A method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model

Abstract

This invention discloses a method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model. First, the location information of the rebar sleeve joints in the rebar image is obtained using the YOLOv8 model. Then, a depth map is generated using the Monodepth2 model to obtain rebar sleeve joints with the same average depth. Next, the MobileSAM model is used to obtain segmented mask images of rebar sleeve joints in the same layer. Finally, a subpixel edge detection algorithm is used to detect the actual spacing of the rebar sleeve joints. This invention combines the YOLOv8 model, Monodepth2 model, MobileSAM model, and subpixel edge detection algorithm to detect the spacing of rebar sleeve joints, especially in multi-layer rebar meshes. This solves the problems in existing technologies where it is difficult to detect the spacing of rebars in multi-layer rebar meshes and easy to calculate the spacing of rebars in the wrong layer, leading to incorrect results, thus improving the accuracy of rebar spacing detection.

Description

Technical Field

[0001] This invention belongs to the field of engineering acceptance technology, specifically relating to a method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model. Background Technology

[0002] In cast-in-place structures, reinforcing steel bars are the primary load-bearing components, used in enormous quantities. Their selection and connection directly affect the safety and service life of the building. In production, the main methods of connecting reinforcing steel bars include lap splices, welding, and mechanical connections. Among these, mechanical connections are considered the "third generation of steel bar joints" following lap splices and welding. Mechanical connections often use steel bar sleeve joints, which offer advantages such as energy saving and are not limited by the composition or type of reinforcing steel. To ensure that the steel bar connections meet the requirements of engineering design and construction specifications, the connections need to be inspected. Improper spacing of the reinforcing bars, whether too large or too small, can cause uneven stress on the cast-in-place slab, thereby threatening the stability and safety of the concrete structure.

[0003] Currently, the inspection and acceptance of reinforcing bars still rely on manual labor. Inspectors use traditional measuring tools such as tape measures to measure the spacing of the reinforcing bars. If the spacing is found to be non-compliant with standards, adjustments are made or reinforcement is added. Due to the large number of reinforcing bars in cast-in-place structures, manual inspection is inefficient, labor-intensive, and has low accuracy, and there is also the possibility of errors and omissions.

[0004] In recent years, fully automated detection of rebar spacing in rebar meshes has been promoted to some extent. Existing technology discloses a single-layer rebar segmentation method and system for intelligent rebar detection and tying. This method extracts and segments single-layer rebar point cloud data from color and depth images of the rebar skeleton, enabling intelligent detection of rebar intersections and spacing calculation. However, this existing technology utilizes color and depth images of the rebar skeleton for point cloud data extraction and segmentation, a complex process with high computational load, making it unsuitable for real-time tasks. Existing technology also discloses a computer vision-based method for the acceptance of concealed engineering rebar structures. This method generates a binarized image of the rebar mesh using deep learning algorithms, draws multiple vertical survey lines between two target rebars, measures the pixel distance between the intersection points of these survey lines and the two target rebars, and takes the average value to calculate the rebar mesh spacing. However, this existing technology is only suitable for single-layer, unobstructed rebar arrangements and is insufficient for multi-layer rebar meshes or situations with a large number of rebars. Existing technology discloses a rebar spacing detection method based on an improved YOLOv7 algorithm. This method detects rebar intersections using the improved YOLOv7 algorithm, calculates the center point based on the coordinates of the intersection detection box, divides the intersections into different sets using a disjoint-set data structure, fits straight lines using the least squares method, and finally calculates the distance between two parallel lines. Multiplying this distance by the image scaling factor yields the actual rebar spacing. However, this existing technology is only suitable for single-layer rebar structures with clear intersections. For rebar structures with large angles or blurred intersections, it may lead to inaccurate fitting, affecting the spacing calculation.

