Displacement monitoring methods and related devices based on artificial intelligence and multi-view vision

Abstract

This invention relates to the field of artificial intelligence and discloses a displacement monitoring method and related device based on artificial intelligence and multi-view vision to improve the accuracy of displacement monitoring. The method includes: performing feature fusion and feature matching on multiple tunnel feature points to obtain multiple feature point pairs, and acquiring the feature point coordinates of each feature point pair; calculating attribute parameters of a visual displacement analyzer based on the feature point coordinates of each feature point pair; performing three-dimensional mapping of the gas pipeline tunnel based on the attribute parameters and multiple detection areas to obtain initial coordinate information, and performing coordinate transformation on the initial coordinate information to obtain first coordinate information; generating second coordinate information based on a second tunnel image, and calculating displacement and vibration frequency data based on the first and second coordinate information; generating a displacement-frequency curve based on the displacement and vibration frequency data, and performing safety monitoring of the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and in particular to a displacement monitoring method and related apparatus based on artificial intelligence and multi-view vision. Background Technology

[0002] For safety monitoring of gas pipeline tunnels, the harsh environment, high dust levels, and insufficient lighting within tunnels necessitate research into intelligent visual monitoring equipment based on visual AI algorithms and key technologies for data post-processing. This research aims to achieve displacement measurement accuracy superior to conventional monitoring methods, meeting tunnel safety monitoring requirements. The monitoring effect is linked to optimized parameter configuration technology on intelligent terminals. Video displacement monitoring differs significantly from video surveillance in data processing and architecture. While providing real-time viewing of the scene, video displacement monitoring utilizes built-in algorithms to identify displacement deformation at the monitored cross-section. The monitoring dimension shifts from image display to image recognition, and from image viewing to data analysis. Therefore, traditional parameter configurations cannot meet or adapt to monitoring needs.

[0003] During the operation and maintenance period of gas pipeline tunnels, real-time monitoring is required to ensure the safe operation of the gas pipeline tunnels. Existing solutions are greatly affected by changes in image background and do not perform well in terms of stability. For high-precision monitoring of structural deformation detection in gas pipeline tunnel operations, it is difficult to achieve the ideal effect, resulting in low displacement monitoring accuracy of existing solutions. Summary of the Invention

[0004] This invention provides a displacement monitoring method and related device based on artificial intelligence and multi-view vision to improve the accuracy of displacement monitoring.

[0005] The first aspect of this invention provides a displacement monitoring method based on artificial intelligence and multi-view vision. The method includes: when monitoring is conducted on a gas pipeline tunnel during its operation and maintenance period, a preset visual displacement analyzer is used to collect images of the tunnel at an initial time and a current time, obtaining a first tunnel image and a second tunnel image; feature extraction is performed on the first tunnel image to obtain multiple tunnel feature points, and feature point coordinates are constructed for each tunnel feature point; feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained; and a displacement monitoring method is calculated based on the feature point coordinates of each feature point pair. The system calculates the attribute parameters of the visual displacement analyzer; performs three-dimensional mapping on the gas pipeline tunnel based on the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and performs coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment; generates the second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculates the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information; generates a displacement-frequency curve based on the displacement and vibration frequency data, and performs safety monitoring on the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the displacement monitoring method based on artificial intelligence and multi-view vision further includes: performing image segmentation on the first tunnel image and the second tunnel image respectively to obtain a first standard image and a second standard image; performing denoising and contrast enhancement processing on the first standard image and the second standard image respectively to obtain a first target image and a second target image; and storing the first target image and the second target image.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the step of performing feature fusion and feature matching on the plurality of tunnel feature points to obtain a plurality of feature point pairs, obtaining the feature point coordinates of each feature point pair, and calculating the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair includes: performing feature fusion on the plurality of tunnel feature points to obtain fused feature points; performing feature point matching on the fused feature points to obtain a plurality of feature point pairs; obtaining the feature point coordinates of each feature point pair; calculating the relative pose relationship of the visual displacement analyzer based on the feature point coordinates of each feature point pair, and using the relative pose relationship as the attribute parameters of the visual displacement analyzer.

[0008] Optionally, in a third implementation of the first aspect of the present invention, the step of performing three-dimensional mapping of the gas pipeline tunnel according to the attribute parameters and a preset plurality of detection areas to obtain initial coordinate information, and performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment, includes: performing three-dimensional reconstruction of the gas pipeline tunnel according to the attribute parameters and a preset plurality of detection areas to obtain a three-dimensional model of the tunnel; performing coordinate mapping on the three-dimensional model of the tunnel to obtain initial coordinate information; and performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0009] Optionally, in a fourth implementation of the first aspect of the present invention, the step of performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment includes: determining the three-dimensional coordinates of the initial coordinate information according to the attribute parameters; constructing a parameter transformation model according to the three-dimensional coordinates; calculating the coordinate transformation relationship according to the parameter transformation model; and performing coordinate transformation on the initial coordinate information according to the coordinate transformation relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the step of generating second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculating displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information, includes: performing coordinate transformation on the second tunnel image to obtain second coordinate information of the gas pipeline tunnel at the current moment; obtaining coordinate data of the same feature points based on the first coordinate information and the second coordinate information; and calculating displacement and vibration frequency data of the gas pipeline tunnel based on the coordinate data of the same feature points.

