Tunnel image dynamic splicing method, device and electronic equipment based on camera array

Abstract

The present application provides a method, device, and electronic device for dynamic tunnel image stitching based on a camera array, relating to the field of tunnel detection technology. The method comprises: acquiring multiple sets of tunnel lining images of multiple target tunnel sections, performing resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform resolution; stitching the multiple target tunnel lining images based on the overlap between the shooting ranges of any one of the multiple image sensing devices and the adjacent image sensing devices to obtain a tunnel cross-sectional image of each target tunnel section; and combining the tunnel cross-sectional images of the multiple target tunnel sections to obtain a tunnel image. The present application can improve the quality of the resulting tunnel images, thereby enabling better analysis of the location and severity of damage such as cracks in the tunnel.

Description

Technical Field

[0001] The present invention relates to the field of tunnel detection technology, and in particular to a method, device and electronic equipment for dynamic splicing of tunnel images based on a camera array. Background Art

[0002] Tunnels play an extremely important role in highway, railway and urban rail transit systems, ensuring their safety.

[0003] The tunnel image detection method in the existing technology is usually as follows: a certain number of image capture devices are mounted on the body of a tunnel inspection vehicle along the direction of the tunnel cross-section to take pictures of the tunnel cross-section to obtain multiple cross-sectional images. A fixed model is used to calibrate the preset position and scaling ratio of the image capture devices to stitch the multiple cross-sectional images together.

[0004] However, since the tunnel inspection vehicle takes cross-sectional images of the tunnel while in motion, various reasons such as uneven road surface, problems with the vehicle's own condition, and changes in tunnel size may cause errors or mismatches between the actual cross-sectional images and the images at the preset positions, thereby affecting the actual image stitching results. Summary of the Invention

[0005] The purpose of the present invention is to address the deficiencies in the above-mentioned prior art and provide a method, device and electronic device for dynamic stitching of tunnel images based on a camera array, so as to improve the dynamic stitching effect of tunnel images.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of the present application are as follows:

[0007] In a first aspect, an embodiment of the present application provides a method for dynamic stitching of tunnel images based on a camera array, the method comprising:

[0008] Acquire multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, and the multiple tunnel lining images are images of each target tunnel section at different angles captured by a camera array composed of multiple image sensing devices;

[0009] performing spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution;

[0010] splicing the multiple target tunnel lining images based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and the shooting range of an adjacent image sensing device to obtain a tunnel cross-section image of each target tunnel cross-section;

[0011] The tunnel cross-section images of the multiple target tunnel cross-sections are combined to obtain a tunnel image.

[0012] Optionally, before acquiring multiple sets of tunnel lining images of multiple target tunnel sections, the method further includes:

[0013] Obtain the locations of multiple tunnel sections;

[0014] Dividing the plurality of tunnel sections into a plurality of intervals according to positions of the plurality of tunnel sections;

[0015] Screening standard tunnel sections for each section;

[0016] According to the contour point cloud data of the standard tunnel section and the contour point cloud data of other tunnel sections in each interval, the other tunnel sections are optimized to obtain the target tunnel section, wherein the contour point cloud data is point cloud data in a scanning coordinate system obtained based on a scanner.

[0017] Optionally, screening the standard tunnel section of each section includes:

[0018] Calculating the contour symmetry of each tunnel section based on the contour point cloud data;

[0019] Calculating the contour smoothness of each tunnel section according to the contour point cloud data;

[0020] Calculating the contour continuity of each tunnel section according to the contour point cloud data;

[0021] The standard tunnel section is screened according to a weighted average of the profile symmetry, the profile smoothness, and the profile continuity.

[0022] Optionally, calculating the contour smoothness of each tunnel section according to the contour point cloud data includes:

[0023] Performing contour filtering on each tunnel section according to the contour point cloud data;

[0024] The difference in the profile of each tunnel section before and after filtering is calculated to determine the profile smoothness.

[0025] Optionally, performing spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution includes:

[0026] obtaining a spatial resolution of the plurality of tunnel lining images;

[0027] determining a standard spatial resolution according to the spatial resolutions of the plurality of tunnel lining images;

[0028] According to the standard spatial resolution, the spatial resolutions of the multiple tunnel lining images are modified to obtain the multiple target tunnel lining images with uniform resolution.

[0029] Optionally, before stitching the multiple target tunnel lining images based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and an adjacent image sensing device, the method further includes:

[0030] According to the shooting ranges of the multiple image sensing devices, the overlap degree between the shooting range of any one of the image sensing devices and the shooting range of the adjacent image sensing device is calculated.

[0031] Optionally, before calculating the overlap between the shooting ranges of any image sensing device and the adjacent image sensing device based on the shooting ranges of the multiple image sensing devices, the method further includes:

[0032] Determining, based on the contour point cloud data of the target tunnel section and the position parameters of the image sensing device, limit point cloud data corresponding to the shooting range of each image sensing device, the limit point cloud data being point cloud data of a left limit point and a right limit point that define the shooting range;

[0033] The shooting range of each image perception device is calculated based on the extreme point cloud data.

[0034] In a second aspect, an embodiment of the present application further provides a tunnel image dynamic splicing device based on a camera array, the device comprising:

[0035] An image acquisition module is configured to acquire multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, and the multiple tunnel lining images are images of each target tunnel section at different angles acquired by a camera array composed of multiple image sensing devices;

[0036] a resolution processing module, configured to perform spatial resolution processing on the plurality of tunnel lining images to obtain a plurality of target tunnel lining images with uniform spatial resolution;

[0037] a cross-sectional image stitching module, configured to stitch the plurality of target tunnel lining images together based on a degree of overlap between a shooting range of any image sensing device among the plurality of image sensing devices and a shooting range of an adjacent image sensing device, to obtain a tunnel cross-sectional image of each target tunnel cross-section;

[0038] The tunnel image stitching module is used to combine the tunnel section images of the multiple target tunnel sections to obtain a tunnel image.

[0039] Optionally, before the image acquisition module, the device further includes:

[0040] A section position acquisition module is used to obtain the positions of multiple tunnel sections;

[0041] An interval division module, configured to divide the plurality of tunnel sections into a plurality of intervals according to the positions of the plurality of tunnel sections;

[0042] Section screening module, used to screen the standard tunnel section of each section;

[0043] A section optimization module is used to optimize the other tunnel sections according to the contour point cloud data of the standard tunnel section and the contour point cloud data of the other tunnel sections in each interval to obtain the target tunnel section, wherein the contour point cloud data is point cloud data in a scanning coordinate system obtained based on a scanner.

