Digital reconstruction method for complex special-shaped component of historical building

Abstract

The invention relates to the technical field of historical building protection and digitization, and discloses a digital reconstruction method for complex special-shaped components of a historical building. Through the high-precision three-dimensional laser scanning technology, the complex geometrical shape of the special-shaped part is comprehensively captured, the limitation of traditional ruler measurement on irregular features such as curved surfaces and recesses is overcome, and data integrity and millimeter-level reduction precision are ensured; an electron microscope and a chromatic aberration analysis technology are combined, material characteristics are analyzed layer by layer from microscopic components to macroscopic colors, the component proportion and weathering traces of a traditional process material are accurately recognized, and a reliable basis is provided for repairing material selection; through digital modeling and a color gradient algorithm, fine color transition and texture details are restored, a material color information base is constructed, and efficient management and accurate matching of data are achieved. The method provides systematic technical support for digital inheritance of cultural heritage, and has both engineering practicability and cultural value continuity.

Description

Technical Field

[0001] The present invention relates to the field of historical building protection and digitization technology, and in particular to a method for digitally reconstructing complex special-shaped parts of historical buildings. Background Art

[0002] Historical buildings carry rich cultural, artistic, and historical values. Their distinctive features, such as carved capitals, dome decorations, and eaves, are often complex in shape and size, made of diverse materials, and rich in color. Traditional surveying and mapping methods face many challenges when dealing with these complex and irregular components:

[0003] Difficulty in shape surveying: Due to the irregular shapes of special-shaped parts, there are a large number of complex geometric features such as curved surfaces, depressions, and protrusions. It is difficult to accurately obtain the dimensions of each part using conventional rulers and compasses, which can easily lead to large errors. In addition, it is difficult to reach and measure some high-altitude or hidden parts, resulting in incomplete surveying and mapping data.

[0004] Material analysis is difficult: Historical buildings experience changes in composition due to weathering and erosion over time. Furthermore, some materials may be complex and shaped by traditional craftsmanship. Simple external observation or conventional chemical analysis methods make it difficult to accurately identify material types and proportions, making it difficult to provide a precise basis for subsequent restoration.

[0005] Low color reproduction accuracy: The color of architectural components is affected by factors such as light, oxidation, and dirt adhesion. The color gradients in different areas are delicate and rich. The human eye and ordinary color measurement instruments have difficulty accurately capturing their true color values, making it impossible to restore their original brilliant colors during restoration or digital display.

[0006] In summary, there is an urgent need for an innovative method that can comprehensively and accurately digitally reconstruct complex and special-shaped components of historical buildings in order to achieve effective protection, inheritance and reuse of historical buildings. To this end, a digital reconstruction method for complex and special-shaped components of historical buildings is proposed. Summary of the Invention

[0007] In view of the shortcomings of the existing technology, the present invention provides a method for digitally reconstructing complex and irregularly shaped components of historical buildings to solve the background technical problems.

[0008] To achieve the above-mentioned object, the present invention provides the following technical solution: a method for digitally reconstructing complex and irregularly shaped parts of historical buildings, comprising the following steps:

[0009] Step 1: Data collection:

[0010] Use 3D laser scanning technology to perform multiple superimposed data scans to obtain point cloud information of special-shaped components of historical buildings;

[0011] De-noising and splicing of 3D laser scanning data to accurately obtain the shape of complex and special-shaped components;

[0012] Step 2: Material and color analysis:

[0013] Use a colorimeter to quickly identify the surface texture and color of special-shaped parts;

[0014] Use electron microscope analyzer to accurately identify material components and obtain material information;

[0015] Step 3: Digital reconstruction:

[0016] The complex and irregularly shaped parts of historical buildings are realistically digitally reconstructed through 3D modeling software, and a color information library of the materials is established.

[0017] Preferably, the method for obtaining point cloud information of special-shaped building components in step 1 specifically includes:

[0018] Use 3D laser scanners to rationally arrange scanning stations based on the size, location, and surrounding environment of special-shaped parts to ensure there are no blind spots in the scanning.

[0019] Perform multi-angle and multi-level scanning on special-shaped parts, ensuring a certain degree of overlap between adjacent scanning stations to obtain dense and complete point cloud data;

[0020] During the scanning process, laser parameters, including but not limited to wavelength and pulse frequency, are adjusted according to the surface material characteristics of the component to improve the quality of point cloud data.

[0021] Preferably, the processing of the three-dimensional laser scanning data in step 1 specifically includes:

[0022] Utilize professional data processing software and statistical filtering algorithms to remove isolated noise points from point cloud data, including but not limited to judging and removing noise points based on neighborhood density, distance threshold, etc.

