Digital display method and system for art exhibition

Abstract

The invention discloses a digital display method and system for art exhibition, and relates to the technical field of digital display, and the method comprises the steps: generating a point cloud data set; performing grid reconstruction on the point cloud data set to generate a three-dimensional digital grid of the exhibit; performing material feature analysis on the three-dimensional digital grid of the exhibit, and extracting a material parameter set; collecting audience position coordinates, calculating an audience-exhibit relative azimuth angle based on the audience position coordinates, and generating a view angle transformation matrix; the view angle transformation matrix is applied to the exhibit three-dimensional digital grid, the material parameter set is superposed for rendering, an exhibit image after view angle transformation is output, and the exhibit image and the exhibition hall background image are synthesized into a mixed reality scene; and dynamically updating the visual angle transformation matrix based on the real-time change of the position coordinates of the audience, and regenerating the mixed reality scene. According to the invention, a multi-sensor cooperative acquisition scheme is adopted, so that the three-dimensional reconstruction precision is remarkably improved; and a multispectral structured light illumination technology is introduced, so that accurate acquisition of material characteristics is realized.

Description

Technical Field

[0001] The present invention relates to the field of digital display technology, and in particular to a digital display method and system for art exhibitions. Background Art

[0002] With the rapid development of digital technology, the field of art exhibitions is facing a major change in display methods. Traditional static display methods can no longer meet the audience's growing demand for interactive experience. In recent years, with the development of 3D scanning technology, virtual reality technology and augmented reality technology, digital display methods have gradually been applied to the field of art exhibitions. However, the current digital display technology still has many shortcomings, mainly reflected in the low accuracy of digital acquisition of exhibits, poor material restoration effect, and poor interactive experience.

[0003] Although there are solutions in the prior art that use 3D scanning and virtual reality technology to display artworks, there are the following specific problems: first, the accuracy of the collected three-dimensional data of the exhibits is insufficient, and it is difficult to restore the details of the exhibits, especially when dealing with artworks with complex textures and fine structures, the collection quality is difficult to guarantee; second, the material rendering effect is not realistic enough, and the texture of the artwork cannot be accurately presented, resulting in a significant difference between the digital display effect and the real object; third, the perspective change response speed is slow and the system delay is large, which seriously affects the audience's immersive experience; fourth, the existing solutions generally lack in-depth analysis of the material characteristics of the exhibits, making it difficult to achieve realistic reproduction of the artwork. Summary of the invention

[0004] In view of the problems existing in the application of existing digital display technology in the field of art exhibitions, such as low acquisition accuracy, unrealistic material restoration, and poor interactive experience, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to achieve high-precision digital acquisition, realistic material rendering and smooth real-time interaction of art exhibits, so as to provide better digital display effects of artworks and audience experience.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a digital display method for art exhibitions, which includes scanning the surface of an exhibit with a depth camera to generate a point cloud data set; reconstructing the grid of the point cloud data set to generate a three-dimensional digital grid of the exhibit; performing material feature analysis on the three-dimensional digital grid of the exhibit to extract a material parameter set; collecting audience position coordinates, calculating the audience-exhibit relative azimuth based on the audience position coordinates, and generating a perspective transformation matrix; applying the perspective transformation matrix to the three-dimensional digital grid of the exhibit, superimposing the material parameter set for rendering, outputting an exhibit image after perspective transformation, and synthesizing the exhibit image with the exhibition hall background image into a mixed reality scene; based on the real-time changes in the audience position coordinates, dynamically updating the perspective transformation matrix to regenerate the mixed reality scene.

[0008] As a preferred solution of the digital display method for art exhibitions described in the present invention, the point cloud data set includes the spatial position coordinates, surface reflection intensity values ​​and normal vectors of multiple sampling points; the three-dimensional digital grid of the exhibit includes a grid vertex set, a topological connection matrix and a vertex normal vector set; the material parameter set includes an ambient light coefficient, a diffuse reflection coefficient, a specular reflection coefficient and a texture mapping matrix.