[0044] Optionally, the cross-section screening module includes:

[0045] A symmetry calculation unit, configured to calculate the contour symmetry of each tunnel section based on the contour point cloud data;

[0046] a smoothness calculation unit, configured to calculate the contour smoothness of each tunnel section based on the contour point cloud data;

[0047] a continuity calculation unit, configured to calculate the contour continuity of each tunnel section based on the contour point cloud data;

[0048] A section screening unit is used to screen the standard tunnel section according to a weighted average value of the profile symmetry, the profile smoothness and the profile continuity.

[0049] Optionally, the smoothness calculation unit includes:

[0050] A contour filtering subunit, configured to perform contour filtering on each tunnel section according to the contour point cloud data;

[0051] The smoothness calculation subunit is used to calculate the difference in the contour of each tunnel section before and after filtering to determine the contour smoothness.

[0052] Optionally, the resolution processing module includes:

[0053] a resolution acquisition unit, configured to acquire the spatial resolution of the plurality of tunnel lining images;

[0054] a standard resolution determining unit, configured to determine a standard spatial resolution according to the spatial resolutions of the plurality of tunnel lining images;

[0055] The resolution unification unit is configured to modify the spatial resolutions of the plurality of tunnel lining images according to the standard spatial resolution to obtain the plurality of target tunnel lining images with unified resolutions.

[0056] Optionally, before the cross-sectional image stitching module, the device further includes:

[0057] The overlap calculation module is used to calculate the overlap between the shooting range of any image sensing device and the shooting range of the adjacent image sensing device according to the shooting ranges of the multiple image sensing devices.

[0058] Optionally, before the overlap calculation module, the device further includes:

[0059] a limit point cloud data determination module, configured to determine, based on the contour point cloud data of the target tunnel section and the position parameters of the image sensing device, limit point cloud data corresponding to the shooting range of each image sensing device, the limit point cloud data being point cloud data of a left limit point and a right limit point that define the shooting range;

[0060] The shooting range calculation module is used to calculate the shooting range of each image perception device based on the limit point cloud data.

[0061] Optionally, the cross-sectional image stitching module includes:

[0062] a pixel cropping unit, configured to perform pixel cropping on the corresponding two target tunnel lining images in a width direction according to the overlap between any one image sensing device and the adjacent image sensing device, to obtain two cropped target tunnel lining images;

[0063] a pixel shifting unit, configured to shift the pixels of the two cropped target tunnel lining images in a height direction to obtain two aligned target tunnel lining images;

[0064] The image cropping unit is used to crop the two aligned target tunnel lining images within the target area to obtain a tunnel cross-section image of each target tunnel cross-section.

[0065] In a third aspect, an embodiment of the present application further provides an electronic device comprising: a processor, a storage medium and a bus, wherein the storage medium stores program instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium through the bus, and the processor executes the program instructions to perform the steps of the method for dynamic stitching of tunnel images based on a camera array as described in any of the above embodiments.

[0066] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for dynamic stitching of tunnel images based on a camera array as described in any of the above embodiments are executed.

[0067] The beneficial effects of this application are:

[0068] Embodiments of the present application provide a method, device, and electronic device for dynamically stitching tunnel images based on a camera array. The method comprises: acquiring multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, each of which is captured by multiple image sensing devices at different angles of each target tunnel section; performing resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform resolution; stitching the multiple target tunnel lining images based on the overlap between the shooting ranges of any one of the multiple image sensing devices and an adjacent image sensing device to obtain a tunnel cross-sectional image of each target tunnel section; and combining the tunnel cross-sectional images of the multiple target tunnel sections to obtain a tunnel image. The solution provided by the present application achieves improved stitching of the resulting tunnel cross-sectional images by dynamically unifying the resolution of the tunnel lining images and stitching the images based on overlap. The tunnel image obtained based on the combined tunnel cross-sectional images can be used to analyze the location and severity of damage, such as cracks, in the tunnel, while maintaining uniform pixel values, to facilitate tunnel protection and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0070] Figure 1 A schematic flow chart of a first method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0071] Figure 2 A schematic structural diagram of a camera array provided in an embodiment of the present application;

[0072] Figure 3 A schematic flow chart of a second method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0073] Figure 4A schematic flow chart of a third method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0074] Figure 5 A schematic flow chart of a fourth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0075] Figure 6 A schematic flow chart of a fifth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0076] Figure 7 A diagram showing the mapping relationship between the image sensing device and the image spatial resolution provided in the embodiment of the present application;

[0077] FIG8( a ) is a schematic diagram of two original pipeline images provided in an embodiment of the present application;

[0078] FIG8( b ) is a schematic diagram of two original pipeline images stitched together according to an embodiment of the present application;

[0079] FIG8( c ) is a schematic diagram of two pipeline-reduced images provided in an embodiment of the present application;

[0080] FIG8( d ) is a schematic diagram of two pipeline-reduced images stitched together according to an embodiment of the present application;

[0081] Figure 9 A schematic flow chart of a sixth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0082] Figure 10 A schematic diagram of the shooting range provided in an embodiment of the present application;

[0083] Figure 11 A schematic flow chart of a seventh method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application;

[0084] Figure 12 A schematic diagram of tunnel image stitching provided in an embodiment of the present application;

[0085] Figure 13 A schematic diagram of a bilinear interpolation method provided in an embodiment of the present application;

[0086] Figure 14 This is a diagram showing the effect of using bilinear interpolation correction provided in an embodiment of the present application;

[0087] Figure 15 A schematic diagram of a tunnel stitching image provided in an embodiment of the present application;

[0088] Figure 16A schematic structural diagram of a tunnel image dynamic splicing device based on a camera array provided in an embodiment of the present application;

[0089] Figure 17 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0090] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions 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.

[0091] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for protection, but merely represents selected embodiments of the present application. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments in the present application without creative work are within the scope of protection of the present application.

[0092] In the description of this application, it should be noted that if the terms "upper", "lower", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the accompanying drawings, or is the orientation or position relationship in which the product of the application is usually placed when in use. It is only for the convenience of describing this application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on this application.

