A deep learning-based automatic crack identification processing method and system

Abstract

The application provides a kind of based on deep learning's fissure automatic identification processing method and system, it is related to fissure identification technical field, the method includes: obtaining the image to be identified corresponding to target rock mass;Wherein, the image to be identified is generated by the cylinder obtained by drilling the inside of target rock mass;Target rock mass is any rock mass;The image to be identified is input to the preset fissure identification model, to obtain the fissure image corresponding to the image to be identified;Fissure image includes main trunk fissure and / or branch fissure;Judge whether main trunk fissure and / or branch fissure exist in fissure image;If main trunk fissure and branch fissure exist in fissure image, remove branch fissure, to obtain the main trunk fissure corresponding to target rock mass;Realize removing branch fissure in fissure image, only keep the purpose of main trunk fissure;To accurately identify the rock mass joint fissure automatically.

Description

Technical Field

[0001] This invention relates to the field of crack identification technology, and in particular to a method and system for automatic crack identification based on deep learning. Background Technology

[0002] Rock mass fractures, as a type of structural plane, exhibit complex development directions, lengths, openings, scales, and interlacing degrees, with their spatial distribution exhibiting strong randomness, anisotropy, and concealment. Specific analysis and research on the characteristics of rock mass joints and fractures are crucial indicators for engineering rock mass classification and quality evaluation. Current technologies utilize methods such as visual identification in the field, trace inference, morphological and distribution models, fractal geometry, and probability statistics to study the morphology, attitude, trace length, and spatial distribution of joints and fractures. However, the foundation of these studies lies in the accurate identification of rock mass joints and fractures. Given the needs for rock mass structural plane analysis and evaluation, as well as engineering rock mass classification, the identification of rock mass fractures is critical. Therefore, how to automatically and accurately identify rock mass joints and fractures has become an urgent technical problem to be solved. Summary of the Invention

[0003] To address the aforementioned technical problems, the technical solution adopted by this invention is as follows: According to a first aspect of this application, a method for automatic crack identification based on deep learning is provided, the method comprising the following steps: S100, acquire the image to be identified corresponding to the target rock mass; wherein, the image to be identified is generated by drilling into the cylinder obtained from the inside of the target rock mass; the target rock mass can be any rock mass; S200, the image to be identified is input into a preset crack recognition model to obtain a crack image corresponding to the image to be identified; the crack image includes a main crack and / or a branch crack; S300, determine whether there are main fractures and / or branch fractures in the fracture image; S400, if there are main fractures and branch fractures in the fracture image, the branch fractures are removed to obtain the main fractures corresponding to the target rock mass.

[0004] According to another aspect of this application, a deep learning-based automatic crack identification and processing system is also provided. The system includes a storage medium and a processor. The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by the processor to implement the deep learning-based automatic crack identification and processing method as described in the first aspect.

[0005] The present invention has at least the following beneficial effects: This invention discloses a deep learning-based automatic fracture identification processing method, which acquires an image to be identified corresponding to a target rock mass. The image to be identified is generated by drilling into a cylinder inside the target rock mass. The target rock mass can be any rock mass. The image to be identified is input into a preset fracture identification model to obtain a fracture image corresponding to the image to be identified. The fracture image includes main fractures and / or branch fractures. The method determines whether main fractures and / or branch fractures exist in the fracture image. If both main fractures and branch fractures exist in the fracture image, the branch fractures are removed to obtain the main fractures corresponding to the target rock mass. This achieves the goal of removing branch fractures from the fracture image and retaining only the main fractures, thereby enabling automated and accurate identification of joint fractures in the rock mass. Attached Figure Description

[0006] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0007] Figure 1 A flowchart illustrating the deep learning-based automatic crack identification and processing method provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the automatic rock mass fracture identification algorithm model provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a crack image provided in an embodiment of the present invention; Figure 4 A flowchart illustrating the deep learning-based rock mass structure surface testing and analysis method provided in this embodiment of the invention; Figure 5 This is a schematic diagram of the crack structure surface provided in an embodiment of the present invention. Detailed Implementation

[0008] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0009] It should be noted that, based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Furthermore, this device and / or practice the method can be implemented using other structures and / or functionalities besides one or more of the aspects set forth herein.

