Cultural relic hyperspectral image reconstruction method and device, hyperspectral area-array camera and medium

Abstract

The invention provides a cultural relic hyperspectral image reconstruction method and device, a hyperspectral area-array camera and a medium, and relates to the technical field of hyperspectral imaging and cultural relic digital protection. RGB images of the surface of a cultural relic are collected under uniform and continuous spectral illumination; selecting a plurality of sampling points in the RGB image, and collecting hyperspectral data at the plurality of sampling points; performing spatial registration on the RGB image and the hyperspectral data, and establishing a corresponding relation between pixel coordinates and spectral coordinates; and according to a corresponding relation between pixel coordinates and spectrum coordinates, carrying out fusion on spatial information of the RGB image and the hyperspectral data, combining a fusion result with prior constraints of a preset cultural relic pigment standard spectrum database to reconstruct a complete hyperspectral image of the surface of the cultural relic, and rapidly generating a complete hyperspectral area array image under the condition of a small amount of sampling. The device has the advantages of light weight, efficient collection and credible result, and can be widely applied to cultural relic protection scenes such as murals, paintings and calligraphies, ceramics and textiles.

Description

Technical Field

[0001] This invention relates to the fields of hyperspectral imaging and digital preservation of cultural relics, and in particular to a method, apparatus, hyperspectral area array camera and medium for reconstructing hyperspectral images of cultural relics using a hyperspectral area array camera. Background Technology

[0002] Hyperspectral imaging technology, as a non-destructive testing method combining imaging and spectral techniques, has shown great application potential in the fields of cultural relic preservation and scientific research. It can simultaneously acquire spatial and continuous spectral information from the surface of cultural relics, enabling precise analysis of pigment types, aging conditions, repair traces, and authenticity. Traditional hyperspectral imaging technologies used in cultural relic preservation primarily rely on scanning-based imaging hyperspectral cameras. These devices construct spectral data cubes through linear or dot-array scanning, achieving high spectral resolution spatial distribution information. However, their inherent scanning mechanism leads to low data acquisition efficiency and a long time required to complete a single image. Furthermore, these systems are typically bulky, complex, and expensive, and are extremely sensitive to environmental vibrations and light source stability. These drawbacks limit their widespread application in cultural relic preservation scenarios such as museums and archaeological sites where rapid and flexible operations are required. Among related technologies, point spectrometers are used for hyperspectral imaging of cultural relics. These devices acquire reflectance spectral data at specific points using fiber optic probes, offering advantages such as portability, flexibility, and high spectral resolution. However, its essential drawback is that it can only obtain spectral information of discrete points and lacks complete spatial information. Summary of the Invention

[0003] This invention provides a method, apparatus, hyperspectral array camera and medium for reconstructing hyperspectral images of cultural relics, which solves the defects of traditional hyperspectral imaging equipment, such as long imaging time or only being able to acquire spectral information of discrete points and lacking complete spatial information.

[0004] This invention provides a method for reconstructing hyperspectral images of cultural relics, comprising: RGB images of the artifact surface were acquired under uniform continuous spectrum illumination. Multiple sampling points are selected in the RGB image, and hyperspectral data at the multiple sampling points are acquired; Spatial registration is performed on the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates; The spatial information of the RGB image and the hyperspectral data are fused according to the correspondence between the pixel coordinates and the spectral coordinates. The fusion result is combined with the prior constraints of the preset cultural relic pigment standard spectral database to reconstruct a complete hyperspectral image of the cultural relic surface.

[0005] The method for reconstructing hyperspectral images of cultural relics according to the present invention further includes, after reconstructing a complete hyperspectral image of the surface of the cultural relic: The spectral curves of each pixel in the complete hyperspectral image are matched with the preset cultural relic pigment standard spectral database to output the pigment type and distribution map on the surface of the cultural relic.

[0006] According to the method for reconstructing cultural relics from hyperspectral images provided by the present invention, the step of spatially registering the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates includes: Obtain the initial pixel coordinates of the sampling point on the RGB image; Within the preset coordinate range of the initial pixel coordinates, by fitting discrete pixel similarity values, an extreme point in a continuous space is found, and the coordinates of the extreme point are used as the precise pixel coordinates of the sampling point; A correspondence between pixel coordinates and spectral coordinates is established based on the precise pixel coordinates and the hyperspectral data.

