Image recognition method and device for over-excavation and under-excavation of tunnel face after blasting by drilling and blasting method

Abstract

The present invention belongs to the field of tunnel engineering, and specifically relates to an image recognition method and device for over-excavation and under-excavation of a tunnel face after blasting using a drill and blast method. The specific technical scheme is as follows: based on an established tunnel face image database after blasting, based on the principle of a semantic segmentation algorithm in image recognition, the tunnel face over-excavation and under-excavation area and background area of ​​the tunnel face image are identified; based on an established tunnel face target image database, based on the principle of a semantic segmentation algorithm in image recognition, the target area and background area of ​​the tunnel face target image are identified; through the pixel length of the target and its position in the tunnel coordinate system, the coordinates of the tunnel face over-excavation and under-excavation area in the tunnel coordinate system are obtained, and finally compared and analyzed with the tunnel blasting design contour line to obtain the over-excavation and under-excavation of the tunnel face after the tunnel blasting. This method reduces the error in the calculation results of tunnel over-excavation and under-excavation, reduces the impact on construction time and construction cost, and overcomes the technical problem of low accuracy in tunnel over-excavation and under-excavation calculation in traditional methods.

Description

Technical Field

[0001] The invention belongs to the field of tunnel engineering, and in particular relates to an image recognition method and device for over-excavation and under-excavation of a tunnel face after blasting of a tunnel using a drill and blast method. Background Art

[0002] In the process of tunnel excavation, drilling and blasting is one of the main methods of excavation, but the disadvantage of drilling and blasting is that once the blasting design is improper, it will inevitably cause a large amount of over-excavation and under-excavation of the tunnel. Tunnel over-excavation and under-excavation are divided into over-excavation and under-excavation. Over-excavation refers to the actual excavation contour line of the tunnel being greater than the designed excavation contour line of the tunnel, and under-excavation refers to the actual excavation contour line of the tunnel being less than the designed excavation contour line of the tunnel. Tunnel over-excavation and under-excavation will not only affect the construction progress, but also affect the quality of the project, and even increase the construction cost. Therefore, a convenient, fast and reliable method for evaluating over-excavation and under-excavation of the tunnel face after tunnel blasting is very important in the process of tunnel excavation.

[0003] At present, most of the calculation methods for tunnel overbreak and underbreak still remain at the level of manual on-site measurement of tunnel overbreak and underbreak. This method has the following problems: manual testing is time-consuming and laborious, and each test will cause large errors in the measurement results due to human factors; each section cannot be measured infinitely, and can only be measured at a few local locations; it is troublesome to draw pictures after the test, and the whole process is cumbersome.

[0004] Therefore, if a convenient, fast and accurate image recognition method for over-excavation and under-excavation of the tunnel face after blasting can be provided, it will have excellent industrial application value. Summary of the invention

[0005] In order to solve the above technical problems, the object of the present invention is to provide an image recognition method and device for over-excavation and under-excavation of a tunnel face after blasting using a drill and blast method.

[0006] To achieve the above-mentioned purpose of the invention, the technical scheme adopted by the present invention is: according to the established tunnel face image database after blasting, based on the principle of semantic segmentation algorithm in image recognition, the tunnel face over-excavation area and background area of ​​the tunnel face image are identified; according to the established tunnel face target image database, based on the principle of semantic segmentation algorithm in image recognition, the target area and background area of ​​the tunnel face target image are identified; through the pixel length of the target and the position in the tunnel coordinate system, the coordinates of the tunnel face over-excavation area in the tunnel coordinate system are obtained, and finally compared and analyzed with the tunnel blasting design contour line to obtain the over-excavation and under-excavation amount of the tunnel face after the tunnel blasting.

[0007] Preferably: the image recognition method for over-excavation and under-excavation of the tunnel face after blasting by drilling and blasting method comprises the following steps:

[0008] Step 101: obtaining a large number of tunnel face image samples after blasting, establishing a tunnel face image sample database, performing deep learning on the tunnel face images, and establishing an over-break and under-break area recognition model for tunnel face images after tunnel blasting;

[0009] Step 102: obtaining a large number of target image samples of the tunnel face after blasting, establishing a target image sample database on the tunnel face, performing deep learning on the target image of the tunnel face, and establishing a target area recognition model on the tunnel face;

[0010] Step 103: Based on the over-break and under-break area identification model and the target area identification model, identify the over-break and under-break amount of the tunnel face after blasting.