[0011] Optionally, in a sixth implementation of the first aspect of the present invention, the displacement monitoring method based on artificial intelligence and multi-view vision further includes: performing coordinate correction on the visual displacement analyzer at preset time intervals to obtain corrected coordinate information; and performing displacement correction on the gas pipeline tunnel based on the corrected coordinate information.

[0012] A second aspect of the present invention provides a displacement monitoring device based on artificial intelligence and multi-view vision. The device comprises: a data acquisition module, used to acquire images of the gas pipeline tunnel at an initial time and a current time when monitoring is carried out during the operation and maintenance period; an extraction module, used to extract features from the first tunnel image to obtain multiple tunnel feature points, and to construct the feature point coordinates of each tunnel feature point; and a processing module, used to perform feature fusion and feature matching on the multiple tunnel feature points to obtain multiple feature point pairs, and to obtain the feature point coordinates of each feature point pair, and to calculate the feature point coordinates based on the feature point coordinates of each feature point pair. The system includes: a visual displacement analyzer's attribute parameters; a conversion module for performing three-dimensional mapping on the gas pipeline tunnel based on the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment; a calculation module for generating the second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculating the displacement and vibration frequency data of the gas pipeline tunnel based on the first and second coordinate information; and a monitoring module for generating a displacement-frequency curve based on the displacement and vibration frequency data, and performing safety monitoring on the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

[0013] Optionally, in a first implementation of the second aspect of the present invention, the displacement monitoring device based on artificial intelligence and multi-view vision further includes: a storage module, configured to perform image segmentation on the first tunnel image and the second tunnel image respectively to obtain a first standard image and a second standard image; perform denoising and contrast enhancement processing on the first standard image and the second standard image respectively to obtain a first target image and a second target image; and store the first target image and the second target image.

[0014] Optionally, in a second implementation of the second aspect of the present invention, the processing module is specifically used to: perform feature fusion on the plurality of tunnel feature points respectively to obtain fused feature points; perform feature point matching on the fused feature points to obtain a plurality of feature point pairs; obtain the feature point coordinates of each feature point pair; calculate the relative pose relationship of the visual displacement analyzer based on the feature point coordinates of each feature point pair, and use the relative pose relationship as the attribute parameter of the visual displacement analyzer.

[0015] Optionally, in a third implementation of the second aspect of the present invention, the conversion module further includes: a reconstruction unit, used to perform three-dimensional reconstruction of the gas pipeline tunnel according to the attribute parameters and a preset plurality of detection areas to obtain a three-dimensional model of the tunnel; a mapping unit, used to perform coordinate mapping on the three-dimensional model of the tunnel to obtain initial coordinate information; and a conversion unit, used to perform coordinate conversion on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial time.

[0016] Optionally, in a fourth implementation of the second aspect of the present invention, the conversion unit is specifically used for: determining the three-dimensional coordinates of the initial coordinate information according to the attribute parameters; constructing a parameter conversion model according to the three-dimensional coordinates; calculating the coordinate conversion relationship according to the parameter conversion model; and performing coordinate conversion on the initial coordinate information according to the coordinate conversion relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0017] Optionally, in a fifth implementation of the second aspect of the present invention, the calculation module is specifically used to: perform coordinate transformation on the second tunnel image to obtain the second coordinate information of the gas pipeline tunnel at the current time; obtain coordinate data of the same feature points based on the first coordinate information and the second coordinate information; and calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the coordinate data of the same feature points.

[0018] Optionally, in a sixth implementation of the second aspect of the present invention, the displacement monitoring device based on artificial intelligence and multi-view vision further includes: a correction module, used to perform coordinate correction on the visual displacement analyzer at preset time intervals to obtain corrected coordinate information; and to perform displacement correction on the gas pipeline tunnel based on the corrected coordinate information.

[0019] A third aspect of the present invention provides a displacement monitoring device based on artificial intelligence and multi-view vision, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the displacement monitoring device based on artificial intelligence and multi-view vision to execute the above-described displacement monitoring method based on artificial intelligence and multi-view vision.

[0020] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described displacement monitoring method based on artificial intelligence and multi-view vision.