[0093] In addition, the terms "first," "second," and the like in the description and claims of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable 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," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0094] It should be noted that, in the absence of conflict, the features in the embodiments of this application can be combined with each other.

[0095] The camera array-based dynamic tunnel image stitching method provided in an embodiment of the present application is applied to an electronic device with a tunnel image stitching function, such as a computer device. The electronic device is connected to a camera array and a scanner respectively composed of multiple image sensing devices to respectively receive tunnel substrate images of multiple target tunnel sections at different angles collected by the multiple image sensing devices, receive contour point cloud data of each tunnel section scanned by the scanner, and perform tunnel image stitching based on the tunnel substrate images and the contour point cloud data.

[0096] Figure 1 The flowchart of the first method for dynamic tunnel image stitching based on a camera array provided in the embodiment of the present application is as follows: Figure 1 As shown, the method includes:

[0097] S10: Acquire multiple sets of tunnel lining images of multiple target tunnel sections.

[0098] Specifically, the tunnel lining image is a detailed image of the tunnel top obtained by an image sensing device taking a picture of the surface of the tunnel top inside the tunnel. Each set of tunnel lining images includes multiple tunnel lining images, which are images of each target tunnel section at different angles captured by a camera array composed of multiple image sensing devices.

[0099] For example, Figure 2 A schematic diagram of the structure of a camera array provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the camera array includes multiple image sensing devices 1 at different angles, covering a 90° range. These devices can be cameras or CCDs (charge coupled devices) capable of capturing images that meet the required clarity. In this embodiment, the number of these devices is 16, capturing 16 images of the tunnel lining.

[0100] In one optional embodiment, before acquiring multiple sets of tunnel lining images of multiple target tunnel sections, the raw images captured by the multiple image sensing devices are preprocessed to remove invalid images outside the tunnel and determine the boundaries for capturing continuous tunnel lining images. This includes removing images captured before entering the tunnel, protective facilities before and after the tunnel entrance, such as rockfall protection devices connecting the tunnel entrances, and the external areas between multiple intervening tunnels.

[0101] S20: performing spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution.

[0102] Specifically, spatial resolution refers to the size of the actual ground feature represented by each pixel in the image. Due to differences in the focal length of the imaging device and the object distance between the imaging device and the tunnel surface during capture, the spatial resolution of tunnel lining images taken from different angles can vary. By performing spatial resolution processing on multiple tunnel lining images, multiple target tunnel lining images with uniform spatial resolution are obtained. Each pixel in these target tunnel lining images represents the same size portion of the tunnel.

[0103] S30: Based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and the shooting range of the adjacent image sensing device, the multiple target tunnel lining images are stitched together to obtain a tunnel cross-section image of each target tunnel cross-section.

[0104] Specifically, since multiple image sensing devices are arranged in a 90° arc structure, the shooting range of any image sensing device overlaps with that of an adjacent image sensing device, resulting in overlapping tunnel portions in the corresponding two target tunnel lining images. If the two target tunnel lining images are directly spliced, the splicing of the overlapping portions will cause the resulting spliced image to be longer than the actual length of the target tunnel section. Therefore, to avoid this situation, it is necessary to splice the overlapping portions of the corresponding two target tunnel lining images based on the degree of overlap between the shooting ranges of any image sensing device and the adjacent image sensing device. By splicing the overlapping portions of multiple target tunnel lining images, a tunnel cross-sectional image of each target tunnel section is obtained.

[0105] S40: combining the tunnel cross-section images of the plurality of target tunnel cross-sections to obtain a tunnel image.

[0106] Specifically, the above-mentioned S30 is used to splice multiple target tunnel lining images of each target tunnel section to obtain tunnel cross-sectional images of multiple target tunnel sections. By combining the tunnel cross-sectional images of multiple target tunnel sections along the tunnel length direction, a tunnel image of the entire tunnel interior surface can be obtained.

[0107] The camera array-based dynamic tunnel image stitching method provided in the embodiments of the present application obtains multiple sets of tunnel lining images of multiple target tunnel sections, wherein the multiple tunnel lining images are images of each target tunnel section captured by multiple image sensing devices at different angles. The multiple tunnel lining images are then resolution processed to obtain multiple target tunnel lining images with uniform resolution. The multiple target tunnel lining images are then stitched together based on the overlap between the shooting ranges of any one of the multiple image sensing devices and an adjacent image sensing device to obtain a tunnel cross-sectional image of each target tunnel section. The tunnel cross-sectional images of the multiple target tunnel sections are then combined to obtain a tunnel image. The method provided in the embodiments of the present application achieves better stitching of the resulting tunnel cross-sectional images by dynamically performing resolution uniform processing on the tunnel lining images and stitching the images based on overlap. The tunnel image obtained based on the combined tunnel cross-sectional images can be used to analyze the location and severity of damage such as cracks in the tunnel while maintaining uniform pixel values, thereby enabling tunnel protection and maintenance.

[0108] Based on the above embodiments, the present application also provides a method for dynamic splicing of tunnel images based on a camera array. Figure 3 A schematic diagram of the flow of the second method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, before the above S10, the method further includes:

[0109] S01: Obtain the locations of multiple tunnel sections.

[0110] Specifically, the entire tunnel is divided into multiple tunnel sections according to a preset acquisition interval, and at least one position for acquiring tunnel lining images is determined in each tunnel section to obtain the positions of multiple tunnel sections.

[0111] S02: Divide the multiple tunnel sections into multiple intervals according to the positions of the multiple tunnel sections.

[0112] Specifically, the multiple tunnel sections are divided into intervals according to the positions of the multiple tunnel sections and the number of tunnel sections to obtain multiple intervals, each interval including one to multiple tunnel sections, and the positions of the tunnel sections in each interval are continuous.

[0113] For example, the number of tunnel sections is m. Consider n tunnel sections as a section S. Multiple sections are represented as S{S1, S2, ..., S1}, where l = ceil(m / n), where ceil represents m divided by n, rounded up. Each section includes no more than n tunnel sections. For example, the number of tunnel sections is 20. Consider three tunnel sections as a section S. With l = ceil(20 / 3) = 7, each of the first six sections includes three tunnel sections, and the seventh section includes two tunnel sections.

[0114] S03: Screen the standard tunnel section of each interval.