[0011] Preferably: the step 101 includes:

[0012] Step 1011: by marking the original image of the tunnel face after blasting, the original image of the tunnel face after blasting is segmented into the tunnel face over-excavation area and the background area, and a tunnel face image sample database after blasting is established;

[0013] Step 1012: Based on the established post-blasting tunnel face image sample database, deep learning is performed on the tunnel face image samples to establish an over-break and under-break area recognition model for the tunnel face image after tunnel blasting.

[0014] Preferably: the step 102 includes,

[0015] Step 1021: by marking the original image of the tunnel face after blasting, the original image of the tunnel face after blasting is segmented into a target area and a background area, and a target image sample database on the tunnel face is established;

[0016] Step 1022: Based on the established database of target image samples on the tunnel face after blasting, deep learning is performed on the target image samples on the tunnel face to establish a target area recognition model on the tunnel face of the drill and blast method.

[0017] Preferably: the step 103 includes:

[0018] Step 1031: obtaining the over-underbreak area and the target area on the tunnel face image according to the tunnel face over-underbreak area recognition model and the target area recognition model;

[0019] Step 1032: based on the over-break and under-break areas and the target area on the tunnel face image, obtaining coordinates of the over-break and under-break areas and the target area in the tunnel coordinate system;

[0020] Step 1033: Based on the coordinates of the over-break and under-break areas and the target area in the tunnel coordinate system, by comparing and analyzing with the tunnel blasting design contour line, the over-break and under-break amount of the tunnel face after blasting is obtained.

[0021] Correspondingly: an image recognition device for over-excavation and under-excavation of a tunnel face after blasting by a drilling and blasting method, comprising an image acquisition module, a first construction module, a second construction module, and an analysis and processing module;

[0022] The image acquisition module acquires the tunnel face image after the tunnel blasting, and transmits the tunnel face image to the construction module 1 and the construction module 2;

[0023] The construction module 1 constructs an over-break and under-break area recognition model of the tunnel face image after blasting of the drilling and blasting method tunnel according to the established tunnel face image sample database after blasting;

[0024] The second construction module constructs a target area recognition model on the tunnel face image of the drilling and blasting method according to the established target image sample database on the tunnel face;

[0025] The analysis and processing module analyzes the over-excavation and under-excavation of the tunnel face after blasting based on the recognition result of the tunnel face image of the first construction module and the recognition result of the tunnel face target image of the second construction module.

[0026] Preferably: the construction module 1 includes a tunnel face image data module and a tunnel face over-excavation and under-excavation identification module;

[0027] The tunnel face image data module receives the tunnel face image transmitted by the image acquisition module, establishes a tunnel face image database, and transmits the tunnel face image to the tunnel face over-excavation and under-excavation area identification module;

[0028] The tunnel face over-excavation and under-excavation identification module receives a tunnel face image, identifies a tunnel face area and a background area in the tunnel face image, performs deep learning on the tunnel face image, and establishes a tunnel face over-excavation and under-excavation area identification model.

[0029] Preferably: the second construction module includes a face target image data module and a face target area recognition module;

[0030] The palm face target image data module receives the palm face image transmitted by the image acquisition module, establishes a palm face target image database, and transmits the palm face target image to the palm face target area recognition module;

[0031] The palm face target area recognition module receives the palm face target image, recognizes the target area and the background area in the palm face target image, performs deep learning on the palm face target image, and establishes a palm face target area recognition model.

[0032] Accordingly: an electronic device, comprising:

[0033] one or more processors;

[0034] A storage device for storing one or more programs;

[0035] When the one or more programs are executed by the one or more processors, the one or more processors implement an image recognition method for over-excavation and under-excavation of a tunnel face after blasting of a tunnel using a drill and blast method.