[0021] In the technical solution provided by this invention, a multi-view vision system composed of the preset visual displacement analyzer is used to monitor the gas pipeline tunnel. By extracting features from the first tunnel image, multiple tunnel feature points are obtained, and the feature point coordinates of each tunnel feature point are constructed. Feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Combined with the displacement information at the initial moment, manual measurement is replaced. Combining the advantages of multi-view vision system, such as real-time operation, high accuracy, and low cost, it is convenient to realize simultaneous monitoring of multiple points, which greatly improves the monitoring accuracy and thus improves the safety monitoring accuracy of gas pipeline tunnels. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of an embodiment of the displacement monitoring method based on artificial intelligence and multi-view vision in this invention.

[0023] Figure 2 This is a schematic diagram of another embodiment of the displacement monitoring method based on artificial intelligence and multi-view vision in this invention.

[0024] Figure 3 This is a schematic diagram of one embodiment of the displacement monitoring device based on artificial intelligence and multi-view vision in this invention.

[0025] Figure 4 This is a schematic diagram of another embodiment of the displacement monitoring device based on artificial intelligence and multi-view vision in this invention.

[0026] Figure 5 This is a schematic diagram of an embodiment of a displacement monitoring device based on artificial intelligence and multi-view vision according to the present invention. Detailed Implementation

[0027] This invention provides a displacement monitoring method and related apparatus based on artificial intelligence and multi-view vision to improve the accuracy of displacement monitoring. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the displacement monitoring method based on artificial intelligence and multi-view vision in this invention includes:

[0029] 101. When monitoring is carried out on the mountain tunnel of the gas pipeline during the operation and maintenance period, the preset visual displacement analyzer is called to collect images of the gas pipeline tunnel at the initial time and the current time, respectively, to obtain the first tunnel image and the second tunnel image.

[0030] It is understood that the executing entity of this invention can be a displacement monitoring device based on artificial intelligence and multi-view vision, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0031] It should be noted that during the monitoring of the gas pipeline tunnel during the operation and maintenance period, a pre-set visual displacement analyzer is used to collect images of the tunnel at the initial moment and the current moment, obtaining the first tunnel image and the second tunnel image. The pre-set visual displacement analyzer includes two image acquisition terminals. These terminals acquire images of the tunnel at the initial moment (first tunnel image) and at the current moment (second tunnel image). Specifically, firstly, a marker with n black squares is fixed to the gas pipeline tunnel. Then, the visual displacement analyzer is fixed on the ground approximately 10-50 meters away from the marker. The two image acquisition terminals are spaced a certain distance apart, with their shooting directions approximately at a 90-degree angle. The shooting angles of the image acquisition terminals are then adjusted so that the visual displacement analyzer can capture the marker. The aperture and focus are then adjusted to ensure the marker image has appropriate brightness and is clear. Finally, the shooting range of the visual displacement analyzer is set to include the marker and an area approximately twice the size of the marker. The two image acquisition terminals can be fixed using observation tripods. This embodiment can simultaneously support the monitoring of multiple targets within the field of view. By considering different evaluation requirements such as the smoke environment inside the tunnel, the construction environment, and the monitoring distance, combined with the tunnel cross-sectional shape and geometric dimensions, and taking into account aspects such as clarity, resolution, frame rate, bitrate, and image configuration, a set of reasonable intelligent camera parameters is established. The intelligent monitoring camera positions and corresponding target positions are then rationally deployed to the site to improve the actual monitoring effect. Based on equipment characteristics and actual monitoring requirements, the principles of physical target placement are studied, and a target-based tunnel settlement monitoring system is established. Video image recognition is highly sensitive to horizontal and vertical displacement deformation. During the monitoring process, video cameras are used to monitor the surrounding environment while simultaneously identifying target displacement. The impact of cameras on monitoring accuracy, weak on-site lighting conditions, and the selection of on-site monitoring angles are fully considered. Through research on target placement, practical problems are solved, and a suitable placement scheme for monitoring arched settlement of tunnel cross-sections is designed. Targeted construction and installation can then greatly improve the sensitivity and accuracy of monitoring.

[0032] 102. Extract features from the first tunnel image to obtain multiple tunnel feature points, and construct the feature point coordinates for each tunnel feature point.

[0033] In this process, feature extraction is performed on the first tunnel image to obtain multiple tunnel feature points, and the feature point coordinates of each tunnel feature point are constructed. The first and second tunnel images are acquired using a visual displacement analyzer, and feature points (the center point of each square, including the center point of the center square) are extracted and sorted for matching. This allows for accurate determination of the relative pose relationship between the first and second tunnel images based on the pixel coordinates of multiple matched feature point pairs. The data processing volume is small, significantly improving efficiency and making it fully adaptable to practical application scenarios. By combining the coordinates of the n square center points and their corresponding pixel coordinates, the intrinsic and extrinsic parameters of the two image acquisition terminals can be determined, thus obtaining the relative pose relationship between the two image acquisition terminals. Then, using the pixel coordinates of the calibration image information of the two image acquisition terminals corresponding to a feature point in the world coordinate system, combined with the relative pose relationship between the two image acquisition terminals, the feature point coordinates of each tunnel feature point can be calculated.