[0115] Specifically, after using the above-mentioned S02 to divide multiple tunnel sections into multiple intervals, a representative section is selected from each interval as the standard tunnel section for each interval. The screening criteria for the standard tunnel section can be the section with the best contour symmetry and / or continuity in each interval. This application does not impose any restrictions on this.

[0116] S04: Optimize the other tunnel sections according to the outline point cloud data of the standard tunnel section and the outline point cloud data of the other tunnel sections in each interval to obtain the target tunnel section.

[0117] Specifically, the contour point cloud data is the point cloud data in the scanning coordinate system obtained by the scanner, such as Figure 2 As shown, the scanner 2 is installed within the sector-shaped area where the multiple image sensing devices 1 are located. The scanner scans each tunnel section at its location and determines the contour point cloud data of each tunnel section based on the signal returned by each tunnel section. The contour point cloud data of each tunnel section is represented by multiple discrete coordinate points. Based on the symmetry and / or continuity of the contour point cloud data of the standard tunnel section, the contour point cloud data of other tunnel sections is optimized for symmetry and / or continuity, and the optimized tunnel section is used as the target tunnel section.

[0118] The camera array-based dynamic tunnel image stitching method provided in the embodiments of the present application obtains the positions of multiple tunnel sections, divides the multiple tunnel sections into multiple intervals based on the positions of the multiple tunnel sections, screens the standard tunnel section in each interval, and optimizes the other tunnel sections based on the contour point cloud data of the standard tunnel section and the contour point cloud data of the other tunnel sections in each interval to obtain the target tunnel section. The method provided in the embodiments of the present application can screen the tunnel sections to determine the most representative standard tunnel section, and optimize the contour point cloud data of the other tunnel sections based on the contour point cloud data of the standard tunnel section, so that the contour of each tunnel section is more optimized, so that the tunnel section image obtained by stitching is more effective.

[0119] Based on the above embodiments, the present application also provides a method for dynamic splicing of tunnel images based on a camera array. Figure 4 A schematic diagram of the process of the third method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown as follows: Figure 4 As shown, the above S03 includes:

[0120] S031: Calculate the contour symmetry of each tunnel section based on the contour point cloud data.

[0121] Specifically, since the tunnel cross-section is structurally close to a semicircular arc, in order to screen out the contour point cloud data obtained by the scanner that better represents the shape of the tunnel cross-section, the contour symmetry of each tunnel cross-section can be determined by determining the center point of the contour point cloud data and calculating the symmetry of the distribution of the contour point cloud data on the left and right sides of the center point.

[0122] In an optional embodiment, a straight line fitting through the center of the semicircle formed by the contour point cloud data can be performed, and the contour symmetry of each tunnel section can be determined based on the slope of the fitted straight line. The larger the slope, the higher the contour symmetry.

[0123] S032: Calculate the contour smoothness of each tunnel section based on the contour point cloud data.

[0124] Specifically, contour smoothness is used to indicate the degree of undulation of contour point cloud data. The contour smoothness of each tunnel section can be determined by calculating the change in angle of the contour curve composed of the contour point cloud data in the tangent direction. The smaller the change in angle in the tangent direction, the higher the contour smoothness.

[0125] S033: Calculate the contour continuity of each tunnel section based on the contour point cloud data.

[0126] Specifically, contour continuity is used to indicate the integrity of the contour point cloud data received by the scanner. To ensure the integrity of the received contour point cloud data, the ratio of the number of received contour point cloud data to a preset number, or the sum or average of the intervals between each two adjacent contour point cloud data in all the contour point cloud data of each tunnel section can be calculated to determine the contour continuity of each tunnel section. This application does not impose any restrictions on this. Among them, the greater the ratio of the number of received contour point cloud data to the preset number, the greater the contour continuity; the smaller the sum or average of the intervals between each two adjacent contour point cloud data in all the contour point cloud data, the greater the contour continuity.

[0127] S034: Screen standard tunnel sections based on the weighted average of profile symmetry, profile smoothness, and profile continuity.

[0128] Specifically, a weighted average value is calculated for the profile symmetry, profile smoothness, and profile continuity, and the tunnel section with the largest weighted average value is used as the standard tunnel section.

[0129] The camera array-based dynamic tunnel image stitching method provided in the embodiment of the present application calculates the contour symmetry of each tunnel section based on the contour point cloud data, calculates the contour smoothness of each tunnel section based on the contour point cloud data, and calculates the contour continuity of each tunnel section based on the contour point cloud data. The standard tunnel section is screened based on the weighted average of the contour symmetry, contour smoothness, and contour continuity. The embodiment of the present application considers the effect of the contour point cloud data of the tunnel section from three aspects: contour symmetry, contour smoothness, and contour continuity, to screen out the optimal standard tunnel section, so that other tunnel sections can be optimized based on the standard tunnel section, thereby improving the contour effect of all tunnel sections and making the effect of the stitched tunnel section image better.

[0130] Based on the above embodiments, the present application also provides a method for dynamic splicing of tunnel images based on a camera array. Figure 5 A schematic diagram of the flow of the fourth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown as follows: Figure 5 As shown, the above S032 includes:

[0131] S0321: Perform contour filtering on each tunnel section based on the contour point cloud data.

[0132] Specifically, the contour point cloud data of each tunnel section is filtered, and contour point cloud data with large deviations are screened out from the contour point cloud data of each tunnel section to obtain a smoother contour. For example, contour filtering can be performed using methods such as median filtering, mean filtering, and Gaussian filtering, which are not limited in this application.

[0133] S0322: Determine the contour smoothness based on the contour difference between the tunnel section before and after filtering.

[0134] Specifically, the optimal section is selected based on the sum of the difference d between the contour curves before and after filtering or the variance s after filtering. The smaller the sum of the difference d between the contour curves before and after filtering, or the smaller the variance s after filtering, the smaller the contour difference of the tunnel section before and after filtering, and the higher the contour smoothness.

[0135] For example, assuming the filtering range is (a, b), the screening formula is:

[0136]

[0137]

[0138] Among them, a and b can be taken according to the range of the horizontal coordinate of the contour point cloud data of each tunnel section.

[0139] The camera array-based dynamic tunnel image stitching method provided in this embodiment performs contour filtering on each tunnel section based on contour point cloud data, calculates the difference between each tunnel section before and after filtering, and determines the contour smoothness. This embodiment utilizes contour filtering to calculate a relatively accurate contour smoothness, facilitating the selection of standard tunnel sections and improving the dynamic tunnel image stitching effect.

