Three-dimensional mural multi-spectral data acquisition method and device

Abstract

The present application relates to the technical field of spectral data acquisition, and provides a three-dimensional mural multi-spectral data acquisition method and device, wherein the acquisition range of the mural is established in a plane coordinate system according to the actual boundary of the mural; the dense point cloud data of the mural is obtained by scanning according to the acquisition range of the mural; the two-dimensional depth map of the mural is obtained according to the dense point cloud data and the acquisition range of the mural; the surface of the mural is segmented into multiple quasi-plane regions by using a clustering algorithm according to the two-dimensional depth map of the mural; the acquisition path and acquisition parameters are planned according to the multiple quasi-plane regions obtained by segmentation; the multi-spectral data of the mural in the corresponding quasi-plane region is respectively acquired according to the acquisition path and acquisition parameters, and the position information when the multi-spectral data is acquired is recorded; the multi-spectral data of the mural obtained by acquisition is spliced according to the position information when the multi-spectral data is acquired, and the complete mural multi-spectral data is obtained. The preset active focusing is realized, and the acquisition efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of spectral data acquisition technology, and in particular to a method and apparatus for acquiring multispectral data of three-dimensional murals. Background Technology

[0002] Multispectral imaging technology is one of the key conservation methods for ancient murals and other cultural relics. It can reveal information invisible to the human eye, such as pigment composition, painting techniques, and hidden defects.

[0003] However, in practical applications, for mural-type cultural relics with the two major characteristics of ultra-large format and non-planarity, the current use of multispectral imaging technology to collect data mainly relies on the following two methods:

[0004] One method is a "one-size-fits-all" scanning with a global constant focal length. This method will leave most areas of the mural out of focus, resulting in blurry images and low data quality.

[0005] Another type is real-time dynamic tracking scan. Although it can guarantee sharpness, the camera needs to make frequent, small-amplitude focus adjustments throughout the scanning process. This not only places extremely high demands on the hardware's response speed and stability, but also results in relatively low acquisition efficiency because the system is always "passively" adapting to surface changes.

[0006] In summary, existing methods for acquiring data on mural artifacts using multispectral imaging technology cannot guarantee data acquisition quality while avoiding the efficiency reduction caused by high-frequency and passive focusing during camera data acquisition. In particular, the lower the hardware response speed and the worse the stability, the lower the acquisition efficiency. Summary of the Invention

[0007] This invention provides a method and apparatus for acquiring multispectral data of three-dimensional murals, which solves the defects of existing technologies that cannot avoid the reduction in efficiency caused by high frequency and passive focus during the data acquisition process while ensuring data acquisition quality. It realizes the active setting of acquisition parameters and path planning, and automatically completes multispectral data acquisition according to the set acquisition parameters and path and regional unified parameters, effectively avoiding the defects of reduced efficiency caused by high frequency and passive focus, while ensuring data acquisition quality at the same time.

[0008] This invention provides a method for acquiring multispectral data of three-dimensional murals, comprising the following steps:

[0009] Based on the actual boundaries of the murals, the area for collecting the murals is established in a planar coordinate system;

[0010] Based on the area of ​​the mural collection, dense point cloud data of the mural was obtained by scanning;

[0011] Based on the dense point cloud data and the acquisition range of the mural, a two-dimensional depth map of the mural was obtained;

[0012] Based on the two-dimensional depth map of the mural, a clustering algorithm was used to segment the surface of the mural into multiple quasi-planar regions; each quasi-planar region has similar spatial pose and depth information.

[0013] Based on the multiple quasi-planar regions obtained from the segmentation, the acquisition path and acquisition parameters are planned; among them, the acquisition parameters include the camera focal length and camera pose, and the acquisition path traverses all quasi-planar regions;

[0014] Based on the acquisition path and acquisition parameters, multispectral data of the murals in the corresponding quasi-planar areas were acquired, and the location information when acquiring the multispectral data was recorded.