[0034] 103. Perform feature fusion and feature matching on multiple tunnel feature points respectively to obtain multiple feature point pairs, and obtain the feature point coordinates of each feature point pair. Calculate the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair.

[0035] Specifically, feature fusion and feature matching are performed on multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Specifically, this includes: performing feature fusion on multiple tunnel feature points to obtain fused feature points; performing feature point matching on the fused feature points to obtain multiple feature point pairs; obtaining the feature point coordinates of each feature point pair; calculating the relative pose relationship of the visual displacement analyzer based on the feature point coordinates of each feature point pair, and using the relative pose relationship as the attribute parameters of the visual displacement analyzer.

[0036] 104. Based on the attribute parameters and multiple preset detection areas, perform three-dimensional mapping on the gas pipeline tunnel to obtain the initial coordinate information, and perform coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0037] Specifically, the coordinates of feature points in the first and second tunnel images after initial cropping are obtained respectively. Combining these two images allows for the determination of the three-dimensional coordinates of the feature points in the three-dimensional coordinate system, achieving precise positioning. The process involves determining the three-dimensional coordinates of the initial coordinate information based on attribute parameters; constructing a parameter transformation model based on the three-dimensional coordinates; calculating the coordinate transformation relationship based on the parameter transformation model; and performing coordinate transformation on the initial coordinate information based on the coordinate transformation relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0038] 105. Generate the second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information.

[0039] Specifically, coordinate transformation is performed on the second tunnel image to obtain the second coordinate information of the gas pipeline tunnel at the current moment; coordinate data of the same feature points are obtained based on the first and second coordinate information; displacement and vibration frequency data of the gas pipeline tunnel are calculated based on the coordinate data of the same feature points. By recalibrating the relative pose relationship and the pose change information of the image acquisition terminal, the three-dimensional coordinates of the center point in the current three-dimensional coordinate system can be adjusted to the three-dimensional coordinates in the initial three-dimensional coordinate system. This can eliminate the error caused by the relative pose change between the two image acquisition terminals and improve the accuracy of the monitoring results.

[0040] 106. Generate displacement-frequency curves based on displacement and vibration frequency data, and conduct safety monitoring of gas pipeline tunnels based on displacement-frequency curves to obtain safety monitoring results.

[0041] Specifically, displacement and vibration frequency data are visualized, generating displacement-frequency curves. Safety monitoring of the gas pipeline tunnel is then conducted based on these curves, yielding safety monitoring results. When displacement and frequency exceed abnormal values, the monitoring result indicates a potential safety hazard; conversely, when displacement and frequency do not exceed abnormal values, the monitoring result indicates no safety hazard. These safety monitoring results are generated using a pre-built digital twin engine, creating a visualized image that is then presented in three dimensions.

[0042] In this embodiment of the invention, a multi-view vision system composed of a preset visual displacement analyzer is used to monitor gas pipeline tunnels. By extracting features from the first tunnel image, multiple tunnel feature points are obtained, and the feature point coordinates of each tunnel feature point are constructed. Feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Combined with the displacement information at the initial moment, this replaces manual measurement. Combining the advantages of multi-view vision systems—real-time, high precision, and low cost—it is convenient to achieve simultaneous monitoring of multiple points, greatly improving the monitoring accuracy and thus improving the safety monitoring accuracy of gas pipeline tunnels.

[0043] Please see Figure 2 Another embodiment of the displacement monitoring method based on artificial intelligence and multi-view vision in this invention includes:

[0044] 201. When monitoring is carried out on the mountain tunnel of the gas pipeline during the operation and maintenance period, the preset visual displacement analyzer is called to collect images of the gas pipeline tunnel at the initial time and the current time, respectively, to obtain the first tunnel image and the second tunnel image.

[0045] Optionally, image segmentation is performed on the first tunnel image and the second tunnel image respectively to obtain a first standard image and a second standard image; denoising and contrast enhancement processing are then performed on the first standard image and the second standard image respectively to obtain a first target image and a second target image; the first target image and the second target image are then stored. After the visual displacement analyzer acquires the images, it uses multiple threads to acquire and decode the images. Because the images are acquired by multiple threads, the images in the image queue are not stored sequentially. By performing initial cropping on the first tunnel image and the second tunnel image respectively, the image size can be reduced as much as possible, while ensuring that the corresponding feature points are found after the marker is displaced, thereby improving the calculation speed and reducing memory consumption. Here, since the center points of all the blocks are needed as feature points during calibration, the first and second tunnel images are initially cropped with the center point of the marker (the center point of the center block) as the center. The first and second tunnel images are then cropped to an area twice the area of ​​the marker. This reduces the image size and ensures that all n blocks are located within the first and second tunnel images. Furthermore, it accurately locates the center points of all blocks even with slight displacement of the marker. Using the flash signal from one image acquisition terminal as the trigger signal for the other allows for complete synchronization between the two terminals during shooting. Selecting two images taken at the same time from the cached image information after shooting ensures that the first and second tunnel images used to determine the relative pose between the two terminals are taken at the same moment, thus guaranteeing the accuracy of the results and preventing deviations caused by asynchronous image information. Specifically, two seconds of image information data are cached, and each time, two calibration images with the same shooting time from the previous second of the cached data are selected as synchronized images. Images that are not found to be synchronized are retained until the next second, and if they are still not found in the next second, they are deleted directly.

