Computer vision-based method and system for providing rebar arrangement status

Abstract

To provide a method and system for rebar arrangement state, configured to detect spaces formed between rebars in a captured image of a rebar arrangement, based on computer vision.SOLUTION: A method includes: inputting a rebar image to a pre-trained space detection model so as to estimate mask regions corresponding to spaces between the plurality of rebars; obtaining a plurality of mask regions; classifying the plurality of mask regions into a similar group and an unsimilar group on the basis of similarity in size between the plurality of mask regions; and displaying the plurality of mask regions corresponding to at least one of the similar group or the unsimilar group in the rebar image.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a computer vision-based method and system for providing rebar alignment status.

[0002] This research was conducted with research funds from the Land, Infrastructure, Transport and Tourism Agency (KAIA) supported by the Ministry of Land, Infrastructure, Transport and Tourism (RS-2023-00251002). [Background technology]

[0003] Structural durability is a very important factor for ensuring safety during the design and construction of buildings, bridges, and other civil engineering structures, and in particular, structural durability varies greatly depending on the arrangement and distribution of reinforcing bars.

[0004] Specifically, the spacing at which the rebars are arranged effectively distributes the load of the structure, alleviating the phenomenon of stress concentration at specific points and increasing the structural durability, but if the rebars are not aligned, it can have a significant impact on the supporting capacity of the structure and pose a danger.

[0005] For this reason, various methods have been used in the past to inspect the arrangement of rebars. Typically, an inspector can directly inspect the arrangement of rebars using a visual aid.

[0006] Recently, there has been active research into methods for measuring the spacing between rebars using fixed devices, sensor-based devices, images, and computer vision. In particular, methods for detecting rebar arrangement patterns and features using deep learning models such as convolutional neural networks (CNNs) have been developed. Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention relates to a method and system for providing a rebar arrangement status that detects spaces formed between rebars in an image of the rebar arrangement.

[0008] The present invention also relates to a method and system for providing rebar arrangement status, which classifies spaces detected in an image according to size and provides the arrangement status of rebars. [Means for solving the problem]

[0009] In order to solve the above problem, the method for providing the arrangement status of reinforcing bars according to the present invention may include the steps of receiving a reinforcing bar image captured of a space in which multiple reinforcing bars are arranged, inputting the reinforcing bar image into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the multiple reinforcing bars, and obtaining multiple mask areas shown in the reinforcing bar image, classifying the multiple mask areas into similar groups and dissimilar groups based on the similarity in size between the multiple mask areas, and providing the arrangement status of the multiple reinforcing bars by displaying the multiple mask areas corresponding to at least one of the similar groups and dissimilar groups on the reinforcing bar image.

[0010] In addition, the rebar arrangement status providing system of the present invention includes a storage unit in which rebar images taken of a space in which multiple rebars are arranged are stored, and a control unit that inputs the rebar images into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the multiple rebars, and acquires multiple mask areas shown in the rebar image, and the control unit classifies the multiple mask areas into similar groups and dissimilar groups based on the similarity in size between the multiple mask areas, and displays the multiple mask areas corresponding to at least one of the similar groups and dissimilar groups on the rebar image, thereby providing the arrangement status of the multiple rebars.

[0011] Furthermore, a program stored in a computer-readable recording medium according to the present invention is a program that is executed by one or more processes in an electronic device and is stored in a computer-readable recording medium, and the program may include instructions for performing the following steps: receiving a rebar image of a space in which a plurality of rebars are arranged; inputting the rebar image into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the plurality of rebars, and acquiring a plurality of mask areas shown in the rebar image; classifying the plurality of mask areas into similar groups and dissimilar groups based on the similarity in size between the plurality of mask areas; and displaying the plurality of mask areas corresponding to at least one of the similar groups and dissimilar groups on the rebar image, thereby providing the arrangement status of the plurality of rebars. [Effects of the Invention]

[0012] According to various embodiments of the present invention, a method and system for providing rebar arrangement status detects spaces formed between rebars in an image of the rebar arrangement, classifies the detected spaces according to size, and provides the detected spaces, thereby enabling efficient verification of the structural arrangement of rebar spacing.