[0036] Correspondingly: a computer-readable medium storing a computer program, wherein when the computer program is executed by a processor, an image recognition method for over-excavation and under-excavation of a tunnel face after blasting using a drill and blast method is implemented.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] According to the established post-blasting tunnel face image database, based on the principle of semantic segmentation algorithm in image recognition, the tunnel face over-break and under-break area and background area of ​​the tunnel face image are identified; according to the established tunnel face target image database, based on the principle of semantic segmentation algorithm in image recognition, the target area and background area of ​​the tunnel face target image are identified; through the pixel length of the target and the position in the tunnel coordinate system, the coordinates of the tunnel face over-break area in the tunnel coordinate system are obtained, and finally compared and analyzed with the tunnel blasting design contour line to obtain the over-break and under-break amount of the tunnel face after the tunnel blasting. With the continuous excavation of the tunnel, the revealed tunnel face over-break and under-break data are continuously updated. Through image recognition, the over-break and under-break amount of the tunnel face after each cycle of blasting can be accurately calculated, so as to optimize the blasting design and reduce the project cost. This method reduces the error of the tunnel over-break and under-break calculation results, reduces the impact on construction time and construction cost, and overcomes the technical problem of low accuracy of tunnel over-break and under-break calculation in traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of the method for identifying over-excavation and under-excavation images of the tunnel face after blasting by the drilling and blasting method of the present invention;

[0040] Figure 2 It is a schematic diagram of the marking of the over-break and under-break area of ​​the tunnel face after the tunnel blasting by the drilling and blasting method of the present invention (the over-break and under-break area in the figure is the area surrounded by dots);

[0041] Figure 3 It is a schematic diagram of marking the target area on the tunnel face image of the drilling and blasting method of the present invention;

[0042] Figure 4 It is a schematic diagram of identifying the over-break and under-break area of ​​the tunnel face after blasting of the tunnel by the drilling and blasting method of the present invention;

[0043] Figure 5 It is a schematic diagram of target area recognition on a tunnel face image of the drilling and blasting method of the present invention;

[0044] Figure 6It is a schematic diagram of comparative analysis of the over-excavation and under-excavation identification contour line of the tunnel face after tunnel blasting and the tunnel blasting design contour line;

[0045] Figure 7 It is a schematic diagram of the basic structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art.

[0047] The invention discloses an image recognition method for over- and under-break of a tunnel face after blasting using a drill and blast method. The core idea is: according to an established tunnel face image database after blasting, based on the principle of a semantic segmentation algorithm in image recognition, the tunnel face over- and under-break area and background area of ​​the tunnel face image are identified; according to an established tunnel face target image database, based on the principle of a semantic segmentation algorithm in image recognition, the target area and background area of ​​the tunnel face target image are identified; through the pixel length of the target and the position in the tunnel coordinate system, the coordinates of the over- and under-break area of ​​the tunnel face in the tunnel coordinate system are obtained, and finally compared and analyzed with the tunnel blasting design contour line to obtain the over- and under-break amount of the tunnel face after the tunnel blasting.

[0048] like Figure 1 As shown, an image recognition method for over-excavation and under-excavation of a tunnel face after blasting using a drill and blast method comprises the following steps:

[0049] Step 101: Obtain a large number of tunnel face image samples after blasting, establish a tunnel face image sample database after blasting, perform deep learning on the tunnel face image samples based on the principle of image semantic segmentation, and establish an over-excavation and under-excavation area recognition model for tunnel face images after blasting using the drilling and blasting method.

[0050] Specifically, step 101 includes the following steps:

[0051] Step 1011: by marking the original image of the tunnel face after blasting, the original image of the tunnel face after blasting is divided into the tunnel face over-break area and the background area. The schematic diagram of marking the tunnel face over-break after blasting of the drilling and blasting method tunnel is as follows: Figure 2 As shown, based on a large number of tunnel face images after blasting, a post-blasting tunnel face image sample database is established.

[0052] Step 1012: Based on the established post-blasting tunnel face image sample database and the principle of semantic segmentation algorithm in image recognition, an over-undercut region recognition model of the tunnel face image after tunnel blasting by the drilling and blasting method is established. In the deep learning process, the tunnel face image is used as an input variable, and the tunnel face over-undercut region and background region are used as output variables to establish an over-undercut region recognition model of the tunnel face image. The over-undercut region recognition model of the tunnel face image can recognize the tunnel face over-undercut region and background region in the tunnel face image.