[0005] Therefore, in response to the above problems and technical requirements, it is necessary to develop a new method for detecting rebar spacing that can quickly and efficiently measure the rebar spacing of multi-layer rebar meshes to ensure the quality of reinforced concrete buildings. Summary of the Invention

[0006] To address the aforementioned problems, the present invention aims to provide a method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model. For rebars connected by sleeve joints, depth estimation and the MobileSAM model are used to obtain segmented mask images of the sleeve joints in the same layer. Then, a sub-pixel edge detection algorithm is used to calculate the spacing of the rebar sleeve joints, thus realizing the detection of rebar spacing in multi-layer rebar meshes. This method can be applied to rebar connection projects under complex working conditions, solving the problem of difficulty in detecting rebar spacing in multi-layer rebar meshes in the prior art.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model includes the following steps:

[0009] (1) Real-time acquisition of rebar images and input into the constructed YOLOv8 model of rebar sleeve joint. The model selects the rebar sleeve joint in the rebar image and outputs the coordinates (x, y) of the center point of the rebar sleeve joint frame. center y center ), width of the box, height of the box;

[0010] (2) Input the rebar image into the Monodepth2 model to generate a depth map. The model reads the corresponding (x... center y center ), width, height, and then select the rebar sleeve joint frame in the depth map, and output the coordinates of the lower left corner of the rebar sleeve joint frame (x, y ... min y min ) and the coordinates of the upper right corner (x max y max ), calculate the average depth of each rebar sleeve joint frame and retain rebar sleeve joints with the same average depth and their (x min y min ), (x max y max );

[0011] (3) Combine the images of the reinforcing bars, the sleeve joints of the reinforcing bars with the same average depth, and their (x) min y min ), (x max y max Input the data into the MobileSAM model, segment the rebar sleeve joints, and obtain the segmentation mask image of the rebar sleeve joints in the same layer;

[0012] (4) Use the sub-pixel edge detection algorithm to calculate the coordinates of the center point (c) of the rebar sleeve joint in the segmented mask image. x c y The actual spacing of the rebar sleeve joints was detected.

[0013] In this invention, the location information of each rebar sleeve joint in the rebar image is first obtained using the YOLOv8 model of rebar sleeve joints. Then, based on the location information, a depth map is generated using the Monodepth2 model, and the average depth of each rebar sleeve joint bounding box is calculated to obtain rebar sleeve joints with the same average depth, i.e., rebar sleeve joints located in the same layer. Next, the rebar sleeve joints are segmented using the MobileSAM model to obtain the segmentation mask image of rebar sleeve joints located in the same layer. Then, the accuracy of the edge positioning of the rebar sleeve joints is further improved using the subpixel edge detection algorithm to obtain a more accurate rebar sleeve joint image and calculate the center point coordinates of the rebar sleeve joints, thereby realizing the detection of the rebar sleeve joint spacing in multi-layer rebar mesh.

[0014] In step (1) of this invention, the construction of the YOLOv8 model of the rebar sleeve joint includes the following steps: obtaining the rebar image dataset, inputting the rebar image dataset into the YOLOv8 model, training the YOLOv8 model, and obtaining the YOLOv8 model of the rebar sleeve joint.

[0015] As is common knowledge, rebar image datasets need to be labeled before being input into the YOLOv8 model. Preferably, the rebar image dataset is labeled using labeling software, specifically conventional labeling software and conventional techniques.

[0016] Preferably, data augmentation techniques are used to expand the established rebar image dataset. These techniques are existing technologies, such as mirroring, scaling, cropping, or color space enhancement of the images to increase the size of the dataset and improve the model's generalization ability and robustness.

[0017] Furthermore, the rebar image dataset was uniformly adjusted to a size of 640×640. 640x640 is a commonly used input size for the YOLOv8 model.

[0018] In step (2) of this invention, , , , ;in, The width of the original image. The height of the original image.

[0019] In step (2) of this invention, the formula for calculating the average depth of the steel bar sleeve joint frame is as follows: Where A is the number of pixels and D(x,y) is the depth value at coordinates (x,y).

[0020] In step (2) of this invention, a depth threshold is set to retain steel bar sleeve joints with similar average depths and their (x min y min ), (x max y max The calculation formula for retaining rebar sleeve joints with similar average depths is as follows: ;in, Let be the average depth of the i-th rebar sleeve joint frame. Let j be the average depth of the j-th rebar sleeve joint frame. This is the depth threshold.