[0013] To this end, according to various embodiments of the present invention, the method and system for providing rebar arrangement status can more accurately and precisely detect spaces formed between rebars in the rebar image by enhancing the training rebar image in various ways.

[0014] In particular, according to various embodiments of the present invention, the method and system for providing rebar arrangement status can more accurately detect areas where errors have occurred in the structural arrangement of rebar spacing by repeating the process of classifying spaces formed between rebars according to size several times. [Brief explanation of the drawings]

[0015] [Figure 1] 1 shows an embodiment of a method for providing rebar arrangement status according to the present invention. [Figure 2] 1 illustrates an embodiment of detecting a mask region in a rebar image. [Figure 3] 1 illustrates an embodiment of detecting a mask region in a rebar image. [Figure 4] 1 illustrates one embodiment of classifying a mask region. [Figure 5] 1 illustrates one embodiment of classifying a mask region. [Figure 6] 1 shows a rebar arrangement status providing system according to the present invention. [Figure 7] 1 is a flowchart showing a method for providing a rebar arrangement status according to the present invention. [Figure 8] 1 illustrates an embodiment of detecting a mask region in a rebar image. [Figure 9] 1 illustrates one embodiment of classifying a mask region. [Figure 10] 1 illustrates one embodiment of classifying a mask region. [Figure 11] 1 illustrates one embodiment of classifying a mask region. [Figure 12] 1 illustrates one embodiment for providing rebar alignment. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, the embodiments disclosed herein will be described in detail with reference to the accompanying drawings. Regardless of the drawing numbers, identical or similar components are designated by the same reference numerals, and redundant descriptions thereof will be omitted. The suffixes "module" and "unit" used in the following description are merely used to facilitate the preparation of the specification and do not have any distinct meanings or functions. Furthermore, when describing the embodiments disclosed herein, if it is determined that a detailed description of related publicly known technology may obscure the gist of the embodiments disclosed herein, such a detailed description will be omitted. Furthermore, the accompanying drawings are merely intended to facilitate understanding of the embodiments disclosed herein, and the technical concepts disclosed herein should not be limited by the accompanying drawings. The accompanying drawings should be understood to include all modifications, equivalents, and alternatives within the concept and technical scope of the present invention.

[0017] Terms including ordinal numbers such as first, second, etc. may be used to describe various components, but the components are not limited by the terms. The terms are used only to distinguish one component from another.

[0018] When a component is referred to as being "coupled" or "connected" to another component, it should be understood that it may be directly coupled or connected to the other component, but that there may be other components between them. On the other hand, when a component is referred to as being "directly coupled" or "directly connected" to another component, it should be understood that there are no other components between them.

[0019] Unless otherwise clearly indicated in the context, singular expressions include plural expressions.

[0020] In this application, the terms "comprise" or "have" and the like are intended to specify the presence of any feature, number, step, operation, component, part, or combination thereof described herein, and are to be understood as not precluding the possible presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0021] Figure 1 shows an embodiment of a method for providing rebar alignment status according to the present invention. Figures 2 and 3 show an embodiment of detecting mask regions in a rebar image. Figures 4 and 5 show an embodiment of classifying mask regions. Figure 6 shows a system for providing rebar alignment status according to the present invention.

[0022] Referring to FIG. 1, a system 100 for providing rebar arrangement status according to the present invention can input a rebar image to a pre-trained spatial detection model and estimate a mask region for the rebar image.

[0023] Here, the reinforcing bar image is an image of a space in which a plurality of reinforcing bars are arranged, and may be an image taken by an RGB camera or a black and white camera.

[0024] In this case, the plurality of reinforcing bars may be arranged so as to intersect along different axes, i.e., the reinforcing bar image may be an image captured of a space in which the plurality of reinforcing bars are arranged so as to intersect with each other.

[0025] The mask area may also be a space formed by a plurality of rebars arranged to intersect with each other. For example, in a rebar image in which a plurality of rebars arranged along a first axis and a second axis that intersect perpendicularly are photographed, the mask area may be a rectangular space formed by two rebars arranged adjacent to each other along the first axis and two rebars arranged adjacent to each other along the second axis.