[0015] Based on the location information when collecting multispectral data, the collected multispectral data of the murals are stitched together to obtain complete multispectral data of the murals.

[0016] According to the present invention, a method for acquiring multispectral data of a three-dimensional mural is provided. Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions, including:

[0017] Based on the two-dimensional depth map of the mural, several candidate planes were determined; each candidate plane was determined by randomly selecting three points.

[0018] Based on the candidate planes and the preset tolerance range, calculate the number of interior points for each candidate plane; wherein the distance between the interior points and the candidate planes is within the tolerance range;

[0019] Based on the number of interior points of each candidate plane, the candidate plane with the largest number of interior points is selected as a quasi-plane region;

[0020] Based on the selected quasi-plane regions, remove the points corresponding to the quasi-plane regions from the two-dimensional depth map of the mural; until all quasi-plane regions are found.

[0021] According to the present invention, a method for acquiring multispectral data of three-dimensional murals is provided, which plans acquisition paths and acquisition parameters based on multiple quasi-planar regions obtained by segmentation, including: planning acquisition paths and acquisition parameters within quasi-planar regions, and planning acquisition paths and acquisition parameters between quasi-planar regions.

[0022] According to the present invention, a method for acquiring multispectral data of three-dimensional murals is provided, which includes planning acquisition paths between quasi-planar regions, comprising:

[0023] For each quasi-plane region, calculate the geometric center point or feature point of all quasi-plane regions;

[0024] Based on the geometric center points or feature points of all quasi-plane regions, a traveling salesman problem algorithm is used to plan the acquisition path; the acquisition path includes the acquisition order of each quasi-plane region.

[0025] According to the present invention, a method for acquiring multispectral data of three-dimensional murals includes planning an acquisition path within a quasi-planar region, comprising:

[0026] Calculate the average depth and normal direction of the quasi-plane region based on the quasi-plane region;

[0027] Based on the average depth and normal direction of the quasi-plane region, a scan path covering the quasi-plane region is generated.

[0028] According to the present invention, a method for acquiring multispectral data of a three-dimensional mural is provided, wherein the scanning path covering the quasi-planar area includes a raster path and a serpentine path.

[0029] The method for acquiring multispectral data of a three-dimensional mural according to the present invention further includes: setting acquisition parameters for the corresponding quasi-planar region.

[0030] The present invention also provides a three-dimensional mural multispectral data acquisition device, comprising the following modules:

[0031] The planar coordinate module is used to establish the acquisition range of the mural in a planar coordinate system based on the actual boundaries of the mural.

[0032] The point cloud data module is used to scan and obtain dense point cloud data of the mural based on the acquisition range of the mural;

[0033] The depth map module is used to obtain a two-dimensional depth map of the mural based on dense point cloud data and the acquisition range of the mural.

[0034] The planar segmentation module is used to segment the surface of the mural into multiple quasi-planar regions based on the two-dimensional depth map of the mural using a clustering algorithm; each quasi-planar region has similar spatial pose and depth information;

[0035] The planning and acquisition module is used to plan the acquisition path and acquisition parameters based on the multiple quasi-planar regions obtained by segmentation; the acquisition parameters include camera focal length and camera attitude, and the acquisition path traverses all quasi-planar regions;

[0036] The acquisition module is used to acquire multispectral data of the murals in the corresponding quasi-planar areas according to the acquisition path and acquisition parameters, and record the location information when acquiring multispectral data;

[0037] The data fusion module is used to stitch together the multispectral data of the murals based on the location information when the multispectral data was collected, so as to obtain complete multispectral data of the murals.

[0038] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the three-dimensional mural multispectral data acquisition methods described above.

[0039] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the three-dimensional mural multispectral data acquisition methods described above.

[0040] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the three-dimensional mural multispectral data acquisition methods described above.