[0021] In step (4) of this invention, , ;in, , , I(x,y) is the pixel value at coordinates (x,y).

[0022] In step (4) of this invention, the sub-pixel edge detection algorithm is used to calculate the coordinates (c) of the center point of the rebar sleeve joint in the segmented mask image. x c y The pixel pitch of the rebar sleeve joint is obtained, and then converted into the actual world pitch. Preferably, a calibration reference is placed on site, and the actual size L of the reference is used to determine the pixel pitch. real and its pixel size L in the image image To obtain the image scaling ratio Then, by multiplying the pixel size in the image by the image scaling ratio, the actual spacing of the rebar sleeve joint can be detected. .

[0023] Preferably, in step (4), the subpixel edge detection algorithm includes morphological opening operation, Canny edge detection operator, Devernay algorithm, and minimum bounding rectangle algorithm. Morphological opening operation can eliminate small noise and false targets in the segmented mask image, while preserving the main shape features of the rebar sleeve joint, making the image edges smoother and providing a good foundation for subsequent edge detection; Canny edge detection operator can accurately capture edge details and has strong anti-noise ability, ensuring that the extracted edges are clear; Devernay algorithm can accurately locate the edges, making them more continuous and smooth, eliminating the jaggedness problem, and enhancing the accuracy and consistency of the edges; minimum bounding rectangle algorithm can fit the edges and calculate the coordinates of the center point of the rectangle.

[0024] The YOLOv8 model, Monodepth2 model, MobileSAM model, subpixel edge detection algorithm, and method for training the YOLOv8 model used in this invention are all prior art and do not affect the understanding of this invention by those skilled in the art.

[0025] This invention discloses a rebar sleeve joint spacing detection system based on depth estimation and the MobileSAM model, comprising:

[0026] The acquisition module is used to acquire images of the reinforcing mesh;

[0027] The central control module is used to execute the above-mentioned method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model.

[0028] This invention discloses a storage medium storing a processor-executable program for executing the above-described method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model.

[0029] An electronic device includes a memory and a processor, the memory storing a program for executing the above-described method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model.

[0030] This invention discloses the application of the above-mentioned method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model in detecting the spacing of rebar sleeve joints.

[0031] Due to the application of the above technical solutions, the beneficial effects of this invention are as follows: First, based on depth estimation and the MobileSAM model, this invention creatively proposes a method for detecting the spacing of rebar sleeve joints. By estimating and segmenting the depth of the rebar sleeve joints, rebar sleeve joints located in the same layer can be identified. Then, the spacing of the rebar sleeve joints is obtained using a sub-pixel edge detection algorithm. This solves the problem in the prior art that it is difficult to detect the spacing of rebars in multi-layer rebar meshes and that it is easy to calculate the spacing of rebars in the wrong layer, thus improving the accuracy of rebar spacing detection. Second, by applying digital image processing and machine vision detection technology to the detection of the spacing of rebars connected by rebar sleeve joints, this invention effectively reduces the problems of large workload, high labor intensity, low efficiency, and easy misdetection and missed detection that occur when using a measuring tape manually. Third, this invention realizes the automation of rebar spacing detection. The entire acquisition and calculation process does not require manual operation, improving the detection efficiency and accuracy of rebars connected by rebar sleeve joints. In particular, it can detect the spacing of rebars in multi-layer rebar meshes and has wide applicability. Attached Figure Description

[0032] Figure 1 This is a flowchart of the method steps of the present invention.

[0033] Figure 2 This is a diagram showing the location detection results of the rebar sleeve joint in the application experiment.

[0034] Figure 3 This is a depth diagram of the rebar sleeve joint used in the application experiment.

[0035] Figure 4 This is a segmentation mask diagram of the rebar sleeve joint used in the application experiment.

