Three-dimensional modeling method and device for boiler wall, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a three-dimensional modeling method and device for a boiler wall, electronic equipment and a storage medium, and the method comprises the steps: obtaining three-dimensional voxel features of each target block structure of the boiler wall based on a scanning point cloud of the boiler wall, the scanning point cloud comprising multiple scanning point cloud data of each target block structure; obtaining a multi-scale mask feature pattern of each target block structure based on the corresponding three-dimensional voxel feature, wherein the multi-scale mask feature pattern comprises a plurality of mask feature patterns with different scales; and constructing a three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the live-action image of each target block structure. According to the embodiment of the invention, the accuracy of three-dimensional modeling of the boiler wall can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a three-dimensional modeling method, device, electronic equipment and storage medium for a boiler wall. Background Art

[0002] Boiler walls are mostly confined spaces, such as boiler walls in power plants, which are relatively dark inside and have complex structures, making it difficult to reconstruct a three-dimensional model of them. The three-dimensional modeling methods in the prior art include three-dimensional modeling methods based on scanning point clouds and three-dimensional modeling methods based on images. Due to the complex structure of the boiler wall, the method based on scanning point clouds will cause problems such as regional scanning blind spots, and the method based on image modeling will not be able to capture all the details of various structures. Therefore, the accuracy of the prior art method for three-dimensional modeling of boiler walls is poor. Summary of the invention

[0003] The embodiments of the present invention provide a three-dimensional modeling method, device, electronic device and storage medium for a boiler wall, which can improve the accuracy of three-dimensional modeling of the boiler wall.

[0004] In a first aspect, an embodiment of the present invention provides a three-dimensional modeling method, device, electronic device and storage medium for a boiler wall, including:

[0005] Acquire three-dimensional voxel features of each target block structure of the boiler wall based on a scanned point cloud of the boiler wall, wherein the scanned point cloud includes multiple scanned point cloud data of each target block structure;

[0006] Acquire a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, wherein the multi-scale mask feature map includes a plurality of mask feature maps of different scales; and

[0007] The three-dimensional model of the boiler wall is constructed based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure.

[0008] In a second aspect, an embodiment of the present invention provides a three-dimensional modeling device for a boiler wall, comprising:

[0009] A three-dimensional voxel feature acquisition module, used to acquire three-dimensional voxel features of each target block structure of the boiler wall based on a scan point cloud of the boiler wall, wherein the scan point cloud includes multiple scan point cloud data of each target block structure;

[0010] A multi-scale mask feature map acquisition module, used for acquiring a multi-scale mask feature map of each target block structure based on corresponding three-dimensional voxel features, wherein the multi-scale mask feature map includes a plurality of mask feature maps of different scales; and

[0011] The three-dimensional model acquisition module is used to construct the three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure.

[0012] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a three-dimensional modeling method of a boiler wall as described in any one of the embodiments of the present invention is implemented.

[0013] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the three-dimensional modeling method of a boiler wall as described in any one of the embodiments of the present invention.

[0014] An embodiment of the present invention provides a three-dimensional modeling method, device, electronic device and storage medium for boiler wall. The three-dimensional voxel features of each target block structure of the boiler wall are obtained based on the scanning point cloud of the boiler wall, and a multi-scale mask feature map of each target block structure is further obtained based on the corresponding three-dimensional voxel features. Then, the three-dimensional model of the boiler wall is constructed based on the multi-scale mask feature map and the real-scene image of each target block structure. The three-dimensional modeling of the boiler wall can be performed by combining the structural information of the boiler wall contained in the scanning point cloud and the structural information of the boiler wall in the real-scene image, thereby reducing the omission of structural details of the boiler wall when the three-dimensional modeling is performed only based on the scanning point cloud or only based on the image, and improving the accuracy of the three-dimensional modeling of the boiler wall. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a schematic flow chart of a three-dimensional modeling method of a boiler wall provided by an embodiment of the present invention;