[0053] Specifically, an implementation method of establishing an over-break and under-break recognition model for a tunnel face image after blasting using a drill and blast method is described, wherein step 1012 includes the following steps:

[0054] Step 1012.1: Building a deep learning network framework for tunnel face image semantic segmentation;

[0055] Step 1012.2: Optimize the network parameters of step 1012.1, and use the tunnel face image database of step 1012 to train the network weight parameters of step 1012.1 until the total loss no longer decreases, and obtain the model parameters with the minimum test error. The optimization step 1012.1 can be one or more of changing the model learning rate, changing the backbone feature extraction network, changing the loss function calculation method, etc.

[0056] Step 1012.3: Based on the network framework of step 1012.1 and the final network training parameters of step 1012.2, a model for identifying over-excavation and under-excavation areas of the tunnel face image after tunnel blasting is established.

[0057] Step 102: For the target image of the face after blasting, a corresponding target image sample library on the face is established, and deep learning is performed on the target image samples of the face based on the principle of semantic segmentation algorithm in image recognition, and a target area recognition model on the face of a drilling and blasting tunnel is established. It should be noted that the target can be a brightly colored pole of a fixed length, a brightly colored flat plate of a fixed area, or several brightly colored dots of a fixed diameter. That is, the target should be distinguishable from the face area by color, and the target is an object that exists on the target image of the face after blasting. By using the ratio of the actual length of the target and the pixel of the target image, a proportional conversion is performed between the face target coordinates and the pixel coordinate system.

[0058] Specifically, step 102 includes the following steps:

[0059] Step 1021: By marking the original image of the tunnel face target after blasting, the original image of the tunnel face after blasting is divided into a target area and a background area. The schematic diagram of the marking of the target on the tunnel face image of the drilling and blasting method is as follows: Figure 3 As shown, based on a large number of target images on the tunnel face, a target image sample database on the tunnel face is established.

[0060] Step 1022: Based on the established database of target image samples on the tunnel face after blasting and based on the principle of semantic segmentation algorithm in image recognition, a target area recognition model on the tunnel face of the drilling and blasting method is established. In the deep learning process, the tunnel face target image is used as the input variable, and the target area and background area are used as the output variables to establish the target area recognition model on the tunnel face.

[0061] Specifically, an implementation method of establishing a target area recognition model on a tunnel face of a drill and blast method is as follows: Step 1012 includes the following steps:

[0062] Step 1022.1: Building a deep learning network framework for semantic segmentation of target images on the face;

[0063] Step 1022.2: Optimize the network parameters of step 1022.1, and use the tunnel face target image database of step 1022 to train the network weight parameters of step 1022.1 until the total loss no longer decreases, and obtain the model parameters with the minimum test error. The optimization step 1022.1 can be one or more of changing the model learning rate, changing the backbone feature extraction network, changing the loss function calculation method, etc.

[0064] Step 1022.3: Based on the network framework of step 1022.1 and the final network training parameters of step 1022.2, a target area recognition model on the tunnel face is established.

[0065] Step 103: Based on the over-break and under-break area recognition model of the tunnel face image after blasting by the drill and blast method and the target area recognition model on the tunnel face image of the drill and blast method, the over-break and under-break amount of the tunnel face after blasting is identified through the pixel length of the target and the position in the tunnel coordinate system.

[0066] Specifically, step 103 includes the following steps:

[0067] Step 1031: for the tunnel face image after blasting of the same tunnel, the over-break and under-break areas and the target area on the tunnel face image are obtained according to the tunnel face over-break and under-break area recognition model and the target area recognition model.

[0068] Step 1032: Based on the over-excavation area and the target area on the face image, the coordinates of the over-excavation area and the target area in the tunnel coordinate system are obtained through the pixel length of the target and the position in the tunnel coordinate system. The first is the positioning problem. The tunnel coordinate system always takes the face as the coordinate system plane. The target (taking a pole as an example) is vertically placed at the intersection of the tunnel centerline, the face, and the step plane. The intersection is taken as the origin, the horizontal direction is the X-axis direction, and the vertical direction is the Y-axis direction. The second is the coordinate system conversion problem. According to the conversion ratio between the pixel length of the target and the actual length, the image coordinate system and the tunnel coordinate system are converted proportionally with the origin as the reference.