[0036] Figure 5 This is a diagram of the center point of the steel bar sleeve joint used in the application experiment. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0038] The YOLOv8 model, Monodepth2 model, MobileSAM model, subpixel edge detection algorithm, and method for training the YOLOv8 model used in this invention are all prior art and do not affect the understanding of this invention by those skilled in the art. The inventiveness of this invention lies in combining the YOLOv8 model, Monodepth2 model, MobileSAM model, and subpixel edge detection algorithm to detect the spacing of rebar sleeve joints, especially in multi-layer rebar meshes. This solves the problem in existing technologies where it is difficult to detect the spacing of rebars in multi-layer rebar meshes and easy to calculate the spacing of rebars in the wrong layers, leading to incorrect results, and improves the accuracy of rebar spacing detection.

[0039] Example 1

[0040] A method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model, see [link to relevant documentation]. Figure 1 This includes the following steps:

[0041] (1) Obtain the rebar image dataset:

[0042] Since there is no publicly available training set of rebar sleeve joint images online, it is necessary to build a rebar sleeve joint image acquisition platform. The rebar sleeve joint image acquisition platform in this invention is a conventional platform equipped with a corresponding camera. After the rebar sleeve joint image acquisition platform is built, the camera is used to acquire images of multi-layer rebar mesh. Images are taken at different time periods and under different light intensities, totaling 231 images.

[0043] The collected images of rebar sleeve joints were labeled with bounding boxes using the labelme software. Since the images in the dataset have only one class of labels, class_id (rebar sleeve joint), class_id was defined as 0. After labeling, a JSON file with the same name as the image file was generated. The JSON file was then converted to YOLO format, and a corresponding TXT file was generated.

[0044] Data augmentation techniques were used to expand the established dataset, such as mirroring, scaling, cropping, or color space enhancement of the images to increase the size of the dataset. After the operation was completed, 2310 images of rebar sleeve joints were obtained and distributed to the training set, validation set, and test set in a ratio of 8:1:1, resulting in 1848 images in the training set, 231 images in the validation set, and 231 images in the test set.

[0045] Specific steps to expand the dataset:

[0046] Mirror flip: Flip the labeled image horizontally or vertically;

[0047] Zooming in, zooming out, and cropping: randomly cropping the image while preserving the labeled area, or zooming in and out of the labeled area;

[0048] Color space enhancement: Adjusts the hue, saturation, and brightness of the entire image;

[0049] All 2310 images of rebar sleeve joints were adjusted to a size of 640×640.

[0050] (2) Construct the YOLOv8 model of the rebar sleeve joint:

[0051] Initialize the existing YOLOv8 model, set the number of iterations to 300, the iteration batch size to 64, and select the SGD optimizer;

[0052] The dataset of 2310 labeled images of rebar sleeve joints was input into the YOLOv8 model to train the YOLOv8 model and obtain the YOLOv8 model of rebar sleeve joints.

[0053] (3) Use the rebar sleeve joint image acquisition platform to acquire rebar images in real time and input them into the constructed rebar sleeve joint YOLOv8 model. The model will select the rebar sleeve joint in the rebar image and output the class_id and center point coordinates (x, y, y) of the rebar sleeve joint bounding box. center y center The rectangle's width and height are set up and numbered from 1 in order from left to right and top to bottom, and then stored in a txt file.

[0054] (4) Input the rebar image into the existing Monodepth2 model to generate a depth map. The model reads the txt file output by the YOLOv8 model of the rebar sleeve joint, then selects the rebar sleeve joint in the depth map and outputs the coordinates of the lower left corner of the rebar sleeve joint box (x). min y min ) and the coordinates of the upper right corner (x max y max ), calculate the average depth of each rebar sleeve joint frame and retain rebar sleeve joints with similar average depths in the txt file and their (x min y min ), (x max y max ).

[0055] (5) Input the steel bar image and the txt file obtained in step (4) into the existing MobileSAM model, segment the steel bar sleeve joint, and obtain the segmentation mask image of the steel bar sleeve joint in the same layer.

[0056] (6) Calculate the coordinates (c) of the center point of the rebar sleeve joint in the segmented mask image using existing subpixel edge detection algorithms. x c y The pixel spacing of the rebar sleeve joint is obtained. Using the rebar sleeve joint as a calibration reference, the actual length of the rebar sleeve joint is 60mm, and its pixel length in the image is L. image The pixel spacing of the rebar sleeve joint is D. image This allows us to obtain the image scaling ratio. This allows us to obtain the actual spacing of the rebar sleeve joints. .