[0041] This invention provides a method and apparatus for acquiring multispectral data of a three-dimensional mural. The method involves establishing the acquisition range of the mural in a planar coordinate system based on its actual boundaries; scanning the mural to obtain dense point cloud data based on this acquisition range; obtaining a two-dimensional depth map of the mural based on the dense point cloud data and the acquisition range; segmenting the mural surface into multiple quasi-planar regions using a clustering algorithm based on the two-dimensional depth map; planning acquisition paths and parameters based on these quasi-planar regions; acquiring multispectral data of the mural within the corresponding quasi-planar regions according to the acquisition paths and parameters, and recording the location information during data acquisition; and stitching the acquired multispectral data together to obtain complete multispectral data of the mural. This method enables proactive setting of acquisition parameters and path planning, and automatically completes multispectral data acquisition according to regionally unified parameters based on the set acquisition parameters and paths. This effectively avoids the efficiency reduction caused by high-frequency, passive focusing, while simultaneously ensuring data acquisition quality. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the multispectral data acquisition method for three-dimensional murals provided by the present invention.

[0044] Figure 2 This is a schematic diagram of the structure of the three-dimensional mural multispectral data acquisition device provided by the present invention.

[0045] Figure 3This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0047] The following is combined Figures 1 to 3 This invention describes a method for acquiring multispectral data of three-dimensional murals.

[0048] Figure 1 This is a flowchart illustrating the three-dimensional mural multispectral data acquisition method provided by the present invention, as shown below. Figure 1 As shown, the method includes the following:

[0049] Step 101: Based on the actual boundaries of the mural, establish the acquisition range of the mural in a plane coordinate system;

[0050] Step 102: Based on the collection area of ​​the mural, scan to obtain the dense point cloud data of the mural;

[0051] Step 103: Based on the dense point cloud data and the acquisition range of the mural, obtain the two-dimensional depth map of the mural;

[0052] Step 104: Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions; each quasi-planar region has similar spatial pose and depth information;

[0053] Step 105: Based on the multiple quasi-plane regions obtained from the segmentation, plan the acquisition path and acquisition parameters; among which, the acquisition parameters include the camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions;

[0054] Step 106: According to the acquisition path and acquisition parameters, acquire the multispectral data of the murals in the corresponding quasi-planar areas, and record the location information when acquiring the multispectral data;

[0055] Step 107: Based on the location information when collecting multispectral data, stitch together the collected multispectral data of the mural to obtain complete multispectral data of the mural.

[0056] Specifically, in step 101, the actual boundary of the mural is determined, the outer rectangle of the mural is defined as the scanning range, and the world coordinates of its four corner points are recorded. A two-dimensional working plane coordinate system is established with any one of the corners as the origin (0,0).

[0057] Specifically, in step 102, a lidar is used to perform a full-area scan of the mural within a preset scanning range to obtain dense point cloud data containing three-dimensional coordinate information (X,Y,Z).

[0058] Specifically, in step 103, the three-dimensional point cloud data is projected onto the two-dimensional working plane coordinate system to form a two-dimensional depth map with depth information; wherein, the depth information is Z-axis coordinate data.

[0059] Specifically, in step 105, the camera posture can also be the platform posture. Generally, the camera needs to be fixed on a mobile platform, and the control system collects the path to drive the mobile platform, thereby moving the camera and changing its posture.

[0060] Specifically, in step 106, when the camera moves from one quasi-plane region to the next, the system actively and in one go adjusts the camera's focal length and camera attitude according to the acquisition parameters to adapt to the geometric features of the new region and performs efficient scanning with constant parameters within that region.

[0061] In another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions, including:

[0062] Based on the two-dimensional depth map of the mural, several candidate planes were determined; each candidate plane was determined by randomly selecting three points.

[0063] Based on the candidate planes and the preset tolerance range, calculate the number of interior points for each candidate plane; wherein the distance between the interior points and the candidate planes is within the tolerance range;

[0064] Based on the number of interior points of each candidate plane, the candidate plane with the largest number of interior points is selected as a quasi-plane region;

[0065] Based on the selected quasi-plane regions, remove the points corresponding to the quasi-plane regions from the two-dimensional depth map of the mural; identify all quasi-plane regions.