[0069] Step 1033: Based on the coordinates of the over-break and under-break area and the target area in the tunnel coordinate system, the over-break and under-break amount of the tunnel face after blasting is obtained by comparing and analyzing with the tunnel blasting design contour line. Here, the ratio of the actual length of the target and the target image pixel is used to perform proportional conversion between the target coordinates of the face and the pixel coordinate system, so that the over-break and under-break area, the target area, and the blasting design contour line can be set on the same image, and then the over-break and under-break value in the image coordinate system is calculated, and the over-break and under-break value in the image coordinate system is converted into the over-break and under-break value in the tunnel coordinate system according to the conversion ratio, that is, the actual over-break and under-break amount.

[0070] In a specific implementation scheme, the schematic diagram of identifying the over-excavation and under-excavation areas of the tunnel face after blasting by the drilling and blasting method is as follows: Figure 4 As shown in the figure; the target area recognition diagram on the drilling and blasting tunnel face image is as follows Figure 5 As shown in the figure; the schematic diagram of the comparison analysis between the identification contour line of the over-excavation and under-excavation area of ​​the tunnel face after tunnel blasting and the tunnel blasting design contour line is shown in the figure Figure 6 shown.

[0071] As the tunnel is continuously excavated, the data on the over-excavation and under-excavation of the face is continuously updated. Through image recognition, the over-excavation and under-excavation of the face after each cycle of blasting can be accurately calculated, thereby optimizing the blasting design and reducing the project cost. This method reduces the error in the calculation results of tunnel over-excavation and under-excavation, reduces the impact on construction time and construction cost, and overcomes the technical problem of low accuracy of tunnel over-excavation and under-excavation calculation in traditional methods. In addition, this method has created a new image recognition mode for over-excavation and under-excavation of the face after blasting in the drilling and blasting method of tunnels that is "intelligent, efficient, less manpower, and unmanned", which is highly intelligent and not possessed by traditional methods.

[0072] As an implementation of the above-mentioned method, the present application also provides an image recognition device for over-excavation and under-excavation of the tunnel face after blasting by drilling and blasting method. Figure 1Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices. The present application further discloses an image recognition device for over-excavation and under-excavation of the tunnel face after blasting using the drilling and blasting method, comprising an image acquisition module, a construction module one, a construction module two, and an analysis and processing module. During operation, the image acquisition module acquires an image of the tunnel face and outputs the image of the tunnel face to the analysis and processing module; the construction module one establishes an over-excavation and under-excavation area recognition model of the tunnel face image after blasting using the drilling and blasting method; the construction module two establishes a target area recognition model on the tunnel face; the analysis and processing module obtains the over-excavation and under-excavation area recognition model of the construction module one and the target area recognition model of the construction module two, processes and analyzes the tunnel face image, and identifies the over-excavation and under-excavation of the tunnel face after blasting.

[0073] Furthermore, the image acquisition module acquires the tunnel face image after the tunnel blasting, and transmits the tunnel face image to the construction module 1 and the construction module 2. Specifically, the image acquisition module transmits the tunnel face image to the tunnel face image data module and the tunnel face target image data module.

[0074] Furthermore, the construction module 1 is used to construct an over- and under-excavation area recognition model for the tunnel face image after blasting by the drilling and blasting method based on the established tunnel face image sample database after blasting and based on the principles of deep learning and image semantic segmentation.

[0075] In some optional embodiments, construction module one is specifically used to: distinguish between over- and under-excavation areas and background areas of the tunnel face by annotating the original image, and establish a sample database of tunnel face images after blasting based on a large number of tunnel face images after blasting; and construct an over- and under-excavation area recognition model for tunnel face images after blasting by drilling and blasting based on the established sample database of tunnel face images after blasting and based on the principles of deep learning and semantic segmentation.

[0076] Specifically, the construction module 1 includes a face image data module and a face over-excavation and under-excavation area recognition module. The face image data module receives the face image transmitted by the image acquisition module, establishes a face image database, and transmits the face image to the face over-excavation and under-excavation area recognition module. The face over-excavation and under-excavation area recognition module receives the face image, identifies the face area and the background area in the face image based on the principle of image semantic segmentation, and performs deep learning model training on a large number of face images based on a large number of face images to establish a face over-excavation and under-excavation area recognition model.

[0077] Furthermore, the second construction module is used to construct a target area recognition model on the drilling and blasting tunnel face image based on the established target image sample database on the face and based on the principles of deep learning and semantic segmentation.