[0057] In step (4), , , , ;in, The width of the original image. The height of the original image;

[0058] The formula for calculating the average depth of the rebar sleeve joint frame is as follows: Where A is the number of pixels, and D(x,y) is the depth value at coordinates (x,y);

[0059] Set a depth threshold to retain rebar sleeve joints with similar average depths and their (x) min y min ), (x max y max The calculation formula for retaining rebar sleeve joints with similar average depths is as follows: ;in, Let be the average depth of the i-th rebar sleeve joint frame. Let j be the average depth of the j-th rebar sleeve joint frame. This is the depth threshold.

[0060] In step (6), the existing subpixel edge detection algorithms used include morphological opening, Canny edge detection operator, Devernay algorithm, and minimum bounding rectangle algorithm. Morphological opening is used to eliminate small noise and false targets in the segmentation mask image while preserving the main shape features of the rebar sleeve joint, which can make the image edges smoother. Canny edge detection operator is used to extract the main edges of the rebar sleeve joint. Devernay algorithm is used to accurately locate the edges, making them more continuous and smooth, eliminating the jaggedness problem, and enhancing the accuracy and consistency of the edges. Minimum bounding rectangle algorithm is used to fit the edges of the rebar sleeve joint and calculate the coordinates of the center point of the rectangle.

[0061] , ;in, , , I(x,y) is the pixel value at coordinates (x,y);

[0062] The YOLOv8 model of the rebar sleeve joint was evaluated, and the evaluation results are shown in Table 1.

[0063] Table 1 Evaluation results of the YOLOv8 model for rebar sleeve joints

[0064]

[0065] Example 2

[0066] A rebar sleeve joint spacing detection system based on depth estimation and MobileSAM model, comprising:

[0067] The acquisition module is used to acquire images of the reinforcing mesh;

[0068] The central control module is used to execute the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model in Embodiment 1.

[0069] Example 3

[0070] A storage medium storing a processor-executable program for executing the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model of Embodiment 1.

[0071] Example 4

[0072] An electronic device includes a memory and a processor, the memory storing a program for executing the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model of Embodiment 1.

[0073] Application Experiment

[0074] The rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model in Example 1 is used to detect the rebar sleeve joint spacing, including the following steps:

[0075] (1) Input the image of the rebar to be detected into the constructed YOLOv8 model of the rebar sleeve joint. The model selects the rebar sleeve joint in the rebar image and outputs the class_id and center point coordinates (x, y, y) of the rebar sleeve joint bounding box. center y centerThe rectangle's width and height are set, and they are numbered from 1 to 1, in a left-to-right, top-to-bottom order. These values ​​are then stored in a txt file, as shown in Table 2. This table contains the position information for the rebar sleeve joint rectangle. The position detection results for the rebar sleeve joint can be found in [link to relevant documentation]. Figure 2 ;

[0076] Table 2 Location information of the rebar sleeve joint frame

[0077]

[0078] (2) Input the image of the rebar to be detected into the existing Monodepth2 model to generate a depth map. The model reads the txt file output by the YOLOv8 model of the rebar sleeve joint, and then selects the rebar sleeve joint in the depth map and outputs the coordinates of the lower left corner of the rebar sleeve joint box (x). min y min ) and the coordinates of the upper right corner (x max y max ), calculate the average depth of each rebar sleeve joint frame and retain the rebar sleeve joints numbered 1 and 3 in the txt file and their (x min y min ), (x max y max Table 3 shows the location information of the rebar sleeve joint boxes in the depth map. For the depth map of the rebar sleeve joints, please refer to [reference needed]. Figure 3 ;

[0079] Table 3. Location information of the rebar sleeve joint box in the depth map.