[0046] 202. Perform feature extraction on the first tunnel image to obtain multiple tunnel feature points, and construct the feature point coordinates for each tunnel feature point;

[0047] 203. Perform feature fusion and feature matching on multiple tunnel feature points respectively to obtain multiple feature point pairs, and obtain the feature point coordinates of each feature point pair. Calculate the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair.

[0048] Specifically, feature fusion is performed on multiple tunnel feature points to obtain fused feature points; feature point matching is then performed on the fused feature points to obtain multiple feature point pairs; the feature point coordinates of each feature point pair are obtained; the relative pose relationship of the visual displacement analyzer is calculated based on the feature point coordinates of each feature point pair, and the relative pose relationship is used as the attribute parameter of the visual displacement analyzer. The coordinates of feature points in the first and second tunnel images after initial cropping are obtained respectively. Combining the first and second tunnel images allows for the determination of the three-dimensional coordinates of the feature points in the coordinate system of an image acquisition terminal, achieving precise positioning.

[0049] 204. Based on the attribute parameters and multiple preset detection areas, the gas pipeline tunnel is reconstructed in three dimensions to obtain a three-dimensional model of the tunnel;

[0050] 205. Perform coordinate mapping on the 3D model of the tunnel to obtain initial coordinate information;

[0051] 206. Perform coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment;

[0052] Optionally, the three-dimensional coordinates of the initial coordinate information are determined based on the attribute parameters; a parameter transformation model is constructed based on the three-dimensional coordinates; the coordinate transformation relationship is calculated based on the parameter transformation model; and the initial coordinate information is transformed based on the coordinate transformation relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0053] Specifically, the relative pose relationship between the two image acquisition terminals is recalibrated at set intervals, and the three-dimensional coordinates of the center point in the coordinate system of one image acquisition terminal at the current moment are corrected based on the recalibrated relative pose relationship. By recalibrating the relative pose relationship between the two image acquisition terminals and correcting the three-dimensional coordinates of the center point in the coordinate system of one image acquisition terminal at the current moment based on the recalibrated relative pose relationship, errors caused by changes in the relative pose between the two image acquisition terminals can be eliminated, improving the accuracy of the monitoring results.

[0054] 207. Generate the second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information.

[0055] Specifically, coordinate transformation is performed on the second tunnel image to obtain the second coordinate information of the gas pipeline tunnel at the current moment; coordinate data of the same feature points are obtained based on the first and second coordinate information; and displacement and vibration frequency data of the gas pipeline tunnel are calculated based on the coordinate data of the same feature points.

[0056] 208. Generate displacement-frequency curves based on displacement and vibration frequency data, and conduct safety monitoring of gas pipeline tunnels based on displacement-frequency curves to obtain safety monitoring results.

[0057] Optionally, the visual displacement analyzer is recalibrated at preset time intervals to obtain the corrected coordinate information; the displacement of the gas pipeline tunnel is then corrected based on the corrected coordinate information.

[0058] The specific steps for correcting the three-dimensional coordinates of the center point in the coordinate system of an image acquisition terminal include: acquiring first and second image information captured by two image acquisition terminals in real time at set intervals; recalibrating the relative pose relationship between the two image acquisition terminals at the current moment based on the real-time first and second image information, and determining the pose change information of one image acquisition terminal based on the first image information at the current moment and the first image information at the initial moment; calculating the three-dimensional coordinates of the center point in the coordinate system of one image acquisition terminal at the current moment based on the relative pose relationship between the two image acquisition terminals at the current moment, and calculating the three-dimensional coordinates of the center point in the coordinate system of one image acquisition terminal at the initial moment by combining the pose change information of one image acquisition terminal. By using the recalibrated relative pose relationship and the pose change information of one image acquisition terminal, the three-dimensional coordinates of the center point in the coordinate system of one image acquisition terminal at the current moment can be adjusted to the three-dimensional coordinates in the coordinate system of one image acquisition terminal at the initial moment. This eliminates the error caused by the relative pose change between the two image acquisition terminals and improves the accuracy of the monitoring results.