[0007] According to the method for reconstructing hyperspectral images of cultural relics provided by the present invention, the step of fusing the spatial information of the RGB image and the hyperspectral data based on the correspondence between the pixel coordinates and the spectral coordinates includes: Extract the multi-scale spatial features of the RGB image to obtain a spatial feature map; Based on the correspondence between pixel coordinates and spectral coordinates, the hyperspectral data is embedded as a high-precision spectral monitoring signal into the corresponding coordinate position in the spatial feature map to generate a spatial feature map with spectral monitoring signal. A fusion network is constructed, and the spatial feature map with the spectral supervision signal is input into the fusion network to output the fusion result.

[0008] According to the hyperspectral image reconstruction method for cultural relics provided by the present invention, the fusion result includes the initial estimated spectrum of each pixel in the RGB image, and the step of reconstructing a complete hyperspectral image of the cultural relic surface by combining the fusion result with prior constraints of a preset standard spectral database of cultural relic pigments includes: The initial estimated spectrum of each pixel in the RGB image is linearly combined and constrained by the prior constraints of the preset cultural relic pigment standard spectral database. The results of the linear combination constraint fitting are optimized based on the joint optimization function to obtain a complete hyperspectral image of the artifact surface.

[0009] According to the method for reconstructing hyperspectral images of cultural relics provided by the present invention, the joint optimization function includes: a first loss term based on hyperspectral data supervision, a second loss term based on spectral library prior constraints, and a third loss term based on spatial smoothing regularization.

[0010] According to the method for reconstructing hyperspectral images of cultural relics provided by the present invention, the fusion network is a deep learning model, which is trained based on RGB images and point spectral supervision signals.

[0011] This invention provides a device for hyperspectral image reconstruction of cultural relics, comprising: The first acquisition module is used to acquire RGB images of the surface of cultural relics under uniform continuous spectrum illumination. The second acquisition module is used to select multiple sampling points in the RGB image and acquire hyperspectral data at the sampling points; A module is established to perform spatial registration of the RGB image and the hyperspectral data, and to establish the correspondence between pixel coordinates and spectral coordinates; The reconstruction module is used to fuse the spatial information of the RGB image and the hyperspectral data according to the correspondence between the pixel coordinates and the spectral coordinates, and to reconstruct a complete hyperspectral image of the surface of the cultural relic by combining the fusion result with the prior constraints of the preset cultural relic pigment standard spectral database.

[0012] The present invention also provides a hyperspectral area array camera, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the hyperspectral image reconstruction method for cultural relics as described in any of the preceding claims.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for reconstructing hyperspectral images of cultural relics as described above.

[0014] The present invention provides a method, apparatus, hyperspectral area array camera, and medium for reconstructing hyperspectral images of cultural relics. This method involves acquiring RGB images of the cultural relic surface under uniform continuous spectral illumination; selecting multiple sampling points in the RGB images and acquiring hyperspectral data at these points; spatially registering the RGB images and hyperspectral data to establish a correspondence between pixel coordinates and spectral coordinates; fusing the spatial information of the RGB images and the hyperspectral data based on this correspondence; and reconstructing a complete hyperspectral image of the cultural relic surface by combining the fusion result with prior constraints from a pre-defined standard spectral database of cultural relic pigments. This method rapidly generates complete hyperspectral area array images with minimal non-destructive sampling, offering advantages such as lightweight equipment, high acquisition efficiency, and reliable results. It can be widely applied to the preservation of cultural relics such as murals, paintings, ceramics, and textiles. Attached Figure Description

[0015] 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.

[0016] Figure 1 This is a flowchart of the hyperspectral image reconstruction method for cultural relics provided in an embodiment of the present invention; Figure 2 This is a functional structure diagram of the hyperspectral image reconstruction device for cultural relics provided in an embodiment of the present invention; Figure 3 This is a functional structure diagram of a hyperspectral area array camera provided in an embodiment of the present invention. Detailed Implementation

[0017] 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.