[0017] Figure 2 is another schematic flow chart of the three-dimensional modeling method of a boiler wall provided by an embodiment of the present invention;

[0018] Figure 3 is a structural schematic diagram of a three-dimensional modeling device for a boiler wall provided in an embodiment of the present invention;

[0019] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] Figure 1 A flow chart of a three-dimensional modeling method for a boiler wall provided in an embodiment of the present invention is provided. This embodiment is applicable to the scenario of three-dimensional modeling of a boiler water-cooled wall of a power plant. The method can be executed by a three-dimensional modeling device for a boiler wall provided in an embodiment of the present invention, and the device can be implemented in software and / or hardware. In a specific embodiment, the device can be integrated into an electronic device, such as a computer, a server, etc. The following embodiments will be described by taking the device integrated into an electronic device as an example. Reference Figure 1 , the method may specifically include the following steps:

[0023] Step 101, based on the scanning point cloud of the boiler wall, obtain the three-dimensional voxel features of each target block structure of the boiler wall, the scanning point cloud includes multiple scanning point cloud data of each target block structure. This step can facilitate obtaining a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features.

[0024] Preferably, the scanning point cloud is a laser scanning point cloud obtained by scanning with a three-dimensional laser scanner. Specifically, the scanning point cloud can also be point cloud data obtained by scanning with other types of scanners such as a photographic scanner.

[0025] Specifically, the above-mentioned multiple scanning point cloud data can be understood as scanning point cloud data obtained by scanning each target block structure multiple times at different positions using a scanner.

[0026] Specifically, the target block structure may include all block structures obtained by blocking the boiler wall, or may include some block structures among all block structures, for example, a relatively characteristic structure.

[0027] Optionally, the target block structures of the boiler wall include: water-cooled wall tubes, manhole doors, burners, fire viewing holes, furnace wall surfaces and surface coke deposits.

[0028] It is understandable that, due to the complex structure of the boiler wall, it is inevitable that some structures will be missed when scanning and acquiring the scanning point cloud of the boiler wall. Therefore, the target block structure may only include some block structures among all the block structures of the boiler wall.

[0029] Optionally, the process of acquiring the three-dimensional voxel features of each target block structure of the boiler wall based on the scanning point cloud of the boiler wall includes:

[0030] Based on the corresponding target block structure, semantic annotation is performed on multiple scanned point cloud data of each target block structure to obtain multiple groups of annotated point clouds of each target block structure; and three-dimensional voxel features are extracted from each corresponding group of annotated point cloud data to obtain multiple three-dimensional voxel features of each target block structure.

[0031] Optionally, the process of semantically annotating the multiple scanning point cloud data of each target block structure based on the corresponding target block structure includes: determining the semantic label of the above-mentioned corresponding target block structure, and establishing an association relationship between the multiple scanning point cloud data corresponding to each target block structure and the corresponding semantic label.

[0032] Step 102, based on the corresponding three-dimensional voxel features, a multi-scale mask feature map of each target block structure is obtained, and the multi-scale mask feature map includes a plurality of mask feature maps of different scales. This step can be helpful for constructing a more accurate three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block and the real scene image.

[0033] Specifically, the above-mentioned mask feature map can be understood as a three-dimensional model of the corresponding structure reconstructed based on the corresponding three-dimensional voxel features.

[0034] Optionally, the process of obtaining a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features includes:

[0035] Multiple three-dimensional voxel features of each target block structure are aligned and fused to obtain fused voxel features of each target block structure; based on multiple preset scales, the corresponding fused voxel features are three-dimensionally reconstructed to obtain a multi-scale mask feature map of each target block structure.

[0036] Optionally, the process of aligning the multiple three-dimensional voxel features of each target block structure includes performing scale alignment processing and texture alignment processing on the multiple three-dimensional voxel features.