[0010] Example 1: The following will refer to Figure 1 The flowchart shown is a deep learning-based automatic crack identification and processing method, which introduces a deep learning-based automatic crack identification and processing method.

[0011] The deep learning-based automatic crack identification method may include the following steps: S100, acquire the image to be identified corresponding to the target rock mass; wherein, the image to be identified is generated by drilling into the cylinder obtained inside the target rock mass; the target rock mass is any rock mass.

[0012] In this embodiment, it should be noted that the fracture surface of the rock mass usually runs through the entire interior of the rock mass. Therefore, a cylinder can be drilled at a preset location in the rock mass, and then the periphery of the cylinder can be photographed to obtain the image to be identified corresponding to the target rock mass.

[0013] S200, the image to be identified is input into a preset crack recognition model to obtain a crack image corresponding to the image to be identified; the crack image includes a main crack and / or a branch crack.

[0014] In this embodiment, the preset crack recognition model can be the Mask R-CNN algorithm model, which is obtained through the following method: like Figure 2 As shown, a deep learning-based convolutional neural network model is used, with image crack annotations as samples, to train and inversely train the model in the built Mask R-CNN model, thereby constructing an intelligent rock mass crack recognition and segmentation model to realize intelligent recognition and segmentation of observed images; and the rock mass joint cracks are skeletonized by using sine wave fitting to realize the three-dimensional reconstruction of the rock mass crack morphology.

[0015] The details are as follows: 1) Construction of Joint and Fracture Dataset Crack morphology annotation rules: Select points along the edge of the joint crack and connect them to form a closed region; export according to 1m length and 300 pixels; adopt the training set, validation set and test set with a ratio of 8:1:1. This ratio can balance the size and retain some data for verifying the performance of the model; it can better capture noise and outliers in the dataset and make the model more robust.

[0016] 2) Joint and fracture image segmentation model Joint and fracture image recognition consists of five parts: backbone feature extraction, region proposal network, region alignment, fully connected layer, and instance segmentation network.

[0017] Based on feature extraction, the Region Proposal Network (RPN) is used to identify and filter effective feature layers, and to crop proposed images for suspected targets. The size of all cropped proposal images is adjusted using ROI alignment, ultimately resulting in square images. Based on the cropped proposal images, a regression model is used to determine whether the cropped regions contain target joints or fissures. Predicted boxes are then used to extract and decode effective feature layers, thus obtaining the final predicted boxes.

[0018] The segmentation of rock mass joints and fractures consists of four parts: fracture labeling, scale clipping, binarization, and fracture cutting. The final predicted bounding box is used to crop the target image of the effective feature layer, and the cropped image is then passed to the Mask instance segmentation network for instance segmentation to obtain the segmented image. Joint and fracture feature skeletonization is based on pixel segmentation, peeling away pixels layer by layer, deleting irrelevant pixels, and retaining target pixels. Linear delineations are used to outline the morphology of rock mass joints and fractures, emphasizing important fracture features and simplifying general shapes. Complex pixels are simplified into binary images of connected single pixels to distinguish the edge and center points of the fracture skeleton. The morphological features of rock mass joints and fractures are obtained by depicting the central axis of the image.

[0019] The development of joints and fractures in rock masses is complex and interwoven, with numerous micro-fractures developing around the main joints and fractures. When characterizing the joint and fracture framework, the target fractures cannot be highlighted; therefore, branch deletion is necessary to delineate and characterize the target joints and fractures. The rock mass joint and fracture framework extraction uses an image lookup table method to refine edge points. A 3×3 sliding window is used to process the binarized image, and relevant pixels are removed from the 28 pixels using a pixel matrix.

[0020] S300, determine whether there are main fractures and / or branch fractures in the fracture image.

[0021] In this embodiment, it can be understood that the branch cracks in the crack image obtained in the above steps are useless cracks, i.e. noise, and it is necessary to first determine whether there are main cracks and / or branch cracks in the crack image.

[0022] Furthermore, step S300 may include the following steps: S310, Obtain the endpoints of the curve corresponding to each crack in the crack image to obtain an endpoint list A = (A1, A2, ..., A...). i A n ), i=1, 2,...,n; among them, A i Let be the i-th endpoint in the crack image, and n be the number of endpoints in the crack image.