[0078] In some optional embodiments, construction module two is specifically used for: distinguishing the target area from the background area by annotating the original image, and establishing a target image sample database on the tunnel face based on a large number of target images on the tunnel face; and constructing a target area recognition model on the tunnel face image of the drilling and blasting method based on the established target image sample database on the tunnel face and based on the principles of deep learning and image semantic segmentation.

[0079] Specifically, the second construction module includes a palm face target image data module and a palm face target area recognition module. The palm face target image data module receives the palm face image transmitted by the image acquisition module, establishes a palm face target image database, and transmits the palm face target image to the palm face target area recognition module. The palm face target area recognition module receives the palm face target image, identifies the target area and background area in the palm face target image based on the principle of image semantic segmentation, performs deep learning on a large number of palm face target images based on a large number of palm face target images, and establishes a palm face target area recognition model.

[0080] Furthermore, the analysis and processing module obtains the tunnel face over-excavation and under-excavation area recognition model of construction module one and the tunnel face target area recognition model of construction module two, and based on the recognition results of the tunnel face image by the tunnel face over-excavation and under-excavation area recognition module and the recognition results of the tunnel face target area recognition module, analyzes the over-excavation and under-excavation of the tunnel face after blasting through the pixel length of the target and the position in the tunnel coordinate system.

[0081] In some optional embodiments, the analysis and processing module is specifically used to: obtain the over-excavation area and the target area on the image according to the over-excavation and under-excavation recognition model of the face and the target area recognition model; based on the over-excavation and under-excavation areas on the image and the target area, obtain the coordinates of the over-excavation and under-excavation areas in the tunnel coordinate system through the pixel length of the target and the position in the tunnel coordinate system; based on the coordinates of the over-excavation and under-excavation areas in the tunnel coordinate system, obtain the over-excavation and under-excavation amount of the tunnel face after blasting by comparing and analyzing with the tunnel blasting design contour line.

[0082] refer to Figure 7 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0083] like Figure 7 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage device 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the electronic device 900 are also stored. The processing device 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0084] Typically, the following devices may be connected to the I / O interface 905: an input device 906 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 907 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 908 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 909. The communication device 909 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0085] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 909, or installed from a storage device 908, or installed from a ROM 902. When the computer program is executed by the processing device 901, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0086] It should be noted that the computer-readable medium of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0087] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0088] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0089] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: constructs an over-excavation and under-excavation area recognition model of the tunnel face image after blasting of the drilling and blasting method based on the established tunnel face image sample database and the principles of deep learning and semantic segmentation; constructs a target area recognition model on the tunnel face image of the drilling and blasting method based on the established target image sample database on the tunnel face and the principles of deep learning and semantic segmentation; based on the over-excavation area recognition model of the tunnel face image after blasting of the drilling and blasting method tunnel and the target area recognition model on the tunnel face image of the drilling and blasting method tunnel, the over-excavation and under-excavation amount of the tunnel face after blasting is identified through the pixel length of the target and the position in the tunnel coordinate system.

[0090] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0091] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0092] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of the unit does not limit the unit itself in some cases. For example, the construction module 1 may also be described as "a unit for constructing an over-excavation and under-excavation area recognition model of the tunnel face image after blasting by the drilling and blasting method based on the established tunnel face image sample database and the principles of deep learning and semantic segmentation."

[0093] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0094] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0095] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0096] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0097] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0098] The embodiments described above are only descriptions of the preferred modes of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations, modifications, and substitutions made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. An image recognition method for over-excavation and under-excavation of a tunnel face after blasting using a drill and blast method, characterized in that: According to the established post-blasting tunnel face image database, based on the principle of semantic segmentation algorithm in image recognition, the tunnel face over-excavation and under-excavation areas and background areas of the tunnel face image are identified; According to the established tunnel face target image database and based on the principle of semantic segmentation algorithm in image recognition, the target area and background area of ​​the tunnel face target image are identified; the coordinates of the over-excavation and under-excavation areas of the tunnel face in the tunnel coordinate system are obtained through the pixel length of the target and its position in the tunnel coordinate system. Finally, they are compared and analyzed with the tunnel blasting design contour line to obtain the over-excavation and under-excavation of the tunnel face after tunnel blasting.