[0037] It is understandable that when scanning the same block structure of the boiler wall at different positions, the obtained scanning point cloud often includes different structural information. Fusion of the three-dimensional voxel features corresponding to multiple scanning power supply data can help obtain richer structural information; however, because the distances and angles of scanning the same block structure at different positions are different, the scales and textures of the corresponding three-dimensional voxel features are also different. Therefore, feature fusion can only be performed after the scales and textures of the above-mentioned multiple three-dimensional voxel features are aligned.

[0038] Optionally, the process of performing three-dimensional reconstruction on the corresponding fused voxel features based on multiple preset scales to obtain a multi-scale mask feature map of each target block structure includes: performing three-dimensional reconstruction on the above fused voxel features using a three-dimensional convolutional neural network (3D-CNN).

[0039] Step 103, constructing a three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure. This step is based on step 101 and step 102, by obtaining the three-dimensional voxel features of each target block structure of the boiler wall based on the scanned point cloud of the boiler wall, and further obtaining the multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, and then constructing the three-dimensional model of the boiler wall based on the multi-scale mask feature map and the real scene image of each target block structure, and can combine the structural information of the boiler wall contained in the scanned point cloud and the structural information of the boiler wall in the real scene image to perform three-dimensional modeling of the boiler wall, thereby reducing the omission of structural details of the boiler wall when performing three-dimensional modeling based only on the scanned point cloud or only on the image, and improving the accuracy of three-dimensional modeling of the boiler wall.

[0040] Specifically, the real scene images of the above-mentioned target block structures can be understood as images obtained by photographing the target block structures and including image information of the target block structures.

[0041] Specifically, the real scene images of the above-mentioned target block structures may include one real scene image or multiple real scene images.

[0042] Optionally, the process of constructing the three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure includes:

[0043] One or more mask feature maps having a scale similar to that of the corresponding real scene image are obtained from the multi-scale mask feature maps of each target block structure, and a corresponding target block structure is constructed based on the one or more mask feature maps and the corresponding real scene image.

[0044] Optionally, the process of constructing the three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure includes:

[0045] Based on the multi-scale mask feature map of each target block structure, the three-dimensional structural position of the corresponding real scene image is determined, and based on the three-dimensional structural position of the real scene image of each target block structure, the three-dimensional structural position of other real scene images of the boiler wall is determined, and based on the three-dimensional structural position of the real scene image of each target block structure and the three-dimensional structural position of other real scene images of the boiler wall, a three-dimensional model of the boiler wall is constructed.

[0046] The following further describes a three-dimensional modeling method for a boiler wall provided by another embodiment of the present invention. Figure 2 As shown, the following steps may be included:

[0047] Step 201 : acquiring three-dimensional voxel features of each target block structure of the boiler wall based on a scanned point cloud of the boiler wall, wherein the scanned point cloud includes multiple scanned point cloud data of each target block structure.

[0048] Step 202: acquiring a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, wherein the multi-scale mask feature map includes a plurality of mask feature maps of different scales.

[0049] Step 203 , extracting features of the real scene images of each target block structure to obtain features of the real scene images of each target block structure.

[0050] Optionally, the process of extracting features from the real scene images of each target block structure to obtain features of the real scene images of each target block structure includes:

[0051] By utilizing the feature extraction layer of the pre-trained multi-layer convolutional neural network, image feature extraction is performed on the real scene images of the above-mentioned each target block structure to obtain the real scene image features of the above-mentioned each target block structure.

[0052] Step 204 , fusing the real scene image features of each target block structure with the multi-scale mask feature map to obtain the three-dimensional voxel semantic features of each target block structure.

[0053] Optionally, the process of fusing the real scene image features of each target block structure with the multi-scale mask feature map to obtain the three-dimensional voxel semantic features of each target block structure includes:

[0054] The feature fusion layer of the multi-layer convolutional neural network is used to fuse the real scene image features and multi-scale mask feature maps of each target block structure.

[0055] Specifically, the multi-layer convolutional neural network can be constructed based on the network structure of Visual Geometry Group (VGG) and Alex Net.