[0009] As a preferred solution of the digital display method for art exhibitions described in the present invention, generating a point cloud data set includes the following steps: arranging a plurality of depth cameras, the depth cameras are arranged in an array, and the angles between the depth cameras are adjustable to form an expandable annular scanning array; placing the annular scanning array on a lifting bracket, and driving the annular scanning array to move in a vertical direction by the lifting bracket to form a spiral scanning trajectory; each depth camera emits infrared structured light, which is projected onto the surface of the exhibit to form a speckle pattern, and the deformation characteristics of the speckle pattern are collected; according to the deformation characteristics of the speckle pattern, the spatial position coordinates of each sampling point on the surface of the exhibit are calculated; by analyzing the near-infrared light intensity received at each sampling point, the reflection intensity value of the exhibit surface is obtained; based on the spatial position relationship between adjacent sampling points, the normal vector of the surface where each sampling point is located is fitted by the least squares method; the spatial position coordinates of each sampling point, the reflection intensity value of the exhibit surface and the normal vector are organized as structured data to generate a point cloud data set.

[0010] As a preferred solution of the digital display method for art exhibitions described in the present invention, generating a three-dimensional digital grid of an exhibit includes the following steps: preprocessing a point cloud data set, removing noise points, and performing spatial index encoding on the sampling points through octree spatial segmentation; taking the leaf node of the octree as the center, setting a search radius, extracting a local point set within the neighborhood of each leaf node, and calculating the covariance matrix of the local point set; performing eigenvalue decomposition on the covariance matrix to obtain a main eigenvector, and using the main eigenvector as an estimated value of the normal vector of the local surface; constructing an implicit function based on the distribution characteristics of the local point set, and obtaining the zero isosurface of the exhibit surface by solving the Poisson equation; uniformly sampling on the zero isosurface to generate a grid vertex set; using the Delaunay triangulation algorithm to connect the grid vertex set into a triangular grid, and establishing a topological connection matrix between the grid vertices; for each vertex of the triangular grid, calculating the vertex normal vector set based on the weighted average of the normal vectors of adjacent triangles.

[0011] As a preferred solution of the digital display method for art exhibitions described in the present invention, the extraction of material parameter sets includes the following steps: arranging a multi-light source array around the exhibit, the multi-light source array includes light source units of different wavelengths, and generating structured lighting by controlling the switching sequence of each light source unit; collecting reflected light intensity images of the three-dimensional digital grid of the exhibit under different lighting conditions, comparing the reflected light intensity images with the incident light intensity, and generating reflectivity data; spatially mapping the reflectivity data according to the grid vertex positions, constructing a reflectivity distribution map, and segmenting the material area of ​​the exhibit surface based on the reflectivity distribution map; for each material area, determining the ambient light coefficient by analyzing the reflectivity under dark light conditions; under diffuse light illumination conditions, analyzing the directional distribution of the reflected light intensity and calculating the diffuse reflection coefficient; using a high light illumination method, measuring the specular reflection peak and its attenuation law, and extracting the specular reflection coefficient; establishing a UV coordinate system based on the geometric characteristics of the exhibit surface, and optimizing and generating a texture mapping matrix through the minimum energy principle.

[0012] As a preferred solution of the digital display method for art exhibitions described in the present invention, generating a perspective transformation matrix includes the following steps: arranging an infrared depth camera array on the top of the exhibition hall, the infrared depth camera array includes multiple depth camera units, and the field of view of each depth camera unit overlaps with each other; acquiring a depth image through the infrared depth camera array, extracting the audience foreground target, and performing human key point detection on the foreground target; extracting the audience head feature points based on the detected human key points, and using the three-dimensional coordinates of the head feature points as the audience viewpoint position coordinates; establishing an exhibition hall world coordinate system, taking the exhibit center point as the origin, and calculating the vector between the audience viewpoint position coordinates and the exhibit center point as the observation vector; decomposing the observation vector into a horizontal azimuth and a vertical pitch angle; constructing a rotation matrix based on the horizontal azimuth and the vertical pitch angle, combining the rotation matrix with the translation matrix, and generating a perspective transformation matrix between the audience and the exhibit.

[0013] As a preferred solution of the digital display method for art exhibitions described in the present invention, the generation process of the mixed reality scene is as follows: the perspective transformation matrix is ​​applied to the vertex set of the three-dimensional digital grid of the exhibit, the spatial position of the vertices after the perspective transformation is calculated, and the transformed grid geometry structure is generated; based on the transformed grid geometry structure, a depth buffer is constructed to determine the visible triangle facet set of the grid surface, and the blocked triangle facets are removed; for each facet in the visible triangle facet set, a light transmission equation is calculated according to a material parameter set, wherein the ambient light coefficient, diffuse reflection coefficient and specular reflection coefficient determine the shading effect of the facet; a texture mapping matrix is ​​used to project the texture image onto the visible triangle facets; a real-time exhibition hall background image is collected, the lighting parameters of the exhibition hall are extracted, and the lighting parameters are applied to the exhibit image; the rendered exhibit image is alpha-blended with the exhibition hall background image to generate a mixed reality scene.