[0026] In this case, the mask area may be realized in various shapes such as a diamond or a circle depending on the angle at which the multiple reinforcing bars intersect and the shape of each of the multiple reinforcing bars.

[0027] The spatial detection model may be an artificial neural network trained to estimate mask regions in rebar images, for example, an artificial neural network implemented based on a deep neural network (DNN) and a convolutional neural network (CNN).

[0028] For this purpose, the spatial detection model may be trained by a spatial detection model training method, which may be executed by the rebar alignment status providing system 100 or by a separate training system (or device) depending on the embodiment.

[0029] In this case, the learning system can store training rebar images and augment the training rebar images to generate multiple training rebar images.

[0030] Specifically, the learning system can place photographing devices (e.g., cameras) at different positions based on the space in which multiple rebars are arranged, and store the photographed rebar images as learning rebar images.

[0031] For example, the learning system can store a plurality of learning rebar images 11 taken at positions spaced apart by different distances relative to a space in which a plurality of rebars are arranged.

[0032] As another example, the learning system can store a plurality of learning rebar images 12 taken at positions spaced a predetermined distance apart at different angles relative to the space in which the rebars are arranged.

[0033] As another example, the learning system may store a plurality of learning images of rebars taken under different lighting conditions in a space where a plurality of rebars are arranged.

[0034] As another example, the learning system may store multiple learning images of rebars, each of which is a photograph of a space in which multiple rebars having different dimensions are arranged, or a space in which multiple rebars are arranged at different intervals.

[0035] Furthermore, the learning system can edit the training rebar images and generate multiple training rebar images 13 that are augmented in different shapes.

[0036] For example, the training system can edit the training rebar image to flip it horizontally to generate a flipped training rebar image, which allows the training system to train the spatial detection model to estimate mask regions that exist in various directions and positions.

[0037] As another example, the learning system can edit the training rebar image so that it is rotated by a predetermined angle (e.g., 45 degrees or 135 degrees) to generate a rotated training rebar image, which allows the learning system to train the spatial detection model to accurately estimate the mask region for a wider range of directions in the rebar image.

[0038] As another example, the learning system may edit the training rebar image to adjust the contrast and generate a contrast-adjusted training rebar image. To this end, the learning system may calculate the maximum pixel intensity and the minimum pixel intensity for multiple pixels belonging to the training rebar image and normalize the values ​​of the multiple pixels based on the calculation results, thereby improving the visibility and definition of objects corresponding to multiple rebars in the training rebar image. This allows the learning system to reduce the effects of noise and artifacts contained in the rebar image and train a spatial detection model to subdivide the mask region.

[0039] As another example, the learning system may edit the training rebar image to adjust its saturation, thereby generating a training rebar image with adjusted saturation. To this end, the learning system may adjust the saturation of the training rebar image by adjusting the pixel intensity of a plurality of pixels belonging to the training rebar image. This allows the learning system to train a spatial detection model to more accurately distinguish between areas corresponding to rebar and mask areas in the rebar image.

[0040] Through the above configuration, the learning system can generate multiple training rebar images 13 that have been edited by at least one of inversion, rotation, contrast adjustment, and saturation adjustment for a specific training rebar image.

[0041] Furthermore, the training system can label, for each of the plurality of training rebar images 13, a training mask region corresponding to the space between the plurality of rebars (14).

[0042] Specifically, the learning system can input a bounding box corresponding to a mask area as a learning mask area for each of multiple training rebar images, and label each training rebar image with the training mask area input above.

[0043] This allows the training system to train a spatial detection model to estimate mask regions in rebar images using multiple training rebar images and training mask regions labeled for each of the multiple training rebar images (15).

[0044] Referring to FIG. 2, in one embodiment, a spatial detection model can be trained to detect mask regions in rebar images based on Deep Vision Net (DVNet), Segmentation Pyramid Pooling Network (SPPNet), and Deep CNN Network (DCNet).

[0045] In this regard, referring to FIG. 3, a manner in which a spatial detection model according to an embodiment is implemented can be seen.

[0046] Referring again to FIG. 1, the rebar arrangement status providing system 100 according to the present invention classifies a plurality of mask regions estimated for a rebar image into similar and dissimilar groups based on the similarity in size between each mask region (16), and can provide a rebar arrangement status according to the classification results (17).