[0066] In another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, the acquisition path and acquisition parameters are planned according to the multiple quasi-planar regions obtained by segmentation, including: planning the acquisition path and acquisition parameters within the quasi-planar region, and planning the acquisition path and acquisition parameters between the quasi-planar regions.

[0067] According to another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, planning the acquisition path between quasi-planar regions includes:

[0068] For each quasi-plane region, calculate the geometric center point or feature point of all quasi-plane regions;

[0069] Based on the geometric center points or feature points of all quasi-plane regions, a traveling salesman problem algorithm is used to plan the acquisition path; the acquisition path includes the acquisition order of each quasi-plane region.

[0070] Specifically, algorithms for solving the Traveling Salesman Problem include genetic algorithms and simulated annealing algorithms, which are used to calculate the shortest path to all geometric center points or feature points.

[0071] By employing the Traveling Salesman Problem algorithm for global path optimization, the overall movement distance of the mobile platform is minimized, the motion trajectory is smoother, frequent starts and stops and large changes in attitude are reduced, energy consumption and mechanical wear are lowered, path planning is more intelligent, and mechanical losses are reduced.

[0072] Meanwhile, parameter adjustments between regions are based on precise pre-calculation, making them more reliable than responses measured in real time.

[0073] According to another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, planning the acquisition path within the quasi-planar region includes:

[0074] Calculate the average depth and normal direction of the quasi-plane region based on the quasi-plane region;

[0075] Based on the average depth and normal direction of the quasi-plane region, a scan path covering the quasi-plane region is generated.

[0076] In another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, the scanning path covering the quasi-planar region includes a raster path and a serpentine path.

[0077] In another embodiment of the three-dimensional mural multispectral data acquisition method provided by the present invention, the acquisition path within the planned quasi-plane area is further included: setting acquisition parameters for the corresponding quasi-plane area, including the optimal focal length F and platform attitude θ required for the camera to work within the quasi-plane area.

[0078] The present invention provides a method for acquiring multispectral data of three-dimensional murals. Based on the actual boundaries of the mural, the acquisition range of the mural is established in a planar coordinate system. Based on the acquisition range, dense point cloud data of the mural is obtained by scanning. Based on the dense point cloud data and the acquisition range, a two-dimensional depth map of the mural is obtained. Based on the two-dimensional depth map, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions. Based on the segmented quasi-planar regions, acquisition paths and acquisition parameters are planned. Based on the acquisition paths and parameters, multispectral data of the mural within the corresponding quasi-planar regions is acquired, and the position information during acquisition is recorded. Based on the position information during acquisition, the acquired multispectral data of the mural is stitched together to obtain complete multispectral data of the mural.

[0079] The method of this invention decomposes complex surfaces into simple regions in advance, and the system can make parameter adjustments in one step between regions. This avoids continuous and high-frequency passive focusing during the scanning process, and transforms passive focusing into active setting of focusing parameters, which significantly reduces the focusing adjustment time and greatly improves the acquisition efficiency. In each quasi-plane region, the camera focal length and attitude remain constant, ensuring the uniformity and consistency of image quality within the region, and making the data quality more stable and reliable.

[0080] With a high degree of automation, the system reduces human intervention. The entire process, from 3D scanning and cluster analysis to path planning and execution, can be completed automatically, greatly reducing reliance on operator experience and achieving truly intelligent data acquisition.

[0081] The following describes the three-dimensional mural multispectral data acquisition device provided by the present invention. The three-dimensional mural multispectral data acquisition device described below can be referred to in correspondence with the three-dimensional mural multispectral data acquisition method described above.