[0018] Figure 1 A functional structure diagram of the hyperspectral image reconstruction method for cultural relics provided in the embodiments of the present invention is shown below. Figure 1 As shown, the method for reconstructing hyperspectral images of cultural relics provided in this embodiment of the invention includes: Step 101: Under uniform continuous spectrum illumination, acquire RGB images of the artifact surface; In this embodiment of the invention, the halogen light source module is activated to provide uniform and continuous spectral illumination, thereby acquiring RGB images of the artifact surface.

[0019] Step 102: Select multiple sampling points in the RGB image and collect hyperspectral data at the multiple sampling points; Step 103: Spatial registration is performed on the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates; In this embodiment of the invention, spatial coordinate consistency is established by geometrically registering the point spectrum sampling position with the RGB image, ensuring that the spectral value corresponds one-to-one with the image pixel.

[0020] Step 104: Based on the correspondence between the pixel coordinates and the spectral coordinates, the spatial information of the RGB image and the hyperspectral data are fused, and the fusion result is combined with the prior constraints of the preset cultural relic pigment standard spectral database to reconstruct a complete hyperspectral image of the cultural relic surface.

[0021] Traditional methods for reconstructing hyperspectral images of cultural relics rely on scanning-based imaging hyperspectral devices. These devices construct spectral data cubes through linear or dot-array scanning, achieving high spectral resolution spatial distribution information. However, their inherent scanning mechanism leads to low data acquisition efficiency, resulting in lengthy processing times for completing a single image. Furthermore, these systems are typically bulky, complex, and expensive, and are extremely sensitive to environmental vibrations and light source stability. These drawbacks limit their widespread application in cultural relic preservation scenarios such as museums and archaeological sites where rapid and flexible operations are required. Related technologies utilize point spectrometers for hyperspectral imaging of cultural relics. These devices acquire reflectance spectral data at specific points using fiber optic probes, offering advantages such as portability, flexibility, and high spectral resolution. However, their inherent limitation lies in acquiring only discrete point spectral information, lacking complete spatial information.

[0022] The hyperspectral image reconstruction method for cultural relics provided in this invention involves acquiring an RGB image of the cultural relic surface under uniform continuous spectral illumination; selecting multiple sampling points in the RGB image and acquiring hyperspectral data at these sampling points; spatially registering the RGB image and the hyperspectral data to establish a correspondence between pixel coordinates and spectral coordinates; fusing the spatial information of the RGB image and the hyperspectral data based on the correspondence between pixel coordinates and spectral coordinates; and reconstructing a complete hyperspectral image of the cultural relic surface by combining the fusion result with prior constraints from a preset standard spectral database of cultural relic pigments. This method rapidly generates a complete hyperspectral area array image under conditions of minimal non-destructive sampling, offering advantages such as lightweight equipment, high acquisition efficiency, and reliable results. It can be widely applied to the preservation of cultural relics such as murals, paintings, ceramics, and textiles.

[0023] Based on any of the above embodiments, after reconstructing the complete hyperspectral image of the artifact surface, the method further includes: The spectral curves of each pixel in the complete hyperspectral image are matched with the preset cultural relic pigment standard spectral database to output the pigment type and distribution map on the surface of the cultural relic.

[0024] In this embodiment of the invention, the pixel-level reconstructed spectrum is matched with a standard spectral library to output the pigment category and distribution map, and the standard spectral library is used to complete automatic pigment identification, aging analysis and repair assessment.

[0025] Based on any of the above embodiments, the step of spatially registering the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates includes: Step 201: Obtain the initial pixel coordinates of the sampling point on the RGB image; Step 202: Within the preset coordinate range of the initial pixel coordinates, find an extreme point in a continuous space by fitting discrete pixel similarity values, and use the coordinates of the extreme point as the precise pixel coordinates of the sampling point; Step 203: Establish the correspondence between pixel coordinates and spectral coordinates based on the precise pixel coordinates and the hyperspectral data.