[0047] Here, the similar group may be a group in which the arrangement of the rebars is uniform and the size of the spaces between the rebars is similar, i.e., the similar group may include a plurality of mask regions of similar size among the plurality of mask regions detected in the rebar image.

[0048] For example, a similar group may include a plurality of mask regions that belong to a predetermined range from an average size value for the plurality of mask regions.

[0049] As another example, a similar group may include multiple mask regions that belong to a predetermined range (e.g., the number of mask regions or a percentage of the total mask regions) from the maximum size value (or minimum size value) based on the average size value for the multiple mask regions.

[0050] The dissimilar group may be a group in which the arrangement of the rebars is uneven and the size of the spaces between the rebars is different, or may be a group that includes mask areas excluding the mask areas that belong to the similar group. In other words, the dissimilar group may include a plurality of mask areas that are detected in the rebar image but are excluding the mask areas that belong to the similar group.

[0051] Referring to FIG. 4 , in one embodiment, the rebar alignment status providing system 100 can classify multiple mask regions detected in a rebar image into a first group (e.g., A, B, C) and a second group (e.g., P, Q, R, S, T, U, X, Y, Z) based on the number of pixels in each of the multiple mask regions.

[0052] Next, the rebar arrangement status providing system 100 identifies the group (e.g., the second group) to which more mask areas belong between the first group and the second group, and can further classify the identified group into first subgroups (e.g., P, Q, R, S, T, U) and second subgroups (e.g., X, Y, Z).

[0053] As a result, the rebar arrangement status providing system 100 can identify the group (e.g., the first subgroup) to which more mask areas belong as a similar group among the first subgroup and the second subgroup, and identify the other mask areas excluding the multiple mask areas included in the similar group as dissimilar groups.

[0054] In this regard, referring to FIG. 5, it can be seen how the rebar arrangement status providing system 100 classifies a plurality of mask regions into similar groups and dissimilar groups according to an embodiment.

[0055] Meanwhile, referring to FIG. 6, a rebar arrangement status providing system 100 according to the present invention may include an input unit 110, a storage unit 120, an output unit 130, and a control unit 140.

[0056] The input unit 110 can input rebar images. To this end, the input unit 110 can be connected via a wireless or wired network to another device, system, server, etc. that has rebar images stored therein, and can receive the rebar images from the device, server, etc. Alternatively, the input unit 110 can be connected via a wireless or wired network to a photographing device such as a camera, and can input rebar images photographed by the photographing device.

[0057] The storage unit 120 may store data and commands required for the operation of the rebar arrangement status providing system 100 according to the present invention. For example, the storage unit 120 may store rebar images and information about rebar arrangement statuses generated for the rebar images (e.g., similar groups and dissimilar groups). The storage unit 120 may also store a pre-trained spatial detection model.

[0058] The output unit 130 can output at least one of the rebar image and information about the rebar arrangement state generated for the rebar image. To this end, the output unit 130 can be connected to an output device such as a display device via a wireless or wired network. Thus, the output unit 130 can output the rebar image and information about the rebar arrangement state so that the user can visually confirm them.

[0059] Meanwhile, the output unit 130 can be connected to other devices, systems, and servers via a wireless or wired network. In such a case, the output unit 130 can transmit at least one of the rebar images and information on the rebar arrangement state to the devices, systems, and servers.

[0060] The control unit 140 can control the overall operation of the rebar arrangement status providing system 100 according to the present invention. For example, the control unit 140 can input a rebar image to a spatial detection model to acquire a plurality of mask regions, classify the plurality of mask regions into similar groups and dissimilar groups, and output the rebar arrangement status according to at least one of the similar groups and dissimilar groups.

[0061] Based on the configuration of the rebar arrangement status providing system 100 described above, the rebar arrangement status providing method will be described in more detail below.

[0062] Figure 7 is a flowchart showing a method for providing rebar alignment status according to the present invention. Figure 8 shows an embodiment for detecting mask regions in a rebar image. Figures 9 to 11 show an embodiment for classifying mask regions. Figure 12 shows an embodiment for providing rebar alignment status.