[0023] Scanning artifacts, including but not limited to those caused by ambient light and obstructions, are repaired through manual editing or geometric feature-based patching algorithms. Then, based on the common feature areas between scanning sites, the iterative closest point (ICP) algorithm is used for point cloud stitching, accurately fusing the local point clouds of each site into a complete point cloud model of the special-shaped component, restoring its complex geometric shape.

[0024] Preferably, the identification of the surface texture and color of the special-shaped component in step 2 specifically includes:

[0025] Use a high-precision colorimeter with multi-angle measurement capabilities to measure different areas of the surface of special-shaped parts point by point or in sections;

[0026] For textured surfaces, take into account the texture direction, including but not limited to dense sampling at key locations of texture convexity, concavity, and turning points, measure their color values, and record the corresponding spatial coordinates;

[0027] By analyzing the color data of a large number of sampling points, a color gradient model is constructed to accurately capture the subtle changes in color texture, providing a basis for subsequent color rendering of digital models.

[0028] Preferably, the material information acquisition in step 2 specifically includes:

[0029] Collect tiny samples from inconspicuous locations on special-shaped parts or damaged repair areas, place them in a scanning electron microscope (SEM), and use secondary electrons and backscattered electron signals generated by the interaction between the electron beam and the sample surface to observe the microstructural characteristics of the material, including but not limited to crystal morphology and pore distribution;

[0030] The elemental composition of the sample is analyzed using an energy dispersive spectrometer (EDS). By comparing it with a database of known materials, the type of material, including but not limited to masonry, wood, and metal alloys, as well as the relative content of each component, is determined to understand the material properties and provide a reference for material selection for restoration.

[0031] Preferably, the digital reconstruction of the complex and irregularly shaped parts of the historical building in step 3 specifically includes:

[0032] Import the processed point cloud data into professional 3D modeling software, build a polygon mesh model based on the point cloud, and generate a high-quality, low-polygon 3D model with accurate geometric features through manual topology optimization or automatic reconstruction algorithm;

[0033] Based on the color data obtained by the colorimeter, the software's material editing function is used to give realistic color textures to each surface area of the model, achieving a high-fidelity visual effect.

[0034] Preferably, the establishment of the material color information library in step 3 specifically includes:

[0035] A material color information library is created in professional 3D modeling software, and the material information obtained by the electron microscope analyzer is associated with the corresponding color and stored for easy subsequent retrieval. When restoration or virtual display is required, materials and colors can be matched quickly and accurately to complete the digital reconstruction of complex and special-shaped parts of historical buildings.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] This invention uses high-precision three-dimensional laser scanning technology to comprehensively capture the complex geometric shapes of special-shaped components, overcoming the limitations of traditional ruler and compass measurement for irregular features such as curved surfaces and depressions, ensuring data integrity and millimeter-level restoration accuracy. Combining electron microscopy and color difference analysis technology, it analyzes material properties layer by layer from microscopic components to macroscopic colors, accurately identifying the composition ratios and weathering traces of traditional craft materials, providing a reliable basis for restoration material selection. Through digital modeling and color gradient algorithms, it restores delicate color transitions and texture details, and constructs a material color information library to achieve efficient data management and precise matching. This invention provides systematic technical support for the digital inheritance of cultural heritage, combining engineering practicality with the continuity of cultural value.

[0038] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a flow chart of the method for digitally reconstructing complex and irregularly shaped components of an entire historical building according to the present invention. DETAILED DESCRIPTION

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this technical field without making any creative efforts shall fall within the scope of protection of the present invention.

[0041] See also Figure 1 The present invention provides a digital reconstruction method for complex and irregularly shaped components of historical buildings. This method uses 3D laser scanning technology, material composition and color analysis technology, and digital modeling technology to achieve high-precision, full-process digital reconstruction. Specifically, it includes the following:

[0042] Example 1:

[0043] 1. Data Collection

[0044] 1. 3D laser scanning parameter configuration and site layout

[0045] 1.1 Equipment and parameter selection:

[0046] Scanner type: Select the device based on the size of the target component and the complexity of the environment:

[0047] Small parts (size ≤ 3m): Use a handheld 3D laser scanner with an accuracy of ±0.1mm and a scanning distance of 0.5-1m, suitable for capturing fine surfaces.

[0048] Large components (size > 5m): Use a ground-based long-range scanner with a scanning distance of ≥ 80m and an elevation angle range of ±90°, supporting scanning of complex high-altitude structures.