[0026] This invention significantly compresses the computational scope by acquiring initial pixel coordinates and limiting the search to a preset coordinate range, avoiding the enormous computational overhead of global searches, thereby significantly improving registration efficiency, especially suitable for efficient processing of high-resolution cultural relic images. Employing a sub-pixel-level registration method based on fitting discrete pixel similarity values ​​to find continuous spatial extrema, it overcomes the precision limitations of traditional integer pixel coordinates, achieving precise coordinate positions up to the sub-pixel level. This improvement enables unprecedented precision in the spatial correspondence between hyperspectral data and RGB image pixels, fundamentally eliminating spectral spatial misalignment caused by registration errors. The spectral coordinate correspondence established based on this precise pixel coordinates lays a reliable spatial geometric foundation for subsequent multi-source data fusion and high-precision spectral reconstruction, ensuring that each spectral data point accurately corresponds to the correct physical position on the surface of the cultural relic during the fusion process, thus improving the overall geometric fidelity of hyperspectral image reconstruction.

[0027] Based on any of the above embodiments, the step of fusing the spatial information of the RGB image and the hyperspectral data according to the correspondence between the pixel coordinates and the spectral coordinates includes: Step 301: Extract the multi-scale spatial features of the RGB image to obtain a spatial feature map; Step 302: Based on the correspondence between the pixel coordinates and the spectral coordinates, the hyperspectral data is embedded as a high-precision spectral monitoring signal into the corresponding coordinate position in the spatial feature map to generate a spatial feature map with spectral monitoring signal; Step 303: Construct a fusion network by inputting the spatial feature map with the spectral supervision signal into the fusion network and outputting the fusion result.

[0028] In this embodiment of the invention, the fusion network is a deep learning model, which is trained based on RGB images and point spectral supervision signals.

[0029] In this embodiment of the invention, deep networks are used. Preliminary reconstruction results of the predicted hyperspectral image:

[0030] in, The preliminary reconstruction results of the hyperspectral image are as follows. : RGB image, H is the height of the hyperspectral image, W is the width of the hyperspectral image, and 3 is the number of channels of the hyperspectral image; : point hyperspectral data, B represents the number of spectral bands, with a total of N sampling points. For pixels, λ is the wavelength.

[0031] Based on any of the above embodiments, the fusion result includes the initial estimated spectrum of each pixel in the RGB image, and the step of reconstructing a complete hyperspectral image of the artifact surface by combining the fusion result with prior constraints of a preset cultural relic pigment standard spectral database includes: Step 401: Use the prior constraints of the preset cultural relic pigment standard spectral database to perform linear combination constraint fitting on the initial estimated spectrum of each pixel in the RGB image; Step 402: Optimize the result of the linear combination constraint fitting based on the joint optimization function to obtain a complete hyperspectral image of the artifact surface.

[0032] In this embodiment of the invention, the initial estimate will be... Further fitting of linear combinations of the spectral library: in

[0033] in, For combination coefficients, The a priori standard spectrum.

[0034] In this embodiment of the invention, the joint optimization function includes: a first loss term based on hyperspectral data supervision, a second loss term based on spectral library prior constraints, and a third loss term based on spatial smoothing regularization.

[0035] Let the joint optimization function be:

[0036] in: The first loss term based on hyperspectral data supervision is point spectral supervision, which requires the predicted value to be consistent with the true point spectrum at the point spectrum. The second loss term is based on the spectral library prior constraint: the spectral library constraint, which states that the full spectrum should be a combination fit close to the spectral library. The third loss term based on spatial smoothing regularization, namely the spatial smoothing / edge-preserving regularization term, guides image reconstruction to have spatial continuity or edge sharpness. For true point spectra, and To optimize the coefficients, Regularization terms for spatial smoothing / edge preservation.

[0037] The hyperspectral image reconstruction method for cultural relics provided in this invention utilizes a standard spectral database of pigments in cultural relics to perform linear combination constraint fitting on the initially estimated spectrum. This deeply embeds prior physical knowledge into the computational framework, forcing the reconstructed spectrum to conform to the spectral morphological characteristics of real pigments. This fundamentally avoids spectral reconstruction results with unreasonable physical meaning, significantly improving the interpretability and reliability of the reconstruction results. A joint optimization function incorporating multiple constraint factors is employed. The first loss term ensures that the reconstruction results at the point spectral sampling location are strictly consistent with the measured high-precision data. The second loss term guarantees that the spectral morphology of the entire image conforms to the physical prior of the pigment database. The third loss term introduces spatial context constraints to suppress noise and maintain the clarity of material boundaries. This achieves synergistic optimization driven by data and knowledge, effectively solving the overfitting or non-uniqueness problems that may result from a single supervisory signal. Through unique RGB+point spectrum fusion reconstruction and standard spectral prior constraints, a solid data foundation is provided for subsequent quantitative analyses such as pigment identification and aging assessment, significantly improving the scientific rigor and efficiency of pigment identification and aging assessment in cultural relics.