[0059] In this embodiment of the invention, a multi-view vision system composed of a preset visual displacement analyzer is used to monitor gas pipeline tunnels. By extracting features from the first tunnel image, multiple tunnel feature points are obtained, and the feature point coordinates of each tunnel feature point are constructed. Feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Combined with the displacement information at the initial moment, this replaces manual measurement. Combining the advantages of multi-view vision systems—real-time, high precision, and low cost—it is convenient to achieve simultaneous monitoring of multiple points, greatly improving the monitoring accuracy and thus improving the safety monitoring accuracy of gas pipeline tunnels.

[0060] The displacement monitoring method based on artificial intelligence and multi-view vision in the embodiments of the present invention has been described above. The displacement monitoring device based on artificial intelligence and multi-view vision in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 3 One embodiment of the displacement monitoring device based on artificial intelligence and multi-view vision in this invention includes:

[0061] The acquisition module 301 is used to call a preset visual displacement analyzer to acquire images of the gas pipeline tunnel at the initial moment and the current moment when monitoring is carried out on the mountain tunnel of the gas pipeline during the operation and maintenance period, so as to obtain the first tunnel image and the second tunnel image.

[0062] The extraction module 302 is used to extract features from the first tunnel image to obtain multiple tunnel feature points, and to construct the feature point coordinates of each tunnel feature point respectively;

[0063] The processing module 303 is used to perform feature fusion and feature matching on the multiple tunnel feature points respectively to obtain multiple feature point pairs, and to obtain the feature point coordinates of each feature point pair, and to calculate the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair.

[0064] The conversion module 304 is used to perform three-dimensional mapping on the gas pipeline tunnel according to the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and to perform coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial time.

[0065] The calculation module 305 is used to generate the second coordinate information of the gas pipeline tunnel at the current time based on the second tunnel image, and to calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information.

[0066] The monitoring module 306 is used to generate a displacement-frequency curve based on the displacement and vibration frequency data, and to perform safety monitoring on the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

[0067] In this embodiment of the invention, a multi-view vision system composed of the preset visual displacement analyzer is used to monitor the gas pipeline tunnel. By extracting features from the first tunnel image, multiple tunnel feature points are obtained, and the feature point coordinates of each tunnel feature point are constructed. Feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Combined with the displacement information at the initial moment, this replaces manual measurement. Combining the advantages of multi-view vision systems—real-time, high precision, and low cost—it is convenient to realize simultaneous monitoring of multiple points, greatly improving the monitoring accuracy and thus improving the safety monitoring accuracy of the gas pipeline tunnel.

[0068] Please see Figure 4 Another embodiment of the displacement monitoring device based on artificial intelligence and multi-view vision in this invention includes:

[0069] The acquisition module 301 is used to call a preset visual displacement analyzer to acquire images of the gas pipeline tunnel at the initial moment and the current moment when monitoring is carried out on the mountain tunnel of the gas pipeline during the operation and maintenance period, so as to obtain the first tunnel image and the second tunnel image.

[0070] The extraction module 302 is used to extract features from the first tunnel image to obtain multiple tunnel feature points, and to construct the feature point coordinates of each tunnel feature point respectively;

[0071] The processing module 303 is used to perform feature fusion and feature matching on the multiple tunnel feature points respectively to obtain multiple feature point pairs, and to obtain the feature point coordinates of each feature point pair, and to calculate the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair.

[0072] The conversion module 304 is used to perform three-dimensional mapping on the gas pipeline tunnel according to the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and to perform coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial time.

[0073] The calculation module 305 is used to generate the second coordinate information of the gas pipeline tunnel at the current time based on the second tunnel image, and to calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information.

[0074] The monitoring module 306 is used to generate a displacement-frequency curve based on the displacement and vibration frequency data, and to perform safety monitoring on the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

[0075] Optionally, the displacement monitoring device based on artificial intelligence and multi-view vision further includes:

[0076] The storage module 307 is used to perform image segmentation on the first tunnel image and the second tunnel image respectively to obtain a first standard image and a second standard image; to perform denoising and contrast enhancement processing on the first standard image and the second standard image respectively to obtain a first target image and a second target image; and to store the first target image and the second target image.

[0077] Optionally, the processing module 303 is specifically used for: performing feature fusion on the plurality of tunnel feature points respectively to obtain fused feature points; performing feature point matching on the fused feature points to obtain a plurality of feature point pairs; obtaining the feature point coordinates of each feature point pair; calculating the relative pose relationship of the visual displacement analyzer based on the feature point coordinates of each feature point pair, and using the relative pose relationship as the attribute parameter of the visual displacement analyzer.

[0078] Optionally, the conversion module 304 further includes: a reconstruction unit, used to perform three-dimensional reconstruction of the gas pipeline tunnel according to the attribute parameters and a preset multiple detection areas to obtain a three-dimensional model of the tunnel; a mapping unit, used to perform coordinate mapping on the three-dimensional model of the tunnel to obtain initial coordinate information; and a conversion unit, used to perform coordinate conversion on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial time.