[0080]

[0081] (3) Input the image of the reinforcing bar to be detected and the txt file obtained in step (2) into the existing MobileSAM model, segment the reinforcing bar sleeve joints, and obtain the segmentation mask image of the reinforcing bar sleeve joints in the same layer. See the segmentation mask image of the reinforcing bar sleeve joints. Figure 4 ;

[0082] (4) The subpixel edge detection algorithm calculates the center point coordinates of the rebar sleeve joint numbered 1 in the segmented mask image as (170, 90) and the pixel length as 61.07 mm, and the center point coordinates of the rebar sleeve joint numbered 3 as (195, 611) and the pixel length as 60.99 mm. Therefore, the horizontal pixel spacing of the rebar sleeve joint is 25, and the vertical pixel spacing is 521. The center point of the rebar sleeve joint is shown in [reference needed]. Figure 5 Using the rebar sleeve joint as the calibration reference, and according to the specifications, the actual length of the rebar sleeve joint measured by the workers on site was 60mm, thus obtaining the image scaling ratio. The actual lateral spacing of the rebar sleeve joints was found to be 24.5 mm, and the actual lateral spacing was 510.58 mm. The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting the spacing of rebar sleeve joints based on depth estimation and the MobileSAM model, characterized in that, Includes the following steps: (1) Real-time acquisition of rebar images and input into the constructed YOLOv8 model of rebar sleeve joint. The model selects the rebar sleeve joint in the rebar image and outputs the coordinates (x, y) of the center point of the rebar sleeve joint frame. center y center ), width of the box, height of the box; (2) Input the rebar image into the Monodepth2 model to generate a depth map. The model reads the corresponding (x... center y center ), width, height, and then select the rebar sleeve joint frame in the depth map, and output the coordinates of the lower left corner of the rebar sleeve joint frame (x, y ... min y min ) and the coordinates of the upper right corner (x max y max ), calculate the average depth of each rebar sleeve joint frame and retain rebar sleeve joints with the same average depth and their (x min y min ), (x max y max ); (3) Combine the images of the reinforcing bars, the sleeve joints of the reinforcing bars with the same average depth, and their (x) min y min ), (x max y max Input the data into the MobileSAM model, segment the rebar sleeve joints, and obtain the segmentation mask image of the rebar sleeve joints in the same layer; (4) Use the sub-pixel edge detection algorithm to calculate the coordinates of the center point (c) of the rebar sleeve joint in the segmented mask image. x c y The actual spacing of the rebar sleeve joints was detected.

2. The method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model according to claim 1, characterized in that, In step (1), the construction of the YOLOv8 model of the rebar sleeve joint includes the following steps: obtaining the rebar image dataset, inputting the rebar image dataset into the YOLOv8 model, training the YOLOv8 model, and obtaining the YOLOv8 model of the rebar sleeve joint.

3. The method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model according to claim 1, characterized in that: In step (2), , , , ;in, The width of the original image. The height of the original image.

4. The method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model according to claim 1, characterized in that: In step (2), the formula for calculating the average depth of the steel bar sleeve joint frame is as follows: Where A is the number of pixels and D(x,y) is the depth value at coordinates (x,y).

5. The method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model according to claim 1, characterized in that: In step (2), a depth threshold is set to retain rebar sleeve joints with similar average depths and their (x min y min ), (x max y max The calculation formula for retaining rebar sleeve joints with similar average depths is as follows: ;in, Let be the average depth of the i-th rebar sleeve joint frame. Let j be the average depth of the j-th rebar sleeve joint frame. This is the depth threshold.

6. The method for detecting the spacing of rebar sleeve joints based on depth estimation and MobileSAM model according to claim 1, characterized in that: In step (4), , ;in, , , I(x,y) is the pixel value at coordinates (x,y).

7. A rebar sleeve joint spacing detection system based on depth estimation and MobileSAM model, characterized in that, include: The acquisition module is used to acquire images of the reinforcing mesh; The central control module is used to execute the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model as described in any one of claims 1-6.

8. A storage medium storing a processor-executable program, characterized in that: The program is used to execute the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model as described in any one of claims 1-6.

9. An electronic device, comprising a memory and a processor, characterized in that: The memory stores a program for executing the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model as described in any one of claims 1-6.

10. The application of the rebar sleeve joint spacing detection method based on depth estimation and MobileSAM model as described in claim 1 in detecting the rebar sleeve joint spacing.

Images