[0056] Optionally, the above-mentioned process of fusing the real-scene image features and the multi-scale mask feature map of each target block structure to obtain the three-dimensional voxel semantic features of each target block structure includes: determining the three-dimensional coordinate information of the corresponding real-scene image features based on the multi-scale mask feature map; and aligning and fusing the multi-scale mask feature map with the corresponding real-scene image features based on the three-dimensional coordinate information of the corresponding real-scene image features.

[0057] Specifically, the process of aligning the multi-scale mask feature map with the corresponding real scene image feature includes: aligning and adjusting the contours and textures of the multi-scale mask feature map and the corresponding real scene image feature.

[0058] Optionally, the above-mentioned aligning and fusing the multi-scale mask feature map with the corresponding real scene image features includes:

[0059] For each target block structure, the real scene image features of the current target block structure are aligned and fused with multiple corresponding mask feature maps of similar scale to obtain multiple three-dimensional voxel semantic features of the current target block structure.

[0060] Step 205 : constructing a three-dimensional model of the boiler wall based on the three-dimensional voxel semantic features of each target block structure.

[0061] Optionally, the process of constructing the three-dimensional model of the boiler wall based on the three-dimensional voxel semantic features of each target block structure includes:

[0062] A plurality of prediction modeling results of the current target block structure are respectively constructed based on the corresponding three-dimensional semantic features of the current target block structure; the three-dimensional model of the current target block structure is obtained by taking the intersection of the multiple prediction modeling results; and a three-dimensional model of the boiler wall is constructed based on the three-dimensional model of the current target block structure.

[0063] The embodiments of the present invention can help fully combine the structural information of the boiler wall in the point cloud and the structural information of the boiler wall in the image, so as to better improve the modeling accuracy of the boiler wall.

[0064] Figure 3 1 is a structural diagram of a three-dimensional modeling device for a boiler wall provided in an embodiment of the present invention, and the device is suitable for executing a three-dimensional modeling method for a boiler wall provided in an embodiment of the present invention. Figure 3 As shown, the device may specifically include:

[0065] The three-dimensional voxel feature acquisition module 301 is used to acquire the three-dimensional voxel features of each target block structure of the boiler wall based on the scanned point cloud of the boiler wall, and the scanned point cloud includes multiple scanned point cloud data of each target block structure. This module can facilitate the acquisition of a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features.

[0066] Optionally, the target block structures of the boiler wall include: water-cooled wall tubes, manhole doors, burners, fire viewing holes, furnace wall surfaces and surface coke deposits.

[0067] Optionally, the above-mentioned three-dimensional voxel feature acquisition module 301 can be specifically used to perform semantic annotation on multiple scanned point cloud data of each target block structure based on the corresponding target block structure to obtain multiple groups of annotated point clouds of each target block structure; and to perform three-dimensional voxel feature extraction on each corresponding group of annotated point cloud data to obtain multiple three-dimensional voxel features of each target block structure.

[0068] The multi-scale mask feature map acquisition module 302 is used to acquire the multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, and the multi-scale mask feature map includes multiple mask feature maps of different scales. This module can be used to construct a three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block and the real scene image.

[0069] Optionally, the above-mentioned multi-scale mask feature map acquisition module 302 can be specifically used to align and fuse multiple three-dimensional voxel features of each target block structure to obtain the fused voxel features of each target block structure; perform three-dimensional reconstruction on the corresponding fused voxel features based on multiple preset scales to obtain a multi-scale mask feature map of each target block structure.

[0070] The three-dimensional model acquisition module 303 is used to construct a three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure. This module can be combined with module 301 and module 302, and obtain the three-dimensional voxel features of each target block structure of the boiler wall based on the scanned point cloud of the boiler wall, and further obtain the multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, and then construct the three-dimensional model of the boiler wall based on the multi-scale mask feature map and the real scene image of each target block structure, and can combine the structural information of the boiler wall contained in the scanned point cloud and the structural information of the boiler wall in the real scene image to perform three-dimensional modeling of the boiler wall, which can reduce the omission of structural details of the boiler wall when performing three-dimensional modeling based only on the scanned point cloud or only on the image, and improve the accuracy of three-dimensional modeling of the boiler wall.