[0082] Figure 2 This is a schematic diagram of the structure of the three-dimensional mural multispectral data acquisition device provided by the present invention, as shown below. Figure 2 As shown, the device includes the following:

[0083] The planar coordinate module 201 is used to establish the acquisition range of the mural in the planar coordinate system according to the actual boundary of the mural;

[0084] Point cloud data module 202 is used to scan and obtain dense point cloud data of the mural based on the acquisition range of the mural;

[0085] The depth map module 203 is used to obtain a two-dimensional depth map of the mural based on the dense point cloud data and the acquisition range of the mural.

[0086] The planar segmentation module 204 is used to segment the surface of the mural into multiple quasi-planar regions based on the two-dimensional depth map of the mural using a clustering algorithm; wherein each quasi-planar region has similar spatial pose and depth information;

[0087] The planning and acquisition module 205 is used to plan the acquisition path and acquisition parameters based on the multiple quasi-plane regions obtained by segmentation; wherein, the acquisition parameters include the camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions;

[0088] The acquisition module 206 is used to acquire multispectral data of the murals in the corresponding quasi-planar areas according to the acquisition path and acquisition parameters, and record the location information when acquiring multispectral data;

[0089] The data fusion module 207 is used to stitch together the multispectral data of the murals based on the location information when the multispectral data is collected, so as to obtain complete multispectral data of the murals.

[0090] The present invention provides a method for acquiring multispectral data of three-dimensional murals. Based on the actual boundaries of the mural, the acquisition range of the mural is established in a planar coordinate system. Based on the acquisition range, dense point cloud data of the mural is obtained by scanning. Based on the dense point cloud data and the acquisition range, a two-dimensional depth map of the mural is obtained. Based on the two-dimensional depth map, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions. Based on the segmented quasi-planar regions, acquisition paths and acquisition parameters are planned. Based on the acquisition paths and parameters, multispectral data of the mural within the corresponding quasi-planar regions is acquired, and the position information during acquisition is recorded. Based on the position information during acquisition, the acquired multispectral data of the mural is stitched together to obtain complete multispectral data of the mural.

[0091] The method of this invention decomposes complex surfaces into simple regions in advance, and the system can make parameter adjustments in one step between regions. This avoids continuous and high-frequency passive focusing during the scanning process, and transforms passive focusing into active setting of focusing parameters, which significantly reduces the focusing adjustment time and greatly improves the acquisition efficiency. In each quasi-plane region, the camera focal length and attitude remain constant, ensuring the uniformity and consistency of image quality within the region, and making the data quality more stable and reliable.

[0092] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute a three-dimensional stereoscopic mural multispectral data acquisition method, which includes:

[0093] Step 101: Based on the actual boundaries of the mural, establish the acquisition range of the mural in a plane coordinate system;

[0094] Step 102: Based on the collection area of ​​the mural, scan to obtain the dense point cloud data of the mural;

[0095] Step 103: Based on the dense point cloud data and the acquisition range of the mural, obtain the two-dimensional depth map of the mural;

[0096] Step 104: Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions; each quasi-planar region has similar spatial pose and depth information;

[0097] Step 105: Based on the multiple quasi-plane regions obtained from the segmentation, plan the acquisition path and acquisition parameters; among which, the acquisition parameters include the camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions;

[0098] Step 106: According to the acquisition path and acquisition parameters, acquire the multispectral data of the murals in the corresponding quasi-planar areas, and record the location information when acquiring the multispectral data;

[0099] Step 107: Based on the location information when collecting multispectral data, stitch together the collected multispectral data of the mural to obtain complete multispectral data of the mural.

[0100] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, 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 a 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 of 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.