[0023] like Figure 3 As shown, each crack in the crack image corresponds to a curve with an endpoint, and each endpoint can be obtained to obtain A.

[0024] S311, Based on A, determine the coordinates of each endpoint in A to obtain the endpoint coordinate list ZA = (ZA1, ZA2, ..., ZA...). i , ..., ZA n ); where ZA i For A i Corresponding coordinates; ZA i =(ZA i，x ZA i，y ); ZA i，x Let ZA be the X-axis coordinate of the i-th endpoint in the crack image. i，y Let be the Y-axis coordinate of the i-th endpoint in the crack image.

[0025] In this embodiment, a planar coordinate system corresponding to the crack image can be constructed, and then the coordinates of each endpoint can be obtained to obtain ZA.

[0026] S312, if n=2, and ZA 1，y =ZA 2，y Then, obtain the horizontal distance LA between A1 and A2 as LA = |ZA 1，x -ZA 2，x |

[0027] In this embodiment, it can be understood that the crack image corresponds to the cylinder, which can be understood as the plane corresponding to the side of the cylinder. Therefore, the two ends of the crack image correspond to the cylinder being connected together; if n=2, and ZA 1，y =ZA 2，y This means that A1 and A2 may correspond to the same point on the cylinder. However, if the main crack is not a through-crack on the cylinder, then even if n=2 and ZA 1，y =ZA 2，y A1 and A2 may not correspond to the same point; therefore, it is also necessary to obtain the horizontal distance LA between A1 and A2 for further judgment.

[0028] S313, if LA=D, then it is determined that there is a main fracture and no branch fracture in the fracture image; where D is the circumference of the cylinder obtained from the inside of the target rock mass.

[0029] In this embodiment, when n=2 and ZA 1，y =ZA 2，y Under the premise that LA=D, it can be determined that A1 and A2 correspond to the same point on the cylinder; therefore, it is determined that there is a main crack and no branch cracks in the crack image.

[0030] Furthermore, after step S313, the method may further include the following steps: S314, the cracks corresponding to the curves between A1 and A2 are identified as the main cracks.

[0031] Furthermore, after step S311, the method may further include the following steps: S320, if n>2, then obtain the first preset value Q=1 and the second preset value M=2.

[0032] In this embodiment, if n > 2, it indicates that there may be branch images in the crack image.

[0033] S321. If Q < n-1, proceed to S322; otherwise, proceed to S325.

[0034] S322, if M < n, then proceed to S323; otherwise, proceed to S324.

[0035] S323, if ZA Q，y =ZA M，y Then ZA Q and ZA M Determine a set of intermediate endpoints; otherwise, obtain M=M+1 and proceed to S322. S324, obtain Q=Q+1, M=Q+1; proceed to S322.

[0036] S325, obtain the intermediate endpoints for each group to obtain a list of intermediate endpoint groups B = (B1, B2, ..., B...). j B m ), j=1,2,…,m; where, B j Let m be the number of intermediate endpoints determined in the j-th group.

[0037] In this embodiment, through the above cyclic steps, several sets of intermediate points with the same Y-axis coordinate can be determined, and the curve between two intermediate points with the same Y-axis coordinate may correspond to the main fracture.

[0038] S326, obtain the horizontal distance between any two intermediate endpoints in each group of intermediate endpoints in B, to obtain a horizontal distance list LB = (LB1, LB2, ..., LB...). j , ..., LB m ); among which, LB j For B j The horizontal distance between the two intermediate endpoints.

[0039] S327, iterate through LB, if |LB j If -D|<ΔL, then it is determined that there are main cracks and branch cracks in the crack image; where ΔL is a preset horizontal distance difference threshold.

[0040] In this embodiment, there may be errors in image recognition and coordinate construction. Therefore, ΔL is set to increase the redundancy of the judgment and improve the accuracy of the judgment.

[0041] Furthermore, after step S327, the method may further include the following steps: S328, B j The cracks corresponding to the curves between the two intermediate endpoints are identified as the main cracks.