[0014] In a second aspect, an embodiment of the present invention provides a digital display system for art exhibitions, which includes a depth acquisition module for scanning the surface of an exhibit through a depth camera to generate a point cloud data set; a three-dimensional mesh reconstruction module for meshing the point cloud data set to generate a three-dimensional digital mesh of the exhibit; a material feature analysis module for performing material feature analysis on the three-dimensional digital mesh of the exhibit and extracting a material parameter set; an audience position tracking module for acquiring audience position coordinates, calculating the audience-exhibit relative azimuth based on the audience position coordinates, and generating a perspective transformation matrix; a mixed reality rendering module for applying the perspective transformation matrix to the three-dimensional digital mesh of the exhibit, superimposing the material parameter set for rendering, outputting an exhibit image after perspective transformation, and synthesizing the exhibit image with the exhibition hall background image into a mixed reality scene; and a dynamic scene update module for dynamically updating the perspective transformation matrix based on real-time changes in the audience position coordinates and regenerating the mixed reality scene.

[0015] The beneficial effects of the present invention are as follows: through the above technical solution, the present invention realizes high-precision digital acquisition, realistic rendering and smooth real-time interactive experience of art exhibits. Compared with the existing technology, it has the following advantages: first, the acquisition solution of multi-sensor collaboration is adopted to significantly improve the accuracy of three-dimensional reconstruction; second, the multi-spectral structured light illumination technology is innovatively introduced to achieve accurate acquisition of material properties; third, the improved visual tracking algorithm and real-time rendering optimization strategy are adopted to control the system response delay below the perceptible threshold of the human eye. Experimental results show that the present invention can meet the application requirements of most art exhibition scenes and has good practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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 description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flowchart of a digital display method for art exhibitions;

[0018] Figure 2 Flowchart for generating point cloud datasets for digital display methods of art exhibitions. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0022] Example 1, reference Figure 1-2 , which is the first embodiment of the present invention, and provides a digital display method for art exhibitions, as shown in the flow chart. Figure 1 As shown, including,

[0023] S1: Scan the surface of the exhibit with a depth camera to generate a point cloud dataset.

[0024] The point cloud dataset includes the spatial position coordinates, surface reflection intensity values ​​and normal vectors of multiple sampling points. By comprehensively considering the spatial coordinates, reflection intensity and normal vectors, the geometric shape and surface features of the exhibits can be reconstructed more comprehensively.

[0025] Specifically, the point cloud dataset generation flow chart is as follows Figure 2 As shown, it specifically includes: arranging multiple depth cameras, the depth cameras are arranged in an array, the angle between each depth camera is adjustable (120 degrees in this embodiment), forming an expandable annular scanning array, the selection of the angle is based on a large amount of experimental data, while ensuring the scanning coverage rate and minimizing data redundancy; placing the annular scanning array on a lifting bracket, and driving the annular scanning array to move in a vertical direction by the lifting bracket to form a spiral scanning trajectory, and the moving speed of the lifting bracket is controlled within the range of 2-5mm / s, which is the optimal balance between scanning accuracy and efficiency.

[0026] Furthermore, each depth camera emits infrared structured light, which is projected onto the surface of the exhibit to form a speckle pattern, and the deformation characteristics of the speckle pattern are collected. The wavelength of the infrared light used is 850nm, which is the result of comprehensive consideration of the visual characteristics of the human eye and the sensitivity of the sensor; according to the deformation characteristics of the speckle pattern, the spatial position coordinates of each sampling point on the exhibit surface are calculated using the triangulation principle; by analyzing the intensity of near-infrared light received at each sampling point, the reflection intensity value of the exhibit surface is obtained, and the reflection intensity value is used to characterize the material characteristics of the exhibit surface; based on the spatial position relationship between adjacent sampling points, the least squares method is used to fit the normal vector of the surface where each sampling point is located, and the normal vector is used to describe the orientation of the exhibit surface; the spatial position coordinates of each sampling point, the reflection intensity value of the exhibit surface and the normal vector are organized as structured data to generate a point cloud data set.

[0027] To ensure the accuracy of data collection, the present invention uses the industry-leading depth camera, which has a depth resolution of up to 0.1mm, a field of view of 60 degrees, and a working distance range of 0.5 to 3 meters. The selection of these parameters fully considers the size characteristics of the art exhibits and the display environment requirements.