[0101] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the three-dimensional stereoscopic mural multispectral data acquisition method provided by the above methods, the method comprising:

[0102] Step 101: Based on the actual boundaries of the mural, establish the acquisition range of the mural in a plane coordinate system;

[0103] Step 102: Based on the collection area of ​​the mural, scan to obtain the dense point cloud data of the mural;

[0104] Step 103: Based on the dense point cloud data and the acquisition range of the mural, obtain the two-dimensional depth map of the mural;

[0105] Step 104: Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions; each quasi-planar region has similar spatial pose and depth information;

[0106] Step 105: Based on the multiple quasi-plane regions obtained from the segmentation, plan the acquisition path and acquisition parameters; among which, the acquisition parameters include the camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions;

[0107] Step 106: According to the acquisition path and acquisition parameters, acquire the multispectral data of the murals in the corresponding quasi-planar areas, and record the location information when acquiring the multispectral data;

[0108] Step 107: Based on the location information when collecting multispectral data, stitch together the collected multispectral data of the mural to obtain complete multispectral data of the mural.

[0109] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the three-dimensional stereoscopic mural multispectral data acquisition method provided by the methods described above, the method comprising:

[0110] Step 101: Based on the actual boundaries of the mural, establish the acquisition range of the mural in a plane coordinate system;

[0111] Step 102: Based on the collection area of ​​the mural, scan to obtain the dense point cloud data of the mural;

[0112] Step 103: Based on the dense point cloud data and the acquisition range of the mural, obtain the two-dimensional depth map of the mural;

[0113] Step 104: Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions; each quasi-planar region has similar spatial pose and depth information;

[0114] Step 105: Based on the multiple quasi-plane regions obtained from the segmentation, plan the acquisition path and acquisition parameters; among which, the acquisition parameters include the camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions;

[0115] Step 106: According to the acquisition path and acquisition parameters, acquire the multispectral data of the murals in the corresponding quasi-planar areas, and record the location information when acquiring the multispectral data;

[0116] Step 107: Based on the location information when collecting multispectral data, stitch together the collected multispectral data of the mural to obtain complete multispectral data of the mural.

[0117] The device embodiments described above are merely illustrative. 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 the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of the embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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 method for acquiring multispectral data of three-dimensional murals, characterized in that, include: Based on the actual boundaries of the murals, the area for collecting the murals is established in a planar coordinate system; Based on the acquisition range of the mural, dense point cloud data of the mural was obtained by scanning; Based on the dense point cloud data and the acquisition range of the mural, a two-dimensional depth map of the mural is obtained; Based on the two-dimensional depth map of the mural, a clustering algorithm is used to segment the surface of the mural into multiple quasi-planar regions; wherein each quasi-planar region has similar spatial pose and depth information; Based on the multiple quasi-planar regions obtained from the segmentation, a data acquisition path and acquisition parameters are planned; wherein, the acquisition parameters include camera focal length and camera pose, and the acquisition path traverses all quasi-planar regions; According to the acquisition path and the acquisition parameters, multispectral data of the murals in the corresponding quasi-planar areas are acquired respectively, and the location information when acquiring multispectral data is recorded. Based on the location information when the multispectral data was collected, the multispectral data of the mural was stitched together to obtain complete multispectral data of the mural. The step of segmenting the surface of the mural into multiple quasi-planar regions using a clustering algorithm based on the two-dimensional depth map of the mural includes: Based on the two-dimensional depth map of the mural, several candidate planes are determined; each candidate plane is determined by randomly selecting three points. Based on the candidate planes and a preset tolerance range, the number of interior points of each candidate plane is calculated; wherein the distance between the interior points and the candidate planes is within the tolerance range; Based on the number of interior points of each candidate plane, the candidate plane with the largest number of interior points is selected as a quasi-plane region; Based on the selected quasi-plane region, remove the points corresponding to the quasi-plane region from the two-dimensional depth map of the mural; identify all quasi-plane regions; The step of planning acquisition paths and acquisition parameters based on the multiple quasi-plane regions obtained by segmentation includes: planning acquisition paths and acquisition parameters within quasi-plane regions, and planning acquisition paths and acquisition parameters between quasi-plane regions; The acquisition paths between the planned quasi-planar regions include: For each of the quasi-plane regions, calculate the geometric center point or feature point of all the quasi-plane regions; Based on the geometric center points or feature points of all the quasi-planar regions, a genetic algorithm or simulated annealing algorithm is used to plan the acquisition path; wherein, the acquisition path includes the acquisition order of each quasi-planar region; The acquisition path within the planned quasi-planar area includes: Based on the quasi-plane region, calculate the average depth and normal direction of the quasi-plane region; Based on the average depth and normal direction of the quasi-plane region, a scan path covering the quasi-plane region is generated; The method for acquiring multispectral data of three-dimensional murals further includes: setting acquisition parameters corresponding to the quasi-planar region; When the camera moves from one quasi-plane region to the next, the system actively and in one go adjusts the camera's focal length and camera attitude according to the acquired parameters to adapt to the geometric features of the new region.