[0079] Optionally, the conversion unit is specifically used for: determining the three-dimensional coordinates of the initial coordinate information according to the attribute parameters; constructing a parameter conversion model according to the three-dimensional coordinates; calculating the coordinate conversion relationship according to the parameter conversion model; and performing coordinate conversion on the initial coordinate information according to the coordinate conversion relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

[0080] Optionally, the calculation module 305 is specifically used to: perform coordinate transformation on the second tunnel image to obtain the second coordinate information of the gas pipeline tunnel at the current time; obtain coordinate data of the same feature points based on the first coordinate information and the second coordinate information; and calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the coordinate data of the same feature points.

[0081] Optionally, the displacement monitoring device based on artificial intelligence and multi-view vision further includes:

[0082] The calibration module 308 is used to perform coordinate correction on the visual displacement analyzer at preset time intervals to obtain corrected coordinate information; and to perform displacement correction on the gas pipeline tunnel based on the corrected coordinate information.

[0083] In this embodiment of the invention, a multi-view vision system composed of the preset visual displacement analyzer is used to monitor the gas pipeline tunnel. By extracting features from the first tunnel image, multiple tunnel feature points are obtained, and the feature point coordinates of each tunnel feature point are constructed. Feature fusion and feature matching are performed on the multiple tunnel feature points to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. Combined with the displacement information at the initial moment, this replaces manual measurement. Combining the advantages of multi-view vision systems—real-time, high precision, and low cost—it is convenient to realize simultaneous monitoring of multiple points, greatly improving the monitoring accuracy and thus improving the safety monitoring accuracy of the gas pipeline tunnel.

[0084] above Figure 3 and Figure 4The displacement monitoring device based on artificial intelligence and multi-view vision in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The displacement monitoring device based on artificial intelligence and multi-view vision in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0085] Figure 5 This is a schematic diagram of the structure of a displacement monitoring device based on artificial intelligence and multi-view vision provided in an embodiment of the present invention. The displacement monitoring device 500 based on artificial intelligence and multi-view vision can vary considerably due to different configurations or performance. It may include one or more central processing units (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) for storing application programs 533 or data 532. The memory 520 and storage media 530 can be temporary or persistent storage. The program stored in the storage media 530 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the displacement monitoring device 500 based on artificial intelligence and multi-view vision. Furthermore, the processor 510 may be configured to communicate with the storage media 530 and execute the series of instruction operations in the storage media 530 on the displacement monitoring device 500 based on artificial intelligence and multi-view vision.

[0086] The displacement monitoring device 500 based on artificial intelligence and multi-view vision may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input / output interfaces 560, and / or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 5 The illustrated structure of the displacement monitoring device based on artificial intelligence and multi-view vision does not constitute a limitation on the displacement monitoring device based on artificial intelligence and multi-view vision. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0087] The present invention also provides a displacement monitoring device based on artificial intelligence and multi-view vision. The displacement monitoring device based on artificial intelligence and multi-view vision includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the displacement monitoring method based on artificial intelligence and multi-view vision in the above embodiments.

[0088] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the displacement monitoring method based on artificial intelligence and multi-view vision.

[0089] Furthermore, the computer-readable storage medium may primarily include a stored program area and a stored data area, wherein the stored program area may store the operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of blockchain nodes, etc.

[0090] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0091] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0092] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0093] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A displacement monitoring method based on artificial intelligence and multi-view vision, characterized in that, The displacement monitoring method based on artificial intelligence and multi-view vision includes: When monitoring is carried out on the mountain tunnel of the gas pipeline during the operation and maintenance period, the preset visual displacement analyzer is called to collect images of the gas pipeline tunnel at the initial moment and the current moment, respectively, to obtain the first tunnel image and the second tunnel image; Feature extraction is performed on the first tunnel image to obtain multiple tunnel feature points, and the feature point coordinates of each tunnel feature point are constructed respectively; Feature fusion and feature matching are performed on the multiple tunnel feature points respectively to obtain multiple feature point pairs, and the feature point coordinates of each feature point pair are obtained. The attribute parameters of the visual displacement analyzer are calculated based on the feature point coordinates of each feature point pair. The gas pipeline tunnel is three-dimensionally mapped according to the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and the initial coordinate information is transformed to obtain the first coordinate information of the gas pipeline tunnel at the initial time. The second coordinate information of the gas pipeline tunnel at the current moment is generated based on the second tunnel image, and the displacement and vibration frequency data of the gas pipeline tunnel are calculated based on the first coordinate information and the second coordinate information. A displacement-frequency curve is generated based on the displacement and vibration frequency data, and the gas pipeline tunnel is subjected to safety monitoring based on the displacement-frequency curve to obtain safety monitoring results.