[0028] Preferably, compared with the traditional single-view scanning method, the multi-depth camera array scanning strategy of the present invention can significantly improve the integrity and accuracy of point cloud data. Through the circular array and spiral scanning trajectory, the data blind area can be minimized and the ability to capture the surface details of the exhibits can be improved.

[0029] S2: Reconstruct the mesh of the point cloud dataset to generate a three-dimensional digital mesh of the exhibit.

[0030] Among them, the exhibited three-dimensional digital grid includes a grid vertex set, a topological connection matrix and a vertex normal vector set.

[0031] Specifically, the point cloud dataset is preprocessed, and the statistical outlier analysis method is used to remove noise points. Specifically, the average distance from each point to its nearest neighbor is calculated, and the points that deviate from the mean by more than two standard deviations are identified as noise points. The sampling points are spatially indexed and encoded through octree spatial segmentation, where the depth of octree spatial segmentation can be dynamically adjusted according to the density and complexity of the point cloud dataset to balance the computational efficiency and reconstruction accuracy. The search radius is set with the leaf node of the octree as the center, and the search radius is 3 times the average point spacing. The local point set in the neighborhood of each leaf node is extracted, and the covariance matrix of the local point set is calculated.

[0032] Furthermore, the covariance matrix is ​​decomposed into eigenvalues ​​to obtain the main eigenvector, which is used as the normal vector estimate of the local surface; based on the distribution characteristics of the local point set, an implicit function is constructed, and the zero isosurface of the exhibit surface is obtained by solving the Poisson equation. The zero isosurface is used to characterize the overall shape of the exhibit; when constructing the implicit function, methods such as radial basis function (RBF) or distance transformation function (SDF) can be used to improve the reconstruction ability of complex surface shapes; uniform sampling is performed on the zero isosurface to generate a mesh vertex set, which contains the key feature points of the exhibit surface; the mesh vertex set is connected into a triangular mesh using the Delaunay triangulation algorithm, and a topological connection matrix between mesh vertices is established; for each vertex of the triangular mesh, the vertex normal vector set is calculated based on the weighted average of the normal vectors of the adjacent triangular facets. When calculating the vertex normal vector, a weight factor can be introduced to weight the angle and area between the normal vector of the adjacent triangular facet and the vertex normal vector to improve the accuracy of the normal vector estimation.

[0033] Preferably, this step breaks through the limitations of traditional mesh reconstruction methods, and can significantly improve the geometric accuracy and detail restoration of the exhibit's three-dimensional digital mesh through innovative spatial indexing and surface reconstruction algorithms. Compared with traditional mesh reconstruction technology, the method proposed in the present invention can better handle complex surfaces and irregular geometric shapes.

[0034] S3: Analyze the material characteristics of the exhibit’s three-dimensional digital grid and extract the material parameter set.

[0035] Among them, the material parameter set includes ambient light coefficient, diffuse reflection coefficient, specular reflection coefficient and texture mapping matrix.

[0036] Specifically, a multi-light source array is arranged around the exhibits, and the multi-light source array includes light source units of different wavelengths. The wavelengths of the light sources cover the visible spectrum (380-780nm) with an interval of 20nm. Structured lighting is generated by controlling the switching sequence of each light source unit. In order to reduce errors, the multi-light source array needs to be calibrated for radiometry to ensure the consistency and repeatability of each light source unit. The reflected light intensity image of the three-dimensional digital grid of the exhibit under different lighting conditions is collected, and the reflected light intensity image is compared with the incident light intensity to generate reflectivity data. The reflectivity data is spatially mapped according to the grid vertex position, and a reflectivity distribution map is constructed. The material area of ​​the exhibit surface is segmented based on the reflectivity distribution map.

[0037] Furthermore, for each material area, the ambient light coefficient is determined by analyzing the reflectivity under dark lighting conditions, and the ambient light coefficient is normalized to eliminate the scale differences between different material areas, where the ambient light coefficient represents the material's ability to reflect ambient scattered light; under diffuse light illumination conditions, the directional distribution of the reflected light intensity is analyzed according to Lambert's cosine law, and the diffuse reflection coefficient is calculated, which describes the scattering characteristics of the material; using high light illumination, the specular reflection peak value and its attenuation law are measured, and the specular reflection coefficient is extracted, which represents the gloss of the material; based on the geometric characteristics of the exhibit surface, a UV coordinate system is established, and a texture mapping matrix is ​​generated by optimizing the minimum energy principle. The texture mapping matrix is ​​used to project material details onto the three-dimensional grid surface.