[0042] Understandably, once the main fracture is identified, the remaining fractures are considered branch fractures, thus achieving the purpose of identifying the main fracture and branch fractures.

[0043] S400, if there are main fractures and branch fractures in the fracture image, the branch fractures are removed to obtain the main fractures corresponding to the target rock mass.

[0044] Furthermore, after step S400, the method may further include the following steps: S500: If only the main crack exists in the crack image, the preset branch crack removal operation will not be performed on the crack image.

[0045] In one exemplary embodiment, a deep learning-based automatic crack identification and processing system is provided. The system includes a storage medium and a processor. The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by the processor to implement the deep learning-based automatic crack identification and processing method as described in the above embodiment.

[0046] This embodiment presents a deep learning-based automatic fracture identification method that acquires an image to be identified corresponding to a target rock mass. The image to be identified is generated by drilling into a cylinder inside the target rock mass. The target rock mass can be any rock mass. The image to be identified is input into a preset fracture identification model to obtain a fracture image corresponding to the image to be identified. The fracture image includes main fractures and / or branch fractures. The method determines whether main fractures and / or branch fractures exist in the fracture image. If both main fractures and branch fractures exist in the fracture image, the branch fractures are removed to obtain the main fractures corresponding to the target rock mass. This achieves the goal of removing branch fractures from the fracture image and retaining only the main fractures, thereby enabling automated and accurate identification of joint fractures in the rock mass.

[0047] Furthermore, this embodiment introduces the Mask R-CNN deep learning convolutional neural network algorithm model, which realizes intelligent identification of rock mass joints and fractures through joint and fracture feature capture and skeleton feature extraction. This replaces the manual measurement of joint and fracture occurrence in the original core logging process, improves drilling field operation efficiency, and saves human resource costs.

[0048] Furthermore, this method modularizes and automates professional experience, breaking down professional knowledge barriers and digitizing professional work. This significantly promotes the development of cutting-edge technologies and provides valuable reference experience for the development of digital technologies in various professions. It enables the rapid and efficient completion of similar tasks. In nuclear engineering project surveys, its advantages, such as intuitive test results and comprehensive evaluation of engineering rock masses, have repeatedly earned praise from peer experts and clients. It saves on-site technical personnel's time, allowing them to focus their energy on overall project management, which is beneficial for project implementation. It also reduces labor costs and the workload of technical personnel, and improves the working environment.

[0049] The method and system in this embodiment can be widely used for identifying rock joints and fissures in underground surrounding rock, analyzing rock mass quality, evaluating rock mass integrity, and simultaneously classifying underground surrounding rock and evaluating basic rock mass quality. It can also be applied to rock mass quality analysis in water conservancy and hydropower, bridge and tunnel projects. Example

[0050] To analyze the fracture structure surface corresponding to the main fracture, the corresponding fracture structure surface can be generated based on the main fracture obtained in Example 1, and then analyzed to provide, for example... Figure 4 The method shown may include the following steps: H100, obtain the initial fracture structure surface WA corresponding to the main fracture of the rock mass; where WA is generated based on the fracture image corresponding to the main fracture of the rock mass.

[0051] In this embodiment, as Figure 5As shown, after obtaining the main fracture data, it can be input into existing simulation software to generate the initial fracture structure surface corresponding to the main fracture.

[0052] H200, obtain the tilt angle θ of WA.

[0053] In this embodiment, it can be understood that the main crack is not necessarily horizontal. Therefore, there is a certain tilt angle between the generated initial crack structure surface and the horizontal plane.

[0054] H300, obtain the highest point G1 and lowest point G2 of the curve corresponding to the main fracture in the fracture image; where G1 = (G 1，x G 1，y G2 = (G) 2，x G 2，y ); G 1，x and G 1，y These are the X-axis and Y-axis coordinates of G1, respectively. 2，x and G 2，y These are the X-axis and Y-axis coordinates of G2, respectively.

[0055] In this embodiment, the line shape corresponding to the main fracture is a curve. Therefore, there are a highest point and a lowest point, and the highest point G1 and the lowest point G2 of the curve corresponding to the main fracture in the fracture image can be obtained.

[0056] H400, based on G1 and G2, determine the inclination angle γ of the main fracture γ = arctan(|G 1，y -G 2，y | / |G 1，x -G 2，x |).