2. The method for acquiring multispectral data of three-dimensional murals according to claim 1, characterized in that, The scanning path covering the quasi-planar region includes a raster path and a serpentine path.

3. A three-dimensional mural multispectral data acquisition device, characterized in that, include: The planar coordinate module is used to establish the acquisition range of the mural in a planar coordinate system based on the actual boundaries of the mural. The point cloud data module is used to scan and obtain dense point cloud data of the mural based on the acquisition range of the mural. The depth map module is used to obtain a two-dimensional depth map of the mural based on the dense point cloud data and the acquisition range of the mural. The planar segmentation module is used to segment the surface of the mural into multiple quasi-planar regions based on the two-dimensional depth map of the mural using a clustering algorithm; wherein each quasi-planar region has similar spatial pose and depth information; The planning and acquisition module is used to plan the acquisition path and acquisition parameters based on the multiple segmented quasi-plane regions; wherein, the acquisition parameters include camera focal length and camera attitude, and the acquisition path traverses all quasi-plane regions; The acquisition module is used to acquire multispectral data of the murals in the corresponding quasi-planar areas according to the acquisition path and the acquisition parameters, and record the location information when acquiring the multispectral data; The data fusion module is used to stitch together the multispectral data of the murals based on the location information when the multispectral data was collected, so as to obtain complete multispectral data of the murals. The planar segmentation module is used to determine several candidate planes based on the two-dimensional depth map of the mural; each candidate plane is determined by randomly selecting three points; the number of interior points of each candidate plane is calculated based on the candidate planes and a preset tolerance range; the distance between the interior points and the candidate plane is within the tolerance range; the candidate plane with the most interior points is selected as a quasi-planar region based on the number of interior points of each candidate plane; the points corresponding to the selected quasi-planar region are removed from the two-dimensional depth map of the mural based on the selected quasi-planar region; and all quasi-planar regions are identified. The step of planning acquisition paths and acquisition parameters based on the multiple quasi-plane regions obtained by segmentation includes: planning acquisition paths and acquisition parameters within quasi-plane regions, and planning acquisition paths and acquisition parameters between quasi-plane regions; The acquisition paths between the planned quasi-planar regions include: For each of the quasi-plane regions, calculate the geometric center point or feature point of all the quasi-plane regions; Based on the geometric center points or feature points of all the quasi-planar regions, a genetic algorithm or simulated annealing algorithm is used to plan the acquisition path; wherein, the acquisition path includes the acquisition order of each quasi-planar region; The acquisition path within the planned quasi-planar area includes: Based on the quasi-plane region, calculate the average depth and normal direction of the quasi-plane region; Based on the average depth and normal direction of the quasi-plane region, a scan path covering the quasi-plane region is generated; The method for acquiring multispectral data of three-dimensional murals further includes: setting acquisition parameters corresponding to the quasi-planar region; when the camera moves from one quasi-planar region to the next, the system actively and in one go adjusts the camera focal length and camera attitude according to the acquisition parameters to adapt to the geometric features of the new region.

4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the three-dimensional mural multispectral data acquisition method as described in any one of claims 1 to 2.

5. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the three-dimensional mural multispectral data acquisition method as described in any one of claims 1 to 2.

Images