[0038] Preferably, the material analysis method used in this step breaks through the limitations of traditional material measurement under a single light source, and achieves accurate acquisition of material characteristics through multi-spectral structured light illumination. Experiments have shown that this method can accurately restore more than 90% of common artwork materials, including metals, ceramics, fabrics, etc.

[0039] S4: Collect the audience position coordinates, calculate the audience-exhibit relative azimuth based on the audience position coordinates, and generate a perspective transformation matrix.

[0040] Specifically, an infrared depth camera array is arranged on the top of the exhibition hall. The infrared depth camera array includes multiple depth camera units. The fields of view of each depth camera unit overlap with each other, and the overlapping area of ​​adjacent cameras is not less than 30% to ensure a seamless tracking effect. The depth image is acquired through the infrared depth camera array, and the background difference method is used to extract the audience foreground target. The human key point detection is performed on the foreground target, and the improved OpenPose algorithm is used for human posture estimation. The key point detection accuracy rate is over 95%. Based on the detected human key points, the audience's head feature points are extracted, and the three-dimensional coordinates of the head feature points are used as the audience's viewpoint position coordinates.

[0041] Furthermore, a world coordinate system of the exhibition hall is established, with the center point of the exhibit as the origin, and the vector between the viewer's viewpoint position coordinates and the exhibit center point is calculated as the observation vector; the relative translation vector is calculated according to the world coordinate system of the exhibition hall and the viewer's head coordinates to form a translation matrix; the observation vector is decomposed into a horizontal azimuth and a vertical pitch angle, wherein the horizontal azimuth indicates the viewer's viewing direction in the horizontal plane, and the vertical pitch angle indicates the degree to which the viewer looks up or down; a rotation matrix is ​​constructed according to the horizontal azimuth and the vertical pitch angle, and the rotation matrix is ​​combined with the translation matrix to generate a perspective transformation matrix between the viewer and the exhibit; the perspective transformation matrix is ​​calibrated to eliminate the coordinate system conversion error and ensure the accuracy of the perspective conversion.

[0042] S5: Apply the perspective transformation matrix to the three-dimensional digital grid of the exhibit, superimpose the material parameter set for rendering, output the exhibit image after the perspective transformation, and synthesize the exhibit image and the exhibition hall background image into a mixed reality scene.

[0043] Specifically, the perspective transformation matrix is ​​applied to the vertex set of the three-dimensional digital grid of the exhibit, the spatial position of the vertices after the perspective transformation is calculated, and the transformed grid geometry structure is generated; based on the transformed grid geometry structure, a depth buffer is constructed, and the Z-buffer algorithm is used to determine the set of visible triangles on the grid surface, and the occluded triangles are removed; for each facet in the set of visible triangles, the light transmission equation is calculated according to the material parameter set, where the ambient light coefficient, diffuse reflection coefficient and mirror reflection coefficient determine the shading effect of the facet.

[0044] Furthermore, the high-resolution texture image is projected onto the visible triangular facets using the texture mapping matrix, and a smooth texture transition effect is generated through a bilinear interpolation algorithm. The real-time background image of the exhibition hall is collected, the lighting parameters of the exhibition hall are extracted, and the lighting parameters are applied to the exhibit rendering process, so that the exhibits present a lighting effect consistent with the exhibition hall environment. The rendered exhibit image is anti-aliased to smooth edge pixels and improve image display quality. The processed exhibit image is Alpha blended with the exhibition hall background image to generate a realistic mixed reality scene.

[0045] S6: Based on the real-time changes of the viewer's position coordinates, the perspective transformation matrix is ​​dynamically updated to regenerate the mixed reality scene.

[0046] Specifically, audience motion data is collected at a frequency of 30 frames per second through a depth camera array, the motion trajectory of the audience's head feature points is extracted, and a sequence of audience position coordinates is generated; Kalman filtering is performed on the audience position coordinate sequence to eliminate position coordinate jitter and predict the audience position coordinates at the next moment. The process noise covariance and observation noise covariance of the Kalman filter are obtained through experimental optimization, which can effectively suppress more than 90% of tracking jitter; based on the filtered audience position coordinates, the quaternion interpolation method is used to calculate the gradual process of the perspective transformation matrix to achieve a smooth transition of the perspective; the updated perspective transformation matrix is ​​passed to the graphics processing unit, and the parallel transformation calculation of the mesh vertices is started to accelerate the geometric transformation processing; the material parameter set is kept unchanged, and material rendering is performed on the transformed mesh, and the frame difference technology is used to update only the changed image area; the dynamic changes of the exhibition hall background image are detected, and the background update is synchronized with the exhibit rendering results to maintain the timing consistency of the mixed reality scene; the generated mixed reality scene is output in real time through the display device to ensure that the image refresh delay is less than 16 milliseconds.