[0057] H500, if |θ-γ|<GY, then WA does not need adjustment; otherwise, obtain each preset designated point on the curve corresponding to the main fracture to obtain a list of designated points E=(E1, E2, ..., E...). p , ..., E q ), p=1,2,…,q; where, E p = is the p-th preset designated point on the curve corresponding to the main fracture, q is the number of preset designated points on the curve corresponding to the main fracture, and GY is the preset tilt angle difference threshold.

[0058] In this embodiment, ideally θ and γ should be the same. However, the main crack is not a standard sine curve. Therefore, there may be a deviation between the tilt angle of the structural surface obtained by the software and the tilt angle of the main crack. If the deviation is within the preset tilt angle difference threshold, it can be determined that WA is normal. Otherwise, each preset designated point on the curve corresponding to the main crack is obtained to adjust WA. The designated points are points that are evenly set.

[0059] H600, based on E and WA, adjust WA to obtain the target fracture structure surface, and analyze the target fracture structure surface.

[0060] Furthermore, step H600 may include the following steps: H610, retrieve each specified point in E located above WA to obtain a first list of specified points C = (C1, C2, ..., C...). r C s ), r=1,2,…,s; where, C r s represents the r-th specified point located above WA, and s represents the number of specified points located above WA.

[0061] H620, retrieve each specified point in E located below WA to obtain a second list of specified points F = (F1, F2, ..., F...). u F v ), u=1,2,…,v; where, F u v represents the u-th specified point located below WA, and v represents the number of specified points located below WA.

[0062] In this embodiment, it can be understood that WA is a plane, while the main fracture is a curve, and the designated point on the main fracture will be located above and / or below WA.

[0063] H630, based on C and F, adjust WA to obtain the target fracture structure surface.

[0064] Furthermore, step H630 may include the following steps: H631, if α1≤s / v≤α2, then obtain the perpendicular distance from each specified point in C to WA, to obtain the first perpendicular distance list LC=(LC1, LC2, ..., LC3) for C. r , ..., LC s ); where LC r C r Find the perpendicular distance from WA; and obtain the perpendicular distance from WA to each specified point in F to obtain the second perpendicular distance list LF = (LF1, LF2, ..., LF3) for F. u , ..., LF v ); LF u For F u The vertical distance from WA; α1 is the first preset weight, and α2 is the second preset weight.

[0065] In this embodiment, α1 < 1; α2 > 1; if α1 ≤ s / v ≤ α2, it means that the number of specified points above and below WA is basically the same. However, it cannot be determined whether the vertical distance from WA is uniform. Therefore, the vertical distance from WA of each specified point in C and the vertical distance from WA of each specified point in F are obtained for further judgment.

[0066] H632, based on LC and LF, determine the average distance LC' corresponding to LC and the average distance LF' corresponding to LF.

[0067] H633, if |LC'-LF'|≤LU, then WA will not be adjusted; where LU is the preset average distance difference threshold.

[0068] In this embodiment, if |LC'-LF'|≤LU, it means that the position of WA is appropriate, and WA will not be adjusted.

[0069] Furthermore, after step H633, the method further includes the following steps: H634, if LC'-LF'>LU, then move WA downwards by a distance LQ1 along the direction perpendicular to WA; where LQ1=(LC'-LF') / 2.

[0070] In this embodiment, if LC'-LF'>LU, it means that the position of WA is too high. Therefore, WA is moved down by a distance LQ1 along the direction of the vertical line of WA so that WA can better characterize the structural surface corresponding to the main fracture.

[0071] H635, if LF'-LC'>LU, then move WA upwards by a distance LQ2 along the direction perpendicular to WA; where LQ2=(LF'-LC') / 2.

[0072] In this embodiment, if LF'-LC'>LU, it means that the position of WA is too low. Therefore, WA is moved upward by a distance of LQ2 along the direction of the vertical line of WA so that WA can better characterize the structural surface corresponding to the main fracture.

[0073] Furthermore, step H630 also includes the following steps: H61, if s / v < α1, then move WA downwards several times along the direction perpendicular to WA. Each time it moves, determine whether s / v satisfies α1 ≤ s / v ≤ α2; the distance moved each time is the same.