2. The displacement monitoring method based on artificial intelligence and multi-view vision according to claim 1, characterized in that, The displacement monitoring method based on artificial intelligence and multi-view vision also includes: The first tunnel image and the second tunnel image are segmented respectively to obtain a first standard image and a second standard image; Denoising and contrast enhancement processes are performed on the first standard image and the second standard image respectively to obtain the first target image and the second target image; The first target image and the second target image are stored.

3. The displacement monitoring method based on artificial intelligence and multi-view vision according to claim 1, characterized in that, The process involves performing feature fusion and feature matching on the multiple tunnel feature points to obtain multiple feature point pairs, acquiring the feature point coordinates of each feature point pair, and calculating the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair, including: The multiple tunnel feature points are fused to obtain fused feature points. Feature point matching is performed on the fused feature points to obtain multiple feature point pairs; Obtain the coordinates of the feature points for each feature point pair; The relative pose relationship of the visual displacement analyzer is calculated based on the feature point coordinates of each feature point pair, and the relative pose relationship is used as the attribute parameter of the visual displacement analyzer.

4. The displacement monitoring method based on artificial intelligence and multi-view vision according to claim 1, characterized in that, The step of performing three-dimensional mapping on the gas pipeline tunnel based on the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and then performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment, includes: The gas pipeline tunnel is reconstructed in three dimensions based on the attribute parameters and multiple preset detection areas to obtain a three-dimensional model of the tunnel. The three-dimensional model of the tunnel is mapped to obtain initial coordinate information; The initial coordinate information is transformed to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

5. The displacement monitoring method based on artificial intelligence and multi-view vision according to claim 4, characterized in that, The step of performing coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial moment includes: The three-dimensional coordinates of the initial coordinate information are determined based on the attribute parameters; Construct a parameter transformation model based on the three-dimensional coordinates; Calculate the coordinate transformation relationship based on the parameter transformation model; The initial coordinate information is transformed according to the coordinate transformation relationship to obtain the first coordinate information of the gas pipeline tunnel at the initial moment.

6. The displacement monitoring method based on artificial intelligence and multi-view vision according to claim 1, characterized in that, The step of generating second coordinate information of the gas pipeline tunnel at the current moment based on the second tunnel image, and calculating displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information, includes: Perform coordinate transformation on the second tunnel image to obtain the second coordinate information of the gas pipeline tunnel at the current time; Based on the first coordinate information and the second coordinate information, obtain the coordinate data of the same feature points; The displacement and vibration frequency data of the gas pipeline tunnel are calculated based on the coordinate data of the same feature points.

7. The displacement monitoring method based on artificial intelligence and multi-view vision according to any one of claims 1-6, characterized in that, The displacement monitoring method based on artificial intelligence and multi-view vision also includes: The visual displacement analyzer is corrected at preset time intervals to obtain the corrected coordinate information. The gas pipeline tunnel is displacement corrected based on the corrected coordinate information.

8. A displacement monitoring device based on artificial intelligence and multi-view vision, characterized in that, The displacement monitoring device based on artificial intelligence and multi-view vision includes: The acquisition module is used to collect images of the gas pipeline tunnel at the initial moment and the current moment when monitoring is carried out during the operation and maintenance period. The first tunnel image and the second tunnel image are obtained by calling the preset visual displacement analyzer. The extraction module is used to extract features from the first tunnel image to obtain multiple tunnel feature points, and to construct the feature point coordinates of each tunnel feature point. The processing module is used to perform feature fusion and feature matching on the multiple tunnel feature points respectively to obtain multiple feature point pairs, and to obtain the feature point coordinates of each feature point pair, and to calculate the attribute parameters of the visual displacement analyzer based on the feature point coordinates of each feature point pair. The conversion module is used to perform three-dimensional mapping on the gas pipeline tunnel according to the attribute parameters and multiple preset detection areas to obtain initial coordinate information, and to perform coordinate transformation on the initial coordinate information to obtain the first coordinate information of the gas pipeline tunnel at the initial time. The calculation module is used to generate second coordinate information of the gas pipeline tunnel at the current time based on the second tunnel image, and to calculate the displacement and vibration frequency data of the gas pipeline tunnel based on the first coordinate information and the second coordinate information. The monitoring module is used to generate a displacement-frequency curve based on the displacement and vibration frequency data, and to perform safety monitoring on the gas pipeline tunnel based on the displacement-frequency curve to obtain safety monitoring results.

9. A displacement monitoring device based on artificial intelligence and multi-view vision, characterized in that, The displacement monitoring device based on artificial intelligence and multi-view vision includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the displacement monitoring device based on artificial intelligence and multi-view vision to execute the displacement monitoring method based on artificial intelligence and multi-view vision as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the displacement monitoring method based on artificial intelligence and multi-view vision as described in any one of claims 1-7.

Images