[0063] Referring to FIG. 7, the rebar arrangement status providing system 100 according to the present invention receives a rebar image captured of a space in which multiple rebars are arranged (S100), inputs the rebar image into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the multiple rebars, and can acquire multiple mask areas shown in the rebar image (S200).

[0064] Specifically, the rebar arrangement status providing system 100 receives a rebar image taken of a space in which multiple rebars are arranged in a grid pattern, inputs the received rebar image into a pre-trained space detection model, and can obtain, as mask areas, each of multiple areas corresponding to the spaces between multiple rebars formed by the multiple rebars arranged in a grid pattern.

[0065] Referring to Figure 8, for example, the rebar arrangement status providing system 100 can input a rebar image into a spatial detection model, and in a rebar image 20 for a lattice-like rebar arrangement 23 formed by a plurality of rebars arranged along a first axis 21 and a plurality of rebars arranged along a second axis 22, obtain a rectangular space formed by two adjacent rebars among the plurality of rebars arranged along the first axis 21 and two adjacent rebars among the plurality of rebars arranged along the second axis 22 as a mask area 24.

[0066] Referring again to FIG. 7, the rebar arrangement status providing system 100 according to the present invention can classify a plurality of mask regions into similar groups and dissimilar groups based on the similarity in size between the plurality of mask regions (S300).

[0067] Specifically, the rebar arrangement status providing system 100 counts the number of pixels belonging to each of the multiple mask areas, and can classify the multiple mask areas into a first group and a second group based on the counted number of pixels.

[0068] Referring to FIG. 9, for example, the rebar arrangement status providing system 100 can calculate the average number of pixels for multiple mask areas 31 based on the number of pixels belonging to each of multiple mask areas 31 acquired for the rebar image 30.

[0069] As a result, the rebar arrangement status providing system 100 can identify, among the multiple mask areas 31, multiple mask areas having a larger number of pixels than the average number of pixels (e.g., the first mask area 31a and the third mask area 31c) as the first group 41 (or the second group 42), and identify multiple mask areas having a smaller number of pixels than the average number of pixels (e.g., the second mask area 31b) as the second group 42 (or the first group 41).

[0070] As another example, the rebar alignment status providing system 100 can classify (or cluster) multiple mask regions obtained for a rebar image into a first group and a second group using a K-Means clustering technique.

[0071] Furthermore, the rebar arrangement status providing system 100 counts the number of multiple mask areas classified into a first group and a second group, respectively, identifies one of the first group and the second group that has the largest number of the multiple mask areas counted above, and can classify the multiple mask areas into a first subgroup and a second subgroup based on the number of pixels for each of the multiple mask areas belonging to the group identified above.

[0072] Referring to FIG. 10, for example, when the number of mask areas included in the first group 41 is greater than the number of mask areas included in the second group, the rebar arrangement status providing system 100 can classify the mask areas (e.g., 41a, 41b, 41c, 41d) belonging to the first group 41 into a first subgroup 45 and a second subgroup 46.

[0073] Alternatively, the rebar arrangement status providing system 100 may classify the mask areas belonging to the second group into a first subgroup 45 and a second subgroup 46 if the number of mask areas included in the first group 41 is smaller than the number of mask areas included in the second group.

[0074] In this case, the rebar arrangement status providing system 100 calculates the average number of pixels for the multiple mask areas included in the first group 41 (or the second group) based on the number of pixels for each of the multiple mask areas included in the first group 41 (or the second group), and identifies, among the multiple mask areas included in the first group 41 (or the second group), multiple mask areas with a number of pixels greater than the average number of pixels (e.g., the first mask area 41a, the third mask area 41b, and the seventh mask area 41d, etc.) as the first subgroup 45 (or the second subgroup 46), and identifies multiple mask areas with a number of pixels less than the average number of pixels (e.g., the sixth mask area 41c, etc.) as the second subgroup 46 (or the first subgroup 45).

[0075] Alternatively, the rebar arrangement status providing system 100 can classify (or cluster) the multiple mask regions included in the first group 41 (or the second group) into a first subgroup 45 and a second subgroup 46 using a K-Means clustering technique.