[0074] In this embodiment, if s / v < α1, it means that the main crack at position WA is relatively high. Most of the midpoints of the main crack are located below WA. WA cannot well represent the actual structural surface corresponding to the main crack. Therefore, WA can be moved downwards several times along the direction of the vertical line of WA. Each time it is moved, it is determined whether s / v satisfies α1≤s / v≤α2.

[0075] H62, if s / v satisfies α1≤s / v≤α2, then proceed to H631.

[0076] If s / v satisfies α1≤s / v≤α2, WA can be adjusted using the methods in steps H631-H635.

[0077] Furthermore, step H630 also includes the following steps: H63, if s / v>α2, then move WA upwards several times along the direction perpendicular to WA. Each time it moves, determine whether s / v satisfies α1≤s / v≤α2; the distance moved each time is the same.

[0078] In this embodiment, if s / v > α2, it means that the main crack at position WA is relatively low. Most of the midpoints of the main crack are located above WA. WA cannot well represent the actual structural surface corresponding to the main crack. Therefore, WA can be moved upward several times along the direction of the vertical line of WA. Each time it is moved, it is determined whether s / v satisfies α1≤s / v≤α2.

[0079] H64, if s / v satisfies α1≤s / v≤α2, then proceed to H631.

[0080] If s / v satisfies α1≤s / v≤α2, WA can be adjusted using the methods in steps H631-H635.

[0081] In one exemplary embodiment, a deep learning-based rock mass structure surface testing and analysis system is provided. The system includes a storage medium and a processor. The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by the processor to implement the deep learning-based rock mass structure surface testing and analysis method as described in the above embodiment.

[0082] The deep learning-based rock mass structure surface testing and analysis method of this embodiment obtains the initial fracture structure surface WA corresponding to the main fracture of the rock mass; obtains the tilt angle θ of WA; obtains the highest point G1 and the lowest point G2 of the curve corresponding to the main fracture in the fracture image; determines the tilt angle γ of the main fracture based on G1 and G2; if |θ-γ|<GY, it is determined that WA does not need to be adjusted; otherwise, it obtains each preset designated point on the curve corresponding to the main fracture to obtain a list of designated points E; adjusts WA based on E and WA to obtain the target fracture structure surface; the target fracture structure surface obtained by the method of this invention is more consistent with the actual structure surface corresponding to the main fracture, making the generated main fracture structure surface more accurate.

[0083] A deep learning-based method for testing and analyzing rock mass structural surfaces can achieve 3D reconstruction of borehole rock mass structural surfaces through image preprocessing, dataset construction, model training and parameter tuning, fracture mask prediction, fracture image segmentation and skeletonization, and fracture morphology fitting. This allows for further analysis of the joint number and average spacing, the degree of connection between major structural surfaces, and the type of major structural surfaces within the rock mass structural surface development. It is primarily applied to rock mass quality and integrity assessment. Replacing traditional manual identification and analysis, it modularizes and intelligently integrates professional experience, breaks down professional knowledge barriers, and digitizes professional work, saving human resources, shortening project timelines, and reducing project production costs.

[0084] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0085] Embodiments of the present invention also provide a non-transitory computer-readable storage medium that can be disposed in an electronic device to store at least one instruction or at least one program related to implementing a method in the method embodiments, wherein the at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided in the above embodiments.

[0086] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0087] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0088] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0089] Program code for performing the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0090] Embodiments of the present invention also provide an electronic device, including a processor and the aforementioned non-transitory computer-readable storage medium.

[0091] The electronic device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments in this application.

[0092] Electronic devices are manifested in the form of general-purpose computing devices. Components of an electronic device may include, but are not limited to: at least one processor, at least one memory, and a bus connecting different system components (including memory and processor).

[0093] The memory stores program code that can be executed by the processor, causing the processor to perform the steps in the various embodiments described in this specification.

[0094] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read-only memory (ROM).

[0095] The memory may also include programs / utilities having a set (at least one) of program modules, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0096] A bus can represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus that uses any of the various bus structures.