[0047] Furthermore, the present embodiment also provides a digital display system for art exhibitions, including a depth acquisition module, which is used to scan the surface of the exhibit through a depth camera to generate a point cloud data set; a three-dimensional grid reconstruction module, which is used to reconstruct the grid of the point cloud data set to generate a three-dimensional digital grid of the exhibit; a material feature analysis module, which is used to perform material feature analysis on the three-dimensional digital grid of the exhibit and extract a material parameter set; an audience position tracking module, which is used to collect audience position coordinates, calculate the audience-exhibit relative azimuth based on the audience position coordinates, and generate a perspective transformation matrix; a mixed reality rendering module, which is used to apply the perspective transformation matrix to the three-dimensional digital grid of the exhibit, superimpose the material parameter set for rendering, output the exhibit image after the perspective transformation, and synthesize the exhibit image with the exhibition hall background image into a mixed reality scene; a dynamic scene update module, which is used to dynamically update the perspective transformation matrix based on the real-time changes of the audience position coordinates and regenerate the mixed reality scene.

[0048] In summary, the present invention realizes high-precision digital acquisition, realistic rendering and smooth real-time interactive experience of art exhibits through the above technical solutions. Compared with the existing technology, it has the following advantages: first, the multi-sensor collaborative acquisition solution is adopted to significantly improve the accuracy of three-dimensional reconstruction; second, the multi-spectral structured light illumination technology is innovatively introduced to achieve accurate acquisition of material properties; third, the improved visual tracking algorithm and real-time rendering optimization strategy are adopted to control the system response delay below the perceptible threshold of the human eye. Experimental results show that the present invention can meet the application requirements of most art exhibition scenes and has good practical value.

[0049] Example 2, reference Figure 1-2, which is the second embodiment of the present invention, and this embodiment provides a digital display method for art exhibitions. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0050] To verify the feasibility and performance of the digital display method, the research team conducted a three-month system test and experiment in an art museum. The test site was an 80-square-meter dedicated exhibition hall, and 10 artworks made of different materials, including bronze, porcelain, and fabric, were selected as test objects.

[0051] During the point cloud data collection phase, six Intel RealSense D455 depth cameras were used to form a circular scanning array, with the angle between the cameras set to 120 degrees. The moving speed of the lifting bracket was set to 3mm / s, and it took about 15 minutes to complete a complete scan of a single exhibit. Experimental data showed that under the irradiation of 850nm wavelength infrared structured light, for exhibits within the size of 50cm×50cm×50cm, the average sampling point density reached 0.2mm / point, and the spatial accuracy of the point cloud data was better than 0.1mm.

[0052] During the mesh reconstruction process, the sampling points are divided into voxel units of 2mm×2mm×2mm in size through octree space segmentation, and the search radius is set to 6mm. Experiments show that this parameter setting can effectively balance the reconstruction accuracy and computational efficiency, with a noise point elimination rate of 98.5%, an average side length of the reconstructed triangular mesh of 0.5mm, and can accurately restore more than 95% of the geometric details of the exhibit surface.

[0053] The material feature analysis uses 20 LED light sources with different wavelengths, ranging from 380 to 780 nm, with an interval of 20 nm. The controller generates 2048 different lighting combinations to obtain reflectivity data. The experimental results show that this method can accurately identify and quantify the optical properties of 7 common artwork materials, and the measurement error of material parameters is controlled within 5%.

[0054] In the audience tracking experiment, 8 infrared depth cameras were used to form an overlapping field of view on the top of the exhibition hall, with an overlapping area of ​​35% between adjacent cameras. The improved OpenPose algorithm was used to estimate human posture, and the test was conducted under different lighting conditions and crowd density. The accuracy of head feature point detection reached 96.8%, and the average error of position coordinates was less than 1 cm. After applying the Kalman filter, the tracking jitter of the audience's motion trajectory was reduced by 92.3%, and the prediction accuracy was increased to 98.2%.