[0140] Based on the above embodiments, the present application also provides a method for dynamic splicing of tunnel images based on a camera array. Figure 6 A schematic diagram of the process of the fifth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown as follows: Figure 6 As shown, the above S20 includes:

[0141] S21: Obtain the spatial resolution of multiple tunnel lining images.

[0142] Specifically, Figure 7 The mapping relationship diagram of the image perception device and image spatial resolution provided in the embodiment of the present application is as follows: Figure 7 As shown, the size of the image perception device is c, the focal length is f, and the object distance is d, then the calculation of the spatial resolution R satisfies:

[0143]

[0144] Then the spatial resolution R is: R = cd / f.

[0145] S22: Determine a standard spatial resolution according to the spatial resolutions of the plurality of tunnel lining images.

[0146] Specifically, due to the different focal lengths of the image sensing devices and the object distances during shooting, the spatial resolutions of multiple tunnel lining images of the same target tunnel section are different. The different spatial resolutions result in the same object on the tunnel occupying different pixel sizes in the two images, which will produce discontinuities when the two tunnel lining images are spliced together.

[0147] For example, Figure 8(a) is a schematic diagram of two original pipeline images provided in an embodiment of the present application. As shown in Figure 8(a), there is a pipeline in the target tunnel section, which is the original pipeline in the figure. The pipeline is photographed by two adjacent image sensing devices respectively. The sizes of the two tunnel lining images are both 1104 pixels high and 1376 pixels wide. It can be seen from Figure 8(a) that due to the different spatial resolutions, the pixel widths of the same pipeline in the two tunnel lining images are different. Therefore, if the resolution is not unified, the splicing results of the two tunnel lining images will produce discontinuity.

[0148] For example, FIG8(b) is a schematic diagram of the stitching of two original pipeline images provided in an embodiment of the present application. As shown in FIG8(b), after stitching, the stitching result is 1104 pixels high. Due to the influence of the overlapping area, the width is 2000 pixels. The pipeline does not have an ideal smooth edge shape, but has a size jump, that is, it is thick first and then thin from left to right. Obviously, this does not conform to the characteristics of the actual original pipeline. Therefore, it is necessary to determine the standard spatial resolution to adjust the spatial resolution of multiple tunnel lining images.

[0149] S23: Modify the spatial resolutions of the multiple tunnel lining images according to the standard spatial resolution to obtain multiple target tunnel lining images with uniform resolution.

[0150] Specifically, after determining the standard spatial resolution, the spatial resolutions of multiple tunnel lining images in each group of tunnel lining images are uniformly modified to the standard spatial resolution; if the spatial resolution of the tunnel lining image is smaller than the standard spatial resolution, the size of the tunnel lining image is reduced to enlarge the spatial resolution of the tunnel lining image; if the spatial resolution of the tunnel lining image is larger than the standard spatial resolution, the size of the tunnel lining image is enlarged to reduce the spatial resolution of the tunnel lining image.

[0151] The spatial resolution of the image with a larger spatial resolution is now used as the standard spatial resolution, and the image with a smaller spatial resolution is reduced in size until the two tunnel lining images have the same spatial resolution R.

[0152] It should be noted that if the spatial resolution of an image with a smaller spatial resolution is used as the standard spatial resolution, and an image with a larger spatial resolution is enlarged to reduce the spatial resolution, sub-pixels will appear in the enlarged image, reducing the image clarity. Therefore, the spatial resolution of an image with a larger spatial resolution is selected as the standard spatial resolution to ensure image clarity.

[0153] For example, Figure 8(c) is a schematic diagram of two pipeline reduction images provided in an embodiment of the present application, and Figure 8(d) is a schematic diagram of the two pipeline reduction images after splicing provided in an embodiment of the present application. As shown in Figure 8(d), after the two tunnel lining images with unified resolution are spliced together, the pipeline in the spliced image conforms to the actual original pipeline characteristics, without any size jump, and the original pipeline is presented normally in the spliced result image.

[0154] The camera array-based dynamic tunnel image stitching method provided in embodiments of the present application obtains the spatial resolution of multiple tunnel lining images, determines a standard spatial resolution based on the spatial resolution of the multiple tunnel lining images, and modifies the spatial resolution of the multiple tunnel lining images based on the standard spatial resolution to obtain multiple target tunnel lining images with uniform resolution. The method provided in embodiments of the present application unifies the spatial resolution of multiple tunnel lining images to ensure that objects in the multiple tunnel lining images do not jump in size, thereby ensuring the presentation quality of the stitched tunnel cross-section image.

[0155] Based on the above embodiment, the embodiment of the present application also provides a method for dynamic stitching of tunnel images based on a camera array. Before the above S30, the method also includes: calculating the overlap between the shooting range of any image sensing device and the shooting range of the adjacent image sensing device based on the shooting range of multiple image sensing devices.

[0156] Specifically, the shooting range of each image sensing device is calculated based on the installation position of each image sensing device, the object distance between each image sensing device and the tunnel wall, and the contour point cloud data of each target tunnel section obtained by the scanner. The overlapping range of the two image sensing devices is determined based on the shooting range of any image sensing device and the adjacent image sensing device, and the overlap degree is obtained based on the ratio of the overlapping range to the shooting range.

[0157] Based on the above embodiments, the present application also provides a method for dynamic splicing of tunnel images based on a camera array. Figure 9 A schematic diagram of the flow of the sixth method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown in FIG. Figure 9 As shown, before calculating the overlap between any image sensing device and an adjacent image sensing device based on the shooting ranges of the multiple image sensing devices, the method further includes:

[0158] S51: Determine the limit point cloud data corresponding to the shooting range of each image perception device according to the contour point cloud data of the target tunnel section and the position parameters of the image perception device.

[0159] Specifically, the extreme point cloud data are the point cloud data of the left extreme point and the right extreme point that limit the shooting range, and the position parameter of the image sensing device is the installation angle of the image sensing device relative to the scanner. According to the installation angle of each image sensing device, the contour point cloud data of the left extreme point and the contour point cloud data of the right extreme point are determined from the contour point cloud data of the target tunnel section.