[0071] Optionally, the three-dimensional modeling device of the boiler wall provided in the embodiment of the present invention further includes: an image feature extraction module, which is used to extract features of the real scene images of each target block structure to obtain real scene image features of each target block structure.

[0072] Optionally, the above-mentioned three-dimensional model acquisition module 303 can be specifically used to fuse the real-scene image features of each target block structure and the multi-scale mask feature map to obtain the three-dimensional voxel semantic features of each target block structure; and to construct a three-dimensional model of the boiler wall based on the three-dimensional voxel semantic features of each target block structure.

[0073] Optionally, the above-mentioned three-dimensional model acquisition module 303 can be specifically used to determine the three-dimensional coordinate information of the corresponding real-scene image features based on the multi-scale mask feature map; and align and fuse the multi-scale mask feature map with the corresponding real-scene image features based on the three-dimensional coordinate information of the corresponding real-scene image features.

[0074] Optionally, the three-dimensional model acquisition module 303 can be specifically used to align and fuse the real-scene image features of the current target block structure with multiple corresponding mask feature maps of similar scales for each target block structure to obtain multiple three-dimensional voxel semantic features of the current target block structure.

[0075] Optionally, the above-mentioned three-dimensional model acquisition module 303 can be specifically used to construct multiple predictive modeling results of the current target block structure based on multiple three-dimensional semantic features of the current target block structure; obtain the three-dimensional model of the current target block structure by taking the intersection of multiple predictive modeling results; and construct a three-dimensional model of the boiler wall based on the three-dimensional model of the current target block structure.

[0076] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the functional modules described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0077] An embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the three-dimensional modeling method of the boiler wall provided in any of the above embodiments is implemented.

[0078] An embodiment of the present invention further provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the three-dimensional modeling method of the boiler wall provided in any of the above embodiments is implemented.

[0079] The embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the three-dimensional modeling method of a boiler wall as described in any one of the embodiments of the present invention.

[0080] Reference below Figure 4 , which shows a schematic diagram of the structure of a computer system 400 of an electronic device suitable for implementing an embodiment of the present invention. Figure 4 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 invention.

[0081] like Figure 4 As shown, the computer system 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage part 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the system 400 are also stored. The CPU 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0082] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed, so that a computer program read therefrom is installed into the storage section 408 as needed.

[0083] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, the above-mentioned functions defined in the system of the present invention are executed.

[0084] It should be noted that the computer-readable medium shown in the present invention 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 invention, 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 invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. 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. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, 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 appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0085] 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 invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two 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 box in the block diagram or flow chart, and the combination of the boxes in the block diagram 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.

[0086] The modules and / or units involved in the embodiments of the present invention may be implemented in software or hardware. The modules and / or units described may also be arranged in a processor, for example, it may be described as follows: a processor includes a 3D voxel feature acquisition module, a multi-scale mask feature map acquisition module, and a 3D model acquisition module. The names of these modules do not, in some cases, limit the modules themselves.

[0087] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining three-dimensional voxel features of each target block structure of the boiler wall based on the scanned point cloud of the boiler wall, the scanned point cloud includes multiple scanned point cloud data of each target block structure; obtaining multi-scale mask feature maps of each target block structure based on the corresponding three-dimensional voxel features, the multi-scale mask feature map includes multiple mask feature maps of different scales; and constructing a three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure.

[0088] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A three-dimensional modeling method for a boiler wall, characterized in that: include: Acquire three-dimensional voxel features of each target block structure of the boiler wall based on a scanned point cloud of the boiler wall, wherein the scanned point cloud includes multiple scanned point cloud data of each target block structure; Acquire a multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel features, wherein the multi-scale mask feature map includes a plurality of mask feature maps of different scales; as well as The three-dimensional model of the boiler wall is constructed based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure.