2. The image recognition method for over-excavation and under-excavation of a tunnel face after blasting by drill and blast method according to claim 1 is characterized by: The following steps are included: Step 101: obtaining a large number of tunnel face image samples after blasting, establishing a tunnel face image sample database, performing deep learning on the tunnel face images, and establishing a tunnel face over-break and under-break area recognition model for tunnel face images after tunnel blasting; Step 102: obtaining a large number of tunnel face target image samples after blasting, establishing a tunnel face target image sample database, performing deep learning on the tunnel face target images, and establishing a tunnel face target area recognition model; Step 103: Based on the tunnel face over-break and under-break area identification model and the tunnel face target area identification model, identify the over-break and under-break amount of the tunnel face after blasting.

3. The image recognition method for over-excavation and under-excavation of a tunnel face after blasting by drill and blast method according to claim 2 is characterized by: The step 101 includes: Step 1011: by marking the original image of the tunnel face after blasting, the original image of the tunnel face after blasting is segmented into the tunnel face over-excavation area and the background area, and a tunnel face image sample database after blasting is established; Step 1012: Based on the established post-blasting tunnel face image sample database, deep learning is performed on the tunnel face image samples to establish a tunnel face over- and under-break area recognition model for the tunnel face image after tunnel blasting.

4. The image recognition method for over-excavation and under-excavation of a tunnel face after blasting by drill and blast method according to claim 2 is characterized by: The step 102 includes, Step 1021: by marking the original image of the tunnel face after blasting, the original image of the tunnel face after blasting is segmented into a target area and a background area, and a target image sample database on the tunnel face is established; Step 1022: Based on the established target image sample database on the tunnel face after blasting, deep learning is performed on the tunnel face target image samples to establish a tunnel face target area recognition model.

5. The image recognition method for over-break and under-break of tunnel face after blasting by drill and blast method according to claim 2 is characterized by: The step 103 includes: Step 1031: obtaining the over-underbreak area and the target area on the tunnel face image according to the tunnel face over-underbreak area recognition model and the tunnel face target area recognition model; Step 1032: based on the over-break and under-break areas and the target area on the tunnel face image, obtaining coordinates of the over-break and under-break areas and the target area in the tunnel coordinate system; Step 1033: Based on the coordinates of the over-break and under-break areas and the target area in the tunnel coordinate system, by comparing and analyzing with the tunnel blasting design contour line, the over-break and under-break amount of the tunnel face after blasting is obtained.

6. An image recognition device for over-excavation and under-excavation of tunnel face after blasting by drilling and blasting method, characterized in that: It includes an image acquisition module, a construction module 1, a construction module 2, and an analysis and processing module; The image acquisition module acquires the tunnel face image after the tunnel blasting, and transmits the tunnel face image to the construction module 1 and the construction module 2; The construction module 1 constructs a tunnel face over-break and under-break area recognition model of tunnel face images after blasting of a drilling and blasting tunnel according to an established tunnel face image sample database after blasting; The second construction module constructs a target area recognition model for the tunnel face according to the established target image sample database on the tunnel face; The analysis and processing module analyzes the over-excavation and under-excavation of the tunnel face after blasting based on the recognition result of the tunnel face image of the first construction module and the recognition result of the tunnel face target image of the second construction module.

7. The image recognition device for over-excavation and under-excavation of a tunnel face after blasting by the drill and blast method according to claim 6 is characterized by: The construction module 1 includes a tunnel face image data module and a tunnel face over-excavation and under-excavation area identification module; The tunnel face image data module receives the tunnel face image transmitted by the image acquisition module, establishes a tunnel face image database, and transmits the tunnel face image to the tunnel face over-excavation and under-excavation area identification module; The tunnel face over- and under-break identification module receives a tunnel face image, identifies a tunnel face over- and under-break area and a background area in the tunnel face image, performs deep learning on the tunnel face image, and establishes a tunnel face over- and under-break area identification model.

8. The image recognition device for over-excavation and under-excavation of a tunnel face after blasting by drill and blast method according to claim 6, characterized in that: The second construction module includes a face target image data module and a face target area recognition module; The palm face target image data module receives the palm face image transmitted by the image acquisition module, establishes a palm face target image database, and transmits the palm face target image to the palm face target area recognition module; The palm face target area recognition module receives the palm face target image, recognizes the target area and the background area in the palm face target image, performs deep learning on the palm face target image, and establishes a palm face target area recognition model.

9. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

10. A computer readable medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images