[0160] S53: Calculate the shooting range of each image perception device according to the limit point cloud data.

[0161] Specifically, according to the point cloud data of the left limit point and the point cloud coordinates of the right limit point, the arc length value L is calculated as the shooting range of each image perception device.

[0162] For example, Figure 10 A schematic diagram of the shooting range provided in the embodiment of the present application, such as Figure 10 As shown, the field of view of the first image sensing device on the tunnel wall is The field of view of the second image sensing device on the tunnel wall is The overlapping range of the first image sensing device and the second image sensing device is turn up For any point e on occupy The percentage m and occupy The percentage n of the first image perception device is used as the percentage m, the percentage n is used as the percentage of overlap between the first image perception device and the second image perception device, and the percentage n is used as the percentage of overlap between the second image perception device and the first image perception device. Similarly, the overlap between the remaining image perception devices is calculated.

[0163] For example, any point e can be selected When the midpoint is selected, since the field of view of the image perception device to the tunnel wall is the same, the percentage m and the percentage n are equal, that is, the overlap degree of any two adjacent image perception devices is the same.

[0164] For example, given the coordinates of a(x1,y1) and b(x2,y2) (in the circle center coordinate system), the calculation formula for the arc length L between a and b is as follows:

[0165]

[0166]

[0167] The calculation method of the arc length value of the rest of the field of view is the same as the calculation method of the arc length value between a and b, which will not be repeated here. In addition to calculating the arc length value, the embodiment of the present application can also use the method of calculating the chord length value to determine the shooting range, and this application does not limit this.

[0168] In an optional embodiment, the shooting angle of each image perception device is determined based on the extreme point cloud data and the installation angle θ of each image perception device relative to the scanner, and the overlapping shooting angles of the two adjacent image perception devices are determined based on the intersection point cloud data of the shooting ranges of the two adjacent image perception devices. Based on the shooting angle of each image perception device and the corresponding overlapping shooting angle, the degree of overlap between the shooting range of any image perception device and the adjacent image perception device is determined.

[0169] The camera array-based dynamic tunnel image stitching method provided in the embodiment of the present application determines the center coordinate system of the target tunnel section based on the fitted straight line of the target tunnel section, determines the limit point cloud data corresponding to the shooting range of each image perception device based on the contour point cloud data of the target tunnel section, converts the limit point cloud data to the center coordinate system based on the limit point cloud data and the parameters of the corresponding image perception device, and calculates the shooting range of each image perception device based on the limit point cloud data in the center coordinate system. The method of the embodiment of the present application calculates the shooting range of each image perception device through coordinate system conversion, so as to subsequently accurately calculate the overlap of the shooting ranges of the two image perception devices, thereby facilitating the image stitching to obtain a better tunnel cross-section image.

[0170] Based on the above embodiments, the present application also provides a flowchart of a method for dynamic splicing of tunnel images based on a camera array. Figure 11 A flow chart of the seventh method for dynamic tunnel image stitching based on a camera array provided in an embodiment of the present application is shown in FIG. Figure 11 As shown, the above S30 includes:

[0171] S31: performing pixel cropping on the corresponding two target tunnel lining images in the width direction according to the overlap between any image sensing device and an adjacent image sensing device to obtain two cropped target tunnel lining images.

[0172] Specifically, by Figure 10 It can be seen from the shooting range and the corresponding overlap calculation method shown that since the overlap calculation method adopted in the embodiment of the present application does not calculate the percentage of the overlapping field of view to the total field of view, but rather uses the percentage of the sum of half of the overlapping field of view and the non-overlapping field of view to the total field of view to determine the overlap, the two target tunnel images corresponding to the two image sensing devices can be directly cropped in the width direction according to the overlap to obtain two cropped target tunnel lining images.

[0173] S32: Pixel shifting is performed on the two cropped target tunnel lining images in the height direction to obtain two aligned target tunnel lining images.

[0174] Specifically, to ensure the continuity of the same object on the tunnel wall after splicing, it is necessary to perform pixel shifting in the height direction on the two cropped target tunnel lining images so that the same object on the tunnel wall is aligned with each other, thereby obtaining two aligned target tunnel lining images.

[0175] S33: Crop the two aligned target tunnel lining images within the target area to obtain a tunnel cross-section image of each target tunnel cross-section.

[0176] Specifically, as shown in FIG8( d ) above, the two target tunnel lining images after spatial resolution unification have different sizes after spatial resolution unification. It is necessary to crop the two aligned target tunnel lining images within the target area to obtain a tunnel cross-section image of each target tunnel section. The target area is the effective area in the target tunnel lining image. Generally, the size of the target tunnel lining image with the smallest size is used as the target area, and the other target tunnel lining images are cropped.

[0177] For example, Figure 12 A schematic diagram of tunnel image stitching provided in an embodiment of the present application is shown in FIG. Figure 12 As shown in Figure 1, for two adjacent target tunnel lining images, pixel cropping is first performed in the width direction, with an overlap of 0.85 in the figure, and then pixel shifting is performed in the height direction. Finally, the aligned target tunnel lining images are cropped to obtain the tunnel cross-section image of each target tunnel section.

[0178] The camera array-based dynamic tunnel image stitching method provided in an embodiment of the present application performs pixel cropping on two corresponding target tunnel lining images in the width direction, based on the degree of overlap between any image sensing device and its adjacent image sensing device, to obtain two cropped target tunnel lining images. These cropped images are then pixel-shifted in the height direction to obtain two aligned target tunnel lining images. These aligned images are then cropped within the target region to obtain a tunnel cross-sectional image for each target tunnel section. The method provided in an embodiment of the present application, by performing pixel cropping, pixel shifting, and region cropping on the target tunnel lining images, ensures that the resulting tunnel cross-sectional images are of uniform size and continuous, thereby improving the stitching effect of the tunnel cross-sectional images.

[0179] Based on the above embodiments, this embodiment further provides a method for dynamic tunnel image stitching based on a camera array. Before unifying the spatial resolution of two tunnel lining images, the tunnel lining images captured by the image sensing device may suffer from image distortion and contour deformation, so correction is required. For example, image correction methods include, but are not limited to, bilinear interpolation, the nearest neighbor method, or cubic interpolation. This embodiment uses bilinear interpolation for illustration.

