative actual distance between the reinforcing bars based on the three-dimensional coordinates of the upper and lower endpoints. To allow for better installation of the sleeve.
5. The method of claim 4, wherein The process of determining the three-dimensional coordinates of the upper and lower endpoints of the reinforcing bars within the valid image further includes: based on the three-dimensional coordinates of the upper and lower endpoints, performing BIM-based analysis of the actual distribution of the reinforcing bars. The layout is then reconstructed in three dimensions.
6. A reinforcing bar identification apparatus characterized by comprising: The device includes: a first acquisition module, used to acquire images; The preprocessing module is used to preprocess the image to obtain a depth image; The model building module is used to build an optimal multi-class segmentation model, wherein the multi-classes include the reinforcement area and the background area; The rebar recognition module is used to process the depth image through the optimal multi-class segmentation model. The first module obtains the rebar region within the depth image; the second acquisition module is used to acquire a valid image containing the rebar region. The 3D coordinate module is used to determine the 3D coordinates of the upper and lower endpoints of the reinforcing bars within the valid image. The calculation module is used to obtain the length of the reinforcing bar based on the three-dimensional coordinates of the upper and lower endpoints.
Citation Information
Patent Citations
ROI region extraction method based on deep learning image semantic segmentation model
CN109858539A
Object two-dimensional size measurement method and device
CN111524193A
Trinocular vision-based non-contact deformation detection method and system for an insulator
CN113884017A
Vision-based simultaneous automatic identification and size measurement method for multiple aviation luggage
CN114140514A