[0097] The electronic device can also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via input / output (I / O) interfaces. Furthermore, the electronic device can communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter. The network adapter communicates with other modules of the electronic device via a bus. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0098] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0099] Embodiments of the present invention also provide a computer program product including program code, which, when the program product is run on an electronic device, causes the electronic device to perform the steps of the methods described above in various exemplary embodiments of the present invention.

[0100] While specific embodiments of the invention have been described in detail by way of examples, those skilled in the art should understand that the examples are for illustrative purposes only and are not intended to limit the scope of the invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the invention.

Claims

1. A method for automatic crack identification and processing based on deep learning, characterized in that, The method includes the following steps: S100, acquire the image to be identified corresponding to the target rock mass; wherein, the image to be identified is generated by drilling into the cylinder obtained from the inside of the target rock mass; the target rock mass can be any rock mass; S200, the image to be identified is input into a preset crack recognition model to obtain a crack image corresponding to the image to be identified; the crack image includes a main crack and / or a branch crack; S300, determine whether there are main fractures and / or branch fractures in the fracture image; S400, if there are main fractures and branch fractures in the fracture image, remove the branch fractures to obtain the main fractures corresponding to the target rock mass; Step S300 includes the following steps: S310, Obtain the endpoints of the curve corresponding to each crack in the crack image to obtain an endpoint list A = (A1, A2, ..., A...). i A n ), i=1, 2,...,n; among them, A i Let be the i-th endpoint in the crack image, and n be the number of endpoints in the crack image; S311, Based on A, determine the coordinates of each endpoint in A to obtain the endpoint coordinate list ZA = (ZA1, ZA2, ..., ZA...). i , ..., ZA n ); where ZA i For A i Corresponding coordinates; ZA i =(ZA i，x ZA i，y ); ZA i，x Let ZA be the X-axis coordinate of the i-th endpoint in the crack image. i，y Let be the Y-axis coordinate of the i-th endpoint in the crack image; S312, if n=2, and ZA 1，y =ZA 2，y Then, obtain the horizontal distance LA between A1 and A2 as LA = |ZA 1，x -ZA 2，x |; S313, if LA=D, then it is determined that there is a main fracture and no branch fracture in the fracture image; where D is the circumference of the cylinder obtained from the inside of the target rock mass. S314, the cracks corresponding to the curves between A1 and A2 are identified as the main cracks; Following step S311, the method further includes the following steps: S320, if n>2, then obtain the first preset value Q=1 and the second preset value M=2; S321, if Q < n-1, then proceed to S322; otherwise, proceed to S325; S322, if M < n, then proceed to S323; otherwise, proceed to S324; S323, if ZA Q，y =ZA M，y Then ZA Q and ZA M Determine a set of intermediate endpoints; otherwise, obtain M=M+1 and proceed to S322. S324, obtain Q=Q+1, M=Q+1; proceed to S322; S325, obtain the intermediate endpoints for each group to obtain a list of intermediate endpoint groups B = (B1, B2, ..., B...). j B m ), j=1,2,…,m; where, B j Let m be the number of intermediate endpoints determined in the j-th group; S326, obtain the horizontal distance between any two intermediate endpoints in each group of intermediate endpoints in B, to obtain a horizontal distance list LB = (LB1, LB2, ..., LB...). j , ..., LB m ); among which, LB j For B j The horizontal distance between the two intermediate endpoints; S327, iterate through LB, if |LB j If -D|<ΔL, then it is determined that there are main fractures and branch fractures in the fracture image; where ΔL is a preset horizontal distance difference threshold. S328, B j The cracks corresponding to the curves between the two intermediate endpoints are identified as the main cracks.

2. The method for automatic crack identification based on deep learning according to claim 1, characterized in that, The preset crack recognition model includes the Mask R-CNN algorithm model.

3. The method for automatic crack identification and processing based on deep learning according to claim 1, characterized in that, Following step S400, the method further includes the following steps: S500: If only the main crack exists in the crack image, the preset branch crack removal operation will not be performed on the crack image.

4. A deep learning-based automatic crack identification and processing system, characterized in that, The system includes a storage medium and a processor; wherein the storage medium stores at least one instruction or at least one program segment, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement the deep learning-based automatic crack identification processing method as described in any one of claims 1-3.

Images