[0180] Bilinear interpolation is used to correct the image's outer contour deformation and correct the image into a rectangle. Specifically, interpolation is performed from left to right (or from top to bottom) along the width of the image. When interpolating from left to right, the interpolation ratio value is greater than 1 and gradually increases. Figure 13 A schematic diagram of a bilinear interpolation method provided in an embodiment of the present application is shown in FIG. Figure 13As shown in the figure, if we want to get the pixel value of the unknown function I at point P = (x, y), assuming that we know the values of the function I at four points a = (x1, y1), b = (x2, y1), c = (x1, y2), and d = (x2, y2), the bilinear interpolation method first performs linear interpolation in the x direction, and the formula is as follows:

[0181]

[0182]

[0183] Then interpolate in the y direction, the formula is as follows:

[0184]

[0185] The pixel coordinates of the newly added interpolation point P are obtained through interpolation calculation in two directions.

[0186] Figure 14 The embodiment of the present application provides an effect diagram of a correction using bilinear interpolation method, such as Figure 14 As shown, the tunnel lining image is interpolated from left to right to correct the tunnel lining image. The image can be corrected through the embodiment of the present application to ensure the accuracy of the image.

[0187] Figure 15 A schematic diagram of a tunnel splicing image provided in an embodiment of the present application is shown as follows: Figure 15 As shown, the tunnel image of a section of the tunnel is obtained by processing multiple groups of tunnel lining images of multiple target tunnel sections of the left and right lanes of the tunnel using the dynamic tunnel image stitching method based on the camera array provided in an embodiment of the present application. When the millimeter pixel value is unified, the location and severity of damage such as cracks in the tunnel can be analyzed based on the stitched tunnel image. Cracks on the tunnel wall will appear as dark linear patterns at the corresponding positions in the tunnel image.

[0188] Based on any of the above embodiments, the present application also provides a tunnel image dynamic splicing device based on a camera array. Figure 16 A schematic diagram of a tunnel image dynamic splicing device based on a camera array provided in an embodiment of the present application is shown in FIG. Figure 16 As shown, the device includes:

[0189] An image acquisition module 10 is configured to acquire multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, and the multiple tunnel lining images are images of each target tunnel section at different angles acquired by a camera array composed of multiple image sensing devices;

[0190] A resolution processing module 20 is used to perform spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution;

[0191] The cross-sectional image stitching module 30 is configured to stitch together a plurality of target tunnel lining images based on the overlap between the shooting range of any image sensing device and an adjacent image sensing device among the plurality of image sensing devices, to obtain a tunnel cross-sectional image of each target tunnel cross-section;

[0192] The tunnel image stitching module 40 is used to combine tunnel cross-section images of multiple target tunnel cross-sections to obtain a tunnel image.

[0193] Optionally, before the image acquisition module 10, the device further includes:

[0194] A section position acquisition module is used to obtain the positions of multiple tunnel sections;

[0195] An interval division module, for dividing multiple tunnel sections into multiple intervals according to the positions of the multiple tunnel sections;

[0196] Section screening module, used to screen the standard tunnel section of each section;

[0197] The section optimization module is used to optimize other tunnel sections based on the contour point cloud data of the standard tunnel section and the contour point cloud data of other tunnel sections in each interval to obtain the target tunnel section, wherein the contour point cloud data is the point cloud data in the scanning coordinate system obtained by the scanner.

[0198] Optional cross-section screening module, including:

[0199] A symmetry calculation unit is used to calculate the contour symmetry of each tunnel section based on the contour point cloud data;

[0200] A smoothness calculation unit, used for calculating the contour smoothness of each tunnel section based on the contour point cloud data;

[0201] A continuity calculation unit, used to calculate the contour continuity of each tunnel section based on the contour point cloud data;

[0202] The section screening unit is used to screen standard tunnel sections according to the weighted average of profile symmetry, profile smoothness and profile continuity.

[0203] Optional, smoothness calculation unit, including:

[0204] A contour filtering subunit is used to perform contour filtering on each tunnel section based on the contour point cloud data;

[0205] The smoothness calculation subunit is used to calculate the difference in the profile of each tunnel section before and after filtering and determine the profile smoothness.

[0206] Optionally, the resolution processing module 20 includes:

[0207] A resolution acquisition unit for acquiring the spatial resolution of multiple tunnel lining images;

[0208] a standard resolution determination unit, configured to determine a standard spatial resolution based on the spatial resolutions of a plurality of tunnel lining images;

[0209] The resolution unification unit is used to modify the spatial resolution of multiple tunnel lining images according to the standard spatial resolution to obtain multiple target tunnel lining images with unified resolution.

[0210] Optionally, before the cross-sectional image stitching module 30, the device further includes:

[0211] The overlap calculation module is used to calculate the overlap between the shooting range of any image sensing device and the shooting range of the adjacent image sensing device based on the shooting ranges of multiple image sensing devices.

[0212] Optionally, before the overlap calculation module, the device further includes:

[0213] The limit point cloud data determination module is used to determine the limit point cloud data corresponding to the shooting range of each image perception device based on the contour point cloud data of the target tunnel section and the position parameters of the image perception sleeper. The limit point cloud data is the point cloud data of the left limit point and the right limit point of the limited shooting range;

[0214] The shooting range calculation module is used to calculate the shooting range of each image perception device based on the extreme point cloud data.

[0215] Optionally, the cross-sectional image stitching module 30 includes:

[0216] a pixel cropping unit, configured to crop the pixels of the corresponding two target tunnel lining images in the width direction according to the overlap between any image sensing device and an adjacent image sensing device, thereby obtaining two cropped target tunnel lining images;

[0217] A pixel shifting unit is used to shift the pixels of the two cropped target tunnel lining images in the height direction to obtain the two aligned target tunnel lining images;

[0218] The image cropping unit is used to crop the two aligned target tunnel lining images within the target area to obtain a tunnel cross-section image of each target tunnel cross-section.

[0219] The above-mentioned device is used to execute the method provided in the above-mentioned embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0220] The above modules can be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by scheduling program code through a processing element, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0221] Figure 17 A schematic diagram of an electronic device provided in an embodiment of the present application, such as Figure 17 As shown, the electronic device 100 includes: a processor 101, a storage medium 102 and a bus.

[0222] The storage medium 102 stores program instructions executable by the processor 101. When the electronic device 100 is running, the processor 101 communicates with the storage medium 102 via a bus, and the processor 101 executes the program instructions to perform the above method embodiment. The specific implementation methods and technical effects are similar and will not be repeated here.