[0076] As another example, the rebar arrangement status providing system 100 can identify the group to which one of the multiple mask areas classified into the first group and the second group, which has the largest (or smallest) number of pixels counted above, belongs, and classify the multiple mask areas belonging to the identified group into a first subgroup and a second subgroup.

[0077] Furthermore, the rebar arrangement status providing system 100 counts the number of multiple mask areas classified into a first subgroup and a second subgroup, respectively, and identifies either the first subgroup or the second subgroup that has the largest number of multiple mask areas counted above as a similar group, and can identify the other mask areas, excluding the multiple mask areas belonging to the similar group, from the multiple mask areas obtained from the rebar image, as a dissimilar group.

[0078] Referring to FIG. 11 , for example, when the number of mask regions included in the first subgroup 45 (or the second subgroup 46) is greater than the number of mask regions included in the second subgroup 46 (or the first subgroup 45), the rebar arrangement status providing system 100 may identify the first subgroup 45 (or the second subgroup 46) as a similar group 47.

[0079] In addition, the rebar arrangement state providing system 100 can identify one of the first subgroup 45 and the second subgroup 46 (e.g., the first subgroup 45) identified as a similar group 47 and the other subgroup (e.g., the second subgroup 46) as a dissimilar group.

[0080] That is, the rebar arrangement state providing system 100 can identify multiple mask areas (e.g., the first mask area 45a, the third mask area 45b, and the seventh mask area 45c) included in either the first subgroup 45 or the second subgroup 46, whichever subgroup has a larger number of mask areas (e.g., the first subgroup 45), as a similar group 47, and can identify multiple mask areas (e.g., the sixth mask area 46a and the eighth mask area 46b) included in the other subgroup (e.g., the second subgroup 46) as a dissimilar group 48.

[0081] Referring again to FIG. 7, the rebar arrangement status providing system 100 according to the present invention can provide the arrangement status of multiple rebars by displaying multiple mask areas corresponding to at least one of similar groups and dissimilar groups on the rebar image (S400).

[0082] Specifically, the rebar arrangement status providing system 100 can generate a highlight display for multiple mask areas that belong to a dissimilar group among multiple mask areas acquired from a rebar image so that they are distinguished from other mask areas.

[0083] Referring to FIG. 12, for example, the rebar arrangement status providing system 100 can display a plurality of mask areas 51 identified as dissimilar groups in a first color (e.g., red) on a rebar image 50, and can display a plurality of mask areas 52 identified as similar groups in a second color (e.g., green).

[0084] As another example, the rebar arrangement status providing system 100 can selectively display only one of a plurality of mask areas identified as dissimilar groups and a plurality of mask areas identified as similar groups on the rebar image.

[0085] Through the above configuration, the rebar arrangement status providing system 100 according to the present invention detects spaces formed between rebars in an image of the rebar arrangement, classifies the detected spaces according to size, and provides the detected spaces, thereby enabling efficient verification of the structural arrangement of rebar spacing.

[0086] To this end, the rebar arrangement status providing system 100 according to the present invention can detect spaces formed between rebars in the rebar image more accurately and precisely by enhancing the training rebar image in various ways.

[0087] In particular, the rebar arrangement status providing system 100 according to the present invention can more accurately detect areas where errors have occurred in the structural arrangement of rebar spacing by repeating the process of classifying the spaces formed between rebars according to their sizes several times.

[0088] Furthermore, the present invention described above can be realized as a program that is executed by one or more processes in an electronic device and that is stored on a computer-readable recording medium.

[0089] Therefore, the present invention can be realized as computer-readable codes or instructions stored on a program-recorded medium, i.e., various control methods according to the present invention can be provided in the form of an integrated or individual program.

[0090] On the other hand, computer-readable media include any kind of recording device that stores data that can be read by a computer system, such as a hard disk drive (HDD), a solid state disk (SSD), a silicon disk drive (SDD), a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

[0091] Furthermore, the computer-readable medium may include storage, and may be a server or cloud storage accessible by the electronic device via communication. In this case, the computer can download the program according to the present invention from the server or cloud storage via wired or wireless communication.

[0092] Furthermore, in the present invention, the computer described above is an electronic device equipped with a processor, that is, a CPU (Central Processing Unit), and the type of the computer is not particularly limited.