[0055] In the real-time rendering test, an NVIDIA RTX 3080 graphics card was used for graphics processing. For an exhibit model containing 500,000 triangles, the average processing time for perspective transformation and material rendering was 12ms, and the background synthesis time was 3ms. The overall system delay was controlled within 16ms, ensuring a smooth interactive experience. During the week-long public test, more than 3,000 viewers experienced the system, and the satisfaction survey results showed that 93.5% of the audience believed that the system could provide a close-to-real art viewing experience.

[0056] In order to systematically evaluate the advantages of the method of the present invention, the research team conducted detailed comparative experiments with the prior art, and the results are shown in Table 1.

[0057] Table 1 Performance comparison between the present invention and the prior art

[0058] Performance Indicators Method of the present invention Traditional single-view scanning method <![CDATA[Point cloud density (points / mm 2 )]]> 25 10 Geometric accuracy(mm) 0.1 0.3 Material recognition accuracy (%) 95 70 Viewing angle response time (ms) 16 50 Scene update frame rate (fps) 60 20

[0059] From the comparative data in Table 1, it can be seen that the method of the present invention is significantly better than the existing technology in all key performance indicators, especially in terms of point cloud density, geometric accuracy and material recognition accuracy. These advantages are mainly due to the innovative multi-view scanning strategy and high-precision material analysis method, which provides better technical support for the digital display of art exhibitions.

[0060] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A digital display method for art exhibition, characterized in that: include, Scan the surface of the exhibits with a depth camera to generate a point cloud dataset; Reconstructing the grid of the point cloud data set to generate a three-dimensional digital grid of the exhibit; Performing material feature analysis on the three-dimensional digital grid of the exhibit and extracting a material parameter set; Collecting the audience position coordinates, calculating the audience-exhibit relative azimuth based on the audience position coordinates, and generating a viewing angle transformation matrix; Applying the perspective transformation matrix to the three-dimensional digital grid of the exhibit, superimposing the material parameter set for rendering, outputting an image of the exhibit after perspective transformation, and synthesizing the exhibit image with the exhibition hall background image into a mixed reality scene; Based on the real-time changes in the viewer's position coordinates, the viewing angle transformation matrix is ​​dynamically updated to regenerate the mixed reality scene.

2. The digital display method for art exhibition according to claim 1, characterized in that: The point cloud data set includes spatial position coordinates, surface reflection intensity values ​​and normal vectors of multiple sampling points; The three-dimensional digital grid of the exhibit includes a grid vertex set, a topological connection matrix and a vertex normal vector set; The material parameter set includes an ambient light coefficient, a diffuse reflection coefficient, a specular reflection coefficient, and a texture mapping matrix.

3. The digital display method for art exhibition according to claim 1, characterized in that: Generating a point cloud dataset comprises the following steps: Arrange a plurality of depth cameras, wherein the depth cameras are arranged in an array, and the angles between the depth cameras are adjustable to form an expandable annular scanning array; Placing the annular scanning array on a lifting bracket, and driving the annular scanning array to move in a vertical direction by the lifting bracket to form a spiral scanning track; Each depth camera emits infrared structured light, projects it onto the surface of the exhibit to form a speckle pattern, and collects deformation features of the speckle pattern; Calculating the spatial position coordinates of each sampling point on the surface of the exhibit according to the deformation characteristics of the speckle pattern; By analyzing the near-infrared light intensity received at each sampling point, the reflection intensity value of the exhibit surface is obtained; Based on the spatial position relationship of adjacent sampling points, the least square method is used to fit the normal vector of the surface where each sampling point is located; The spatial position coordinates of each sampling point, the reflection intensity value of the exhibit surface and the normal vector are organized into structured data to generate a point cloud dataset.

4. The digital display method for art exhibition according to claim 1, characterized in that: Generating the three-dimensional digital grid of the exhibits comprises the following steps: Preprocess the point cloud data set, remove noise points, and perform spatial index encoding on the sampling points through octree spatial segmentation; Taking the leaf node of the octree as the center, setting the search radius, extracting the local point set in the neighborhood of each leaf node, and calculating the covariance matrix of the local point set; Performing eigenvalue decomposition on the covariance matrix to obtain a main eigenvector, and using the main eigenvector as an estimated value of a normal vector of a local surface; Based on the distribution characteristics of the local point set, an implicit function is constructed, and the zero isosurface of the exhibit surface is obtained by solving the Poisson equation; Uniform sampling on the zero isosurface generates a mesh vertex set; Using a Delaunay triangulation algorithm, the mesh vertex set is connected into a triangular mesh, and a topological connection matrix between mesh vertices is established; For each vertex of the triangular mesh, a vertex normal vector set is calculated based on a weighted average of normal vectors of adjacent triangular facets.