[0223] Optionally, the present invention further provides a program product, such as a computer-readable storage medium, comprising a program, which is used to perform the above method embodiment when executed by a processor.

[0224] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0225] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0226] In addition, the functional units in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional units.

[0227] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor (English: processor) to perform some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (English: Read-Only Memory, abbreviated: ROM), a random access memory (English: Random Access Memory, abbreviated: RAM), a magnetic disk or an optical disk, and other media that can store program code.

[0228] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A tunnel image dynamic stitching method based on camera array, characterized in that: The method comprises: Acquire multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, and the multiple tunnel lining images are images of each target tunnel section at different angles captured by a camera array composed of multiple image sensing devices; performing spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution; splicing the multiple target tunnel lining images based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and the shooting range of an adjacent image sensing device to obtain a tunnel cross-section image of each target tunnel cross-section; combining the tunnel cross-section images of the plurality of target tunnel cross-sections to obtain a tunnel image; Before acquiring multiple sets of tunnel lining images of multiple target tunnel sections, the method further includes: Obtain the locations of multiple tunnel sections; Dividing the plurality of tunnel sections into a plurality of intervals according to positions of the plurality of tunnel sections; Screening the standard tunnel section of each interval according to preset screening criteria, wherein the preset screening criteria include at least one of the following: profile symmetry and profile continuity; According to the contour symmetry and / or contour continuity of the contour point cloud data of the standard tunnel section in each interval, the contour symmetry and / or contour continuity of the contour point cloud data of other tunnel sections are optimized to obtain the target tunnel section, wherein the contour point cloud data is point cloud data in a scanning coordinate system obtained based on a scanner.

2. The method according to claim 1, characterized in that The screening of the standard tunnel section for each interval includes: Calculating the contour symmetry of each tunnel section based on the contour point cloud data; Calculating the contour smoothness of each tunnel section according to the contour point cloud data; Calculating the contour continuity of each tunnel section according to the contour point cloud data; The standard tunnel section is screened according to a weighted average of the profile symmetry, the profile smoothness, and the profile continuity.

3. The method according to claim 2, characterized in that Calculating the contour smoothness of each tunnel section according to the contour point cloud data includes: Performing contour filtering on each tunnel section according to the contour point cloud data; The difference in the profile of each tunnel section before and after filtering is calculated to determine the profile smoothness.

4. The method according to claim 1, wherein Before stitching the multiple target tunnel lining images based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and the shooting range of an adjacent image sensing device, the method further includes: According to the shooting ranges of the multiple image sensing devices, the overlap degree between the shooting range of any one of the image sensing devices and the shooting range of the adjacent image sensing device is calculated.

5. The method according to claim 4, characterized in that Before calculating the overlap between the shooting ranges of any one of the image sensing devices and the adjacent image sensing device based on the shooting ranges of the multiple image sensing devices, the method further includes: Determining, based on the contour point cloud data of the target tunnel section and the position parameters of the image sensing device, limit point cloud data corresponding to the shooting range of each image sensing device, the limit point cloud data being point cloud data of a left limit point and a right limit point that define the shooting range; The shooting range of each image perception device is calculated based on the extreme point cloud data.

6. The method according to claim 1, characterized in that The performing spatial resolution processing on the multiple tunnel lining images to obtain multiple target tunnel lining images with uniform spatial resolution includes: obtaining a spatial resolution of the plurality of tunnel lining images; determining a standard spatial resolution according to the spatial resolutions of the plurality of tunnel lining images; According to the standard spatial resolution, the spatial resolutions of the multiple tunnel lining images are modified to obtain the multiple target tunnel lining images with uniform resolution.

7. The method according to claim 1, characterized in that The step of stitching the multiple target tunnel lining images based on the overlap between the shooting range of any image sensing device among the multiple image sensing devices and the shooting range of an adjacent image sensing device to obtain a tunnel cross-section image of each target tunnel cross-section includes: According to the overlap between any one of the image sensing devices and the adjacent image sensing device, pixel cropping is performed on the corresponding two target tunnel lining images in the width direction to obtain two cropped target tunnel lining images; Pixel shifting is performed on the two cropped target tunnel lining images in a height direction to obtain two aligned target tunnel lining images; The two aligned target tunnel lining images are cropped within the target area to obtain a tunnel cross-section image of each target tunnel cross-section.

8. A tunnel image dynamic splicing device based on a camera array, characterized in that: The device comprises: An image acquisition module is configured to acquire multiple sets of tunnel lining images of multiple target tunnel sections, wherein each set of tunnel lining images includes multiple tunnel lining images, and the multiple tunnel lining images are images of each target tunnel section at different angles acquired by a camera array composed of multiple image sensing devices; a resolution processing module, configured to perform spatial resolution processing on the plurality of tunnel lining images to obtain a plurality of target tunnel lining images having uniform spatial resolution; a cross-sectional image stitching module, configured to stitch the plurality of target tunnel lining images together based on a degree of overlap between a shooting range of any image sensing device among the plurality of image sensing devices and a shooting range of an adjacent image sensing device, to obtain a tunnel cross-sectional image of each target tunnel cross-section; a tunnel image stitching module, configured to combine the tunnel cross-section images of the plurality of target tunnel cross-sections to obtain a tunnel image; The device further comprises: A section position acquisition module is used to obtain the positions of multiple tunnel sections; An interval division module, configured to divide the plurality of tunnel sections into a plurality of intervals according to the positions of the plurality of tunnel sections; A section screening module is used to screen the standard tunnel section of each interval according to a preset screening standard, wherein the preset screening standard includes at least one of the following: profile symmetry and profile continuity; A section optimization module is used to optimize the contour symmetry and / or contour continuity of the contour point cloud data of other tunnel sections based on the contour symmetry and / or contour continuity of the contour point cloud data of the standard tunnel section in each interval to obtain the target tunnel section, wherein the contour point cloud data is point cloud data in a scanning coordinate system obtained based on a scanner.

9. An electronic device, characterized in that: include: A processor, a storage medium, and a bus, wherein the storage medium stores program instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the program instructions to perform the steps of the method for dynamic tunnel image stitching based on a camera array according to any one of claims 1 to 7.

Images