[0093] However, the above detailed description should not be construed as limiting in all respects, but should be considered as an example. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all modifications within the scope of the present invention are included in the scope of the present invention.

Claims

1. A method for providing rebar arrangement status executed by a processor mounted on an electronic device, comprising: receiving a reinforcing bar image captured of a space in which a plurality of reinforcing bars are arranged; A step of inputting the rebar image into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the plurality of rebars, estimating mask areas, and acquiring a plurality of mask areas shown in the rebar image; classifying the plurality of mask regions into similar groups and dissimilar groups based on the similarity in size between the plurality of mask regions; and providing the arrangement state of the plurality of rebars by displaying the plurality of mask areas corresponding to at least one of the similar group and the dissimilar group in the rebar image.

2. The step of classifying the plurality of mask regions into similar groups and dissimilar groups includes: counting the number of pixels belonging to each of the plurality of mask regions; and classifying the plurality of mask regions into a first group and a second group based on the number of the counted plurality of pixels.

3. The step of classifying the plurality of mask regions into similar groups and dissimilar groups includes: counting the number of mask regions classified into the first group and the second group; Identifying one of the first group and the second group that has the largest number of the counted mask regions; and classifying the plurality of mask regions belonging to the identified group into a first subgroup and a second subgroup based on the number of pixels for each of the plurality of mask regions belonging to the identified group.

4. The step of classifying the plurality of mask regions into similar groups and dissimilar groups includes: counting the number of mask regions classified into the first subgroup and the second subgroup; 4. The method for providing rebar arrangement status according to claim 3, further comprising the steps of: identifying one of the first and second subgroups, which has the largest number of counted mask areas, as the similar group; and identifying the other mask areas, excluding the mask areas belonging to the similar group, from the mask areas acquired from the rebar image, as the dissimilar group.

5. The step of acquiring a plurality of mask regions indicated in the rebar image includes: The method for providing a rebar arrangement state according to claim 1, wherein each of a plurality of regions corresponding to the spaces between the plurality of rebars formed by the plurality of rebars arranged in a lattice pattern is acquired as the mask region.

6. The pre-trained spatial detection model It is trained based on the spatial detection model training method, The spatial detection model training method includes: A step of storing training rebar images; a step of generating a plurality of training rebar images by enhancing the training rebar image; labeling each of the plurality of training rebar images with a training mask region corresponding to a space between the plurality of rebars; and training the spatial detection model to estimate the mask region in the rebar image using the plurality of training rebar images and the training mask region labeled in each of the plurality of training rebar images.

7. The training rebar image is The method for providing a reinforcing bar arrangement state according to claim 6, wherein the reinforcing bar images are taken by placing image capturing devices at different positions relative to the space in which the plurality of reinforcing bars are arranged.

8. The step of generating a plurality of training rebar images includes: The method for providing rebar arrangement status as described in claim 6, wherein the plurality of training rebar images are generated by editing the training rebar images by at least one of inversion, rotation, contrast adjustment, and saturation adjustment.

9. a storage unit for storing rebar images captured of a space in which a plurality of rebars are arranged; a control unit that inputs the reinforcing bar image into a space detection model that has been trained in advance to estimate mask areas corresponding to the spaces between the plurality of reinforcing bars, and acquires a plurality of mask areas shown in the reinforcing bar image; The control unit A rebar arrangement status providing system that classifies the multiple mask areas into similar groups and dissimilar groups based on the similarity in size between the multiple mask areas, and displays the multiple mask areas corresponding to at least one of the similar groups and dissimilar groups on the rebar image, thereby providing the arrangement status of the multiple rebars.

10. A program executed by one or more processes in an electronic device and stored on a computer-readable recording medium, The program receiving a reinforcing bar image captured of a space in which a plurality of reinforcing bars are arranged; inputting the rebar image into a space detection model that has been trained in advance to estimate mask regions corresponding to the spaces between the plurality of rebars, and acquiring a plurality of mask regions indicated by the rebar image; classifying the plurality of mask regions into similar groups and dissimilar groups based on the similarity in size between the plurality of mask regions; and providing an arrangement state of the plurality of rebars by displaying the plurality of mask areas corresponding to at least one of the similar group and the dissimilar group on the rebar image.

Images