[0038] The artifact hyperspectral image reconstruction device provided by the present invention is described below. The artifact hyperspectral image reconstruction device described below and the artifact hyperspectral image reconstruction method described above can be referred to in correspondence.

[0039] Figure 2 A functional structural diagram of the cultural relic hyperspectral image reconstruction device provided in an embodiment of the present invention is shown below. Figure 2 As shown, the artifact hyperspectral image reconstruction device provided in this embodiment of the invention includes: The first acquisition module 201 is used to acquire RGB images of the surface of cultural relics under uniform continuous spectrum illumination. The second acquisition module 202 is used to select multiple sampling points in the RGB image and acquire hyperspectral data at the sampling points; Module 203 is used to perform spatial registration of the RGB image and the hyperspectral data, and to establish the correspondence between pixel coordinates and spectral coordinates; The reconstruction module 204 is used to fuse the spatial information of the RGB image and the hyperspectral data according to the correspondence between the pixel coordinates and the spectral coordinates, and to reconstruct a complete hyperspectral image of the surface of the cultural relic by combining the fusion result with the prior constraints of the preset cultural relic pigment standard spectral database.

[0040] The hyperspectral image reconstruction device for cultural relics provided in this invention acquires RGB images of the surface of cultural relics under uniform continuous spectral illumination; selects multiple sampling points in the RGB images and acquires hyperspectral data at the multiple sampling points; spatially registers the RGB images and the hyperspectral data to establish a correspondence between pixel coordinates and spectral coordinates; fuses the spatial information of the RGB images and the hyperspectral data according to the correspondence between pixel coordinates and spectral coordinates, and reconstructs a complete hyperspectral image of the surface of the cultural relics by combining the fusion result with the prior constraints of a preset standard spectral database of cultural relic pigments. It can quickly generate a complete hyperspectral area array image under conditions of minimal non-destructive sampling, and has the advantages of lightweight equipment, high acquisition efficiency, and reliable results. It can be widely applied to the protection of cultural relics such as murals, paintings, ceramics, and textiles.

[0041] Figure 3 An example is a schematic diagram of the physical structure of a hyperspectral area array camera, such as... Figure 3 As shown, the hyperspectral area array camera may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The memory 330 includes computer programs, an operating system, and acquired data. The processor 310 can call logical instructions in the memory 330 to execute a hyperspectral image reconstruction method for cultural relics. This method includes: acquiring an RGB image of the cultural relic surface under uniform continuous spectral illumination; selecting multiple sampling points in the RGB image and acquiring hyperspectral data at the multiple sampling points; spatially registering the RGB image and the hyperspectral data to establish a correspondence between pixel coordinates and spectral coordinates; fusing the spatial information of the RGB image and the hyperspectral data according to the correspondence between pixel coordinates and spectral coordinates; and reconstructing a complete hyperspectral image of the cultural relic surface by combining the fusion result with prior constraints from a preset cultural relic pigment standard spectral database.

[0042] Furthermore, the logical instructions in the aforementioned memory 330 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 related technologies, or a portion of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium.

[0043] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the method for reconstructing hyperspectral images of cultural relics provided by the above methods. The method includes: acquiring an RGB image of the surface of the cultural relic under uniform continuous spectral illumination; selecting multiple sampling points in the RGB image and acquiring hyperspectral data at the multiple sampling points; spatially registering the RGB image and the hyperspectral data to establish a correspondence between pixel coordinates and spectral coordinates; fusing the spatial information of the RGB image and the hyperspectral data according to the correspondence between pixel coordinates and spectral coordinates; and reconstructing a complete hyperspectral image of the surface of the cultural relic by combining the fusion result with prior constraints of a preset standard spectral database of cultural relic pigments.

[0044] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. 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.

[0045] 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 parts that contribute to the related technology, can be embodied in the form of software products. 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 described in the various embodiments or some parts of the embodiments.