[0049] Laser parameter adjustment:

[0050] Light-absorbing materials (such as wood and plaster): Use the near-infrared band (1550nm) to reduce scattering noise, and set the pulse frequency to 100-120kHz to balance point cloud density and scanning efficiency.

[0051] Highly reflective materials (such as stone and metal): A short wavelength (532nm) is used to enhance detail capture, and the pulse frequency is increased to 200kHz to meet long-distance scanning needs.

[0052] 1.2 Scanning site layout strategy

[0053] General principles:

[0054] Overlap: Adjacent sites should maintain a 30%-50% overlap area to ensure seamless point cloud stitching.

[0055] Site spacing: Dynamically adjusted according to the geometric complexity of the components, with a spacing of 0.5-1m for small components and 3-5m for large components.

[0056] Elevation coverage: For high altitude or sunken areas, set up inclined stations (elevation angle 30°-60°) to avoid scanning blind spots.

[0057] Operation process:

[0058] Pre-scan: Quickly obtain the outline of the component and identify key feature points (such as carved edges and seam lines). Formal scan: Scan each station according to the planned path, taking 10-20 minutes per station to generate local point cloud data.

[0059] Real-time quality monitoring: Use the supporting software to check point cloud coverage and noise levels, and promptly scan missing areas.

[0060] Through dynamic parameter adjustment and scientific station layout strategies, high-precision data collection of complex geometric shapes (such as 0.1mm-level carved patterns) is ensured, providing a reliable foundation for subsequent processing.

[0061] 2. Point cloud denoising and stitching algorithm

[0062] 2.1 Denoising

[0063] Noise types and coping strategies:

[0064] Isolated noise: discrete noise points caused by environmental dust or equipment errors;

[0065] Artifact noise: local data loss or distortion caused by reflection, occlusion, or motion interference.

[0066] Algorithm selection and implementation:

[0067] Median filter algorithm: eliminates isolated noise based on statistical principles.

[0068] Expressed as:

[0069] Retention conditions: and d 邻域 ≤d 阈值

[0070] Among them, N 邻域 is the number of points in a neighborhood with a radius of r = 2 mm centered at a certain point; ρ 阈值 = 0.3 means points below this density threshold are considered as noise; d 阈值 =0.5mm is the maximum allowable spacing between neighboring points to prevent accidental deletion of valid points;

[0071] Operation example: For wooden carved parts, reduce noise points by 90% and retain effective point cloud density ≥ 95%.

[0072] B-spline surface repair: For artifact missing areas, data is restored through geometric interpolation;

[0073] Expressed as:

[0074]

[0075] Among them, B i,p (u) is the p-order B-spline basis function, which controls the smoothness of the surface; P i,j Interpolation control points are generated based on the surrounding valid point cloud;

[0076] 2.2 Point Cloud Stitching

[0077] ICP algorithm (Iterative Closest Point):

[0078] Mathematical principle: Point cloud alignment is achieved by minimizing the distance error between corresponding points;

[0079]

[0080] Among them, R is the rotation matrix, which adjusts the direction of the point cloud; t is the translation vector, which adjusts the position of the point cloud; p i ,q i are matching point pairs of adjacent point clouds.

[0081] Operation process:

[0082] Coarse registration: preliminary alignment based on feature points (such as carving corners), with an error of about 1-2mm;

[0083] Fine registration: iterative optimization (10-20 times), the residual converges to ≤0.2mm;

[0084] Technical function: Ensure seamless integration of multi-site clouds and restore the complete geometric shape of components.

[0085] Marker ball assisted stitching:

[0086] Implementation method: Pre-place a reflective marker ball on the surface of the component, and use a total station to accurately measure the coordinates of the ball center (accuracy ±0.05mm);

[0087] Solve the problem of insufficient feature points when stitching at high altitude or complex surfaces, with an error of ≤0.3mm.

[0088] 2. Material and Color Analysis

[0089] 1. Color difference analysis and color gradient model construction

[0090] 1.1 Equipment and sampling design

[0091] 1.11 Colorimeter Selection:

[0092] Functional requirements: Support multi-angle (15°, 45°, 75°) measurement, Lab color space resolution ΔE≤0.4.

[0093] Example device: X-Rite Ci7800, suitable for capturing color on complex textured surfaces.

[0094] 1.12 Sampling strategy:

[0095] Key parts of the texture: densely distributed along convex, concave, and turning points (density: 1-2 points / cm 2 );

[0096] Uniform color area: uniform sampling of different areas (density: 1 point / 10cm 2 );

[0097] Spatial coordinate recording: Each sampling point is associated with three-dimensional coordinates (x, y, z) to ensure accurate mapping of color data to geometric models.

[0098] 1.2 Color Correction