5. The digital display method for art exhibition according to claim 1, characterized in that: The extracting of the material parameter set comprises the following steps: Arranging a multi-light source array around the exhibit, the multi-light source array comprising light source units of different wavelengths, and generating structured lighting by controlling the switching sequence of each light source unit; Collecting reflected light intensity images of the three-dimensional digital grid of the exhibit under different lighting conditions, comparing the reflected light intensity images with the incident light intensity, and generating reflectivity data; The reflectivity data is spatially mapped according to the grid vertex positions to construct a reflectivity distribution map, and the surface of the exhibit is segmented into material regions based on the reflectivity distribution map; For each material area, the ambient light coefficient is determined by analyzing the reflectivity under dark lighting conditions; Under diffuse light illumination conditions, the directional distribution of reflected light intensity is analyzed and the diffuse reflectance coefficient is calculated; Using high light irradiation method, measure the specular reflection peak value and its attenuation law, and extract the specular reflection coefficient; Based on the geometric features of the exhibit surface, a UV coordinate system is established, and the texture mapping matrix is ​​optimized and generated through the minimum energy principle.

6. The digital display method for art exhibition according to claim 1, characterized in that: Generating the perspective transformation matrix comprises the following steps: Arrange an infrared depth camera array on the top of the exhibition hall, wherein the infrared depth camera array includes a plurality of depth camera units, and the fields of view of the depth camera units overlap with each other; Acquire a depth image through the infrared depth camera array, extract the audience foreground target, and perform human key point detection on the foreground target; Based on the detected key points of the human body, the feature points of the audience's head are extracted, and the three-dimensional coordinates of the feature points of the head are used as the coordinates of the viewpoint position of the audience; Establish the exhibition hall world coordinate system, take the center point of the exhibit as the origin, and calculate the vector between the viewer's viewpoint position coordinates and the center point of the exhibit as the observation vector; Decomposing the observation vector into a horizontal azimuth angle and a vertical pitch angle; A rotation matrix is ​​constructed according to the horizontal azimuth angle and the vertical pitch angle, and the rotation matrix is ​​combined with a translation matrix to generate a viewing angle transformation matrix between the audience and the exhibit.

7. The digital display method for art exhibition according to claim 1, characterized in that: The generation process of the mixed reality scene is as follows: Apply the perspective transformation matrix to the vertex set of the exhibit's three-dimensional digital grid, calculate the spatial position of the vertices after the perspective transformation, and generate the transformed grid geometry structure; Based on the transformed mesh geometry, a depth buffer is constructed to determine the set of visible triangles on the mesh surface and remove the obscured triangles. For each face in the visible triangle face set, the light transmission equation is calculated according to the material parameter set, where the ambient light coefficient, diffuse reflection coefficient and specular reflection coefficient determine the shading effect of the face; Use the texture mapping matrix to project the texture image onto the visible triangles; Collecting real-time exhibition hall background images, extracting the exhibition hall's lighting parameters, and applying the lighting parameters to the exhibit images; The rendered exhibit image is alpha blended with the exhibition hall background image to generate a mixed reality scene.

8. A digital display system for art exhibitions, characterized in that: The system includes: The depth acquisition module is used to scan the surface of the exhibits through a depth camera to generate a point cloud data set; A three-dimensional mesh reconstruction module is used to reconstruct the mesh of the point cloud data set to generate a three-dimensional digital mesh of the exhibit; A material feature analysis module, used to perform material feature analysis on the three-dimensional digital grid of the exhibit and extract a material parameter set; The audience position tracking module is used to collect audience position coordinates, calculate the audience-exhibit relative azimuth based on the audience position coordinates, and generate a viewing angle transformation matrix; A mixed reality rendering module, used for applying the perspective transformation matrix to the three-dimensional digital grid of the exhibit, superimposing the material parameter set for rendering, outputting an image of the exhibit after the perspective transformation, and synthesizing the exhibit image with the exhibition hall background image into a mixed reality scene; A dynamic scene update module is used to dynamically update the viewing angle transformation matrix based on the real-time changes in the viewer's position coordinates to regenerate the mixed reality scene.

Images