[0046] 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 reconstructing hyperspectral images of cultural relics, characterized in that, include: RGB images of the artifact surface were acquired under uniform continuous spectrum illumination. Multiple sampling points are selected in the RGB image, and hyperspectral data at the multiple sampling points are acquired; Spatial registration is performed on the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates; The spatial information of the RGB image and the hyperspectral data are fused according to the correspondence between the pixel coordinates and the spectral coordinates. The fusion result is combined with the prior constraints of the preset cultural relic pigment standard spectral database to reconstruct a complete hyperspectral image of the cultural relic surface.

2. The method for reconstructing hyperspectral images of cultural relics according to claim 1, characterized in that, After reconstructing a complete hyperspectral image of the artifact's surface, the method further includes: The spectral curves of each pixel in the complete hyperspectral image are matched with the preset cultural relic pigment standard spectral database to output the pigment type and distribution map on the surface of the cultural relic.

3. The method for reconstructing hyperspectral images of cultural relics according to claim 1, characterized in that, The step of spatially registering the RGB image and the hyperspectral data to establish the correspondence between pixel coordinates and spectral coordinates includes: Obtain the initial pixel coordinates of the sampling point on the RGB image; Within the preset coordinate range of the initial pixel coordinates, by fitting discrete pixel similarity values, an extreme point in a continuous space is found, and the coordinates of the extreme point are used as the precise pixel coordinates of the sampling point; A correspondence between pixel coordinates and spectral coordinates is established based on the precise pixel coordinates and the hyperspectral data.

4. The method for reconstructing hyperspectral images of cultural relics according to claim 1, characterized in that, The step of fusing the spatial information of the RGB image and the hyperspectral data based on the correspondence between the pixel coordinates and the spectral coordinates includes: Extract the multi-scale spatial features of the RGB image to obtain a spatial feature map; Based on the correspondence between pixel coordinates and spectral coordinates, the hyperspectral data is embedded as a high-precision spectral monitoring signal into the corresponding coordinate position in the spatial feature map to generate a spatial feature map with spectral monitoring signal. A fusion network is constructed, and the spatial feature map with the spectral supervision signal is input into the fusion network to output the fusion result.

5. The method for reconstructing hyperspectral images of cultural relics according to claim 1 or 4, characterized in that, The fusion result includes the initial estimated spectrum of each pixel in the RGB image. The step of reconstructing a complete hyperspectral image of the artifact surface by combining the fusion result with prior constraints from a preset database of standard spectral data for artifact pigments includes: The initial estimated spectrum of each pixel in the RGB image is linearly combined and constrained by the prior constraints of the preset cultural relic pigment standard spectral database. The results of the linear combination constraint fitting are optimized based on the joint optimization function to obtain a complete hyperspectral image of the artifact surface.

6. The method for reconstructing hyperspectral images of cultural relics according to claim 5, characterized in that, The joint optimization function includes: a first loss term based on hyperspectral data supervision, a second loss term based on spectral library prior constraints, and a third loss term based on spatial smoothing regularization.

7. The method for reconstructing hyperspectral images of cultural relics according to claim 4, characterized in that, The fusion network is a deep learning model, which is trained based on RGB images and point spectral supervision signals.

8. A device for hyperspectral image reconstruction of cultural relics, characterized in that, include: The first acquisition module is used to acquire RGB images of the surface of cultural relics under uniform continuous spectrum illumination. The second acquisition module is used to select multiple sampling points in the RGB image and acquire hyperspectral data at the sampling points; A module is established to perform spatial registration of the RGB image and the hyperspectral data, and to establish the correspondence between pixel coordinates and spectral coordinates; The reconstruction module is used to fuse the spatial information of the RGB image and the hyperspectral data according to the correspondence between the pixel coordinates and the spectral coordinates, and to reconstruct a complete hyperspectral image of the surface of the cultural relic by combining the fusion result with the prior constraints of the preset cultural relic pigment standard spectral database.

9. A hyperspectral area array camera, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for reconstructing hyperspectral images of cultural relics as described in any one of claims 1 to 7.

10. A non-transitory readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for reconstructing hyperspectral images of cultural relics as described in any one of claims 1 to 7.

Images