2. The three-dimensional modeling method of boiler wall according to claim 1, characterized in that: The method of acquiring the three-dimensional voxel features of each target block structure of the boiler wall based on the scanning point cloud of the boiler wall includes: Based on the corresponding target block structure, semantically annotating the multiple scanned point cloud data of each target block structure to obtain multiple groups of annotated point clouds of each target block structure; and The three-dimensional voxel features of the corresponding groups of annotated point cloud data are extracted respectively to obtain a plurality of three-dimensional voxel features of the target block structures.

3. The three-dimensional modeling method of boiler wall according to claim 2, characterized in that: The step of acquiring the multi-scale mask feature map of each target block structure based on the corresponding three-dimensional voxel feature includes: Aligning and fusing the multiple three-dimensional voxel features of the respective target block structures to obtain fused voxel features of the respective target block structures; and Based on multiple preset scales, the corresponding fused voxel features are three-dimensionally reconstructed to obtain the multi-scale mask feature map of each target block structure.

4. The three-dimensional modeling method of boiler wall according to claim 1, characterized in that: Before constructing the three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure, the three-dimensional modeling method further includes: Extracting features of the real scene images of the target block structures to obtain features of the real scene images of the target block structures; The three-dimensional model of the boiler wall is constructed based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure, including: Fusing the real scene image features of each target block structure with the multi-scale mask feature map to obtain three-dimensional voxel semantic features of each target block structure; and The three-dimensional model of the boiler wall is constructed based on the three-dimensional voxel semantic features of each target block structure.

5. The three-dimensional modeling method of boiler wall according to claim 4, characterized in that: The step of fusing the real scene image features of each target block structure with the multi-scale mask feature map to obtain the three-dimensional voxel semantic features of each target block structure includes: Determining three-dimensional coordinate information of the corresponding real scene image feature based on the multi-scale mask feature map; and The multi-scale mask feature map is aligned and fused with the corresponding real scene image features based on the three-dimensional coordinate information of the corresponding real scene image features.

6. The three-dimensional modeling method of boiler wall according to claim 5, characterized in that: The aligning and fusing the multi-scale mask feature map with the corresponding real scene image features comprises: For each target block structure, aligning and fusing the real scene image features of the current target block structure with a plurality of corresponding mask feature maps of similar scales to obtain a plurality of three-dimensional voxel semantic features of the current target block structure; The constructing the three-dimensional model of the boiler wall based on the three-dimensional voxel semantic features comprises: Constructing and obtaining a plurality of prediction modeling results of the current target block structure based on the plurality of three-dimensional semantic features of the current target block structure respectively; Taking the intersection of the multiple prediction modeling results to obtain a three-dimensional model of the current target block structure, and The three-dimensional model of the boiler wall is constructed based on the three-dimensional model of the current target block structure.

7. The three-dimensional modeling method of boiler wall according to claim 1, characterized in that: The target block structures of the boiler wall include: water-cooled wall tubes, manhole doors, burners, fire-viewing holes, furnace wall surfaces and surface coke.

8. A three-dimensional modeling device for a boiler wall, characterized in that: include: A three-dimensional voxel feature acquisition module, used to acquire three-dimensional voxel features of each target block structure of the boiler wall based on a scan point cloud of the boiler wall, wherein the scan point cloud includes multiple scan point cloud data of each target block structure; A multi-scale mask feature map acquisition module, used to acquire a multi-scale mask feature map of each target block structure based on corresponding three-dimensional voxel features, wherein the multi-scale mask feature map includes a plurality of mask feature maps of different scales; as well as The three-dimensional model acquisition module is used to construct the three-dimensional model of the boiler wall based on the multi-scale mask feature map of each target block structure and the real scene image of each target block structure.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the three-dimensional modeling method of the boiler wall as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the three-dimensional modeling method of a boiler wall as claimed in any one of claims 1 to 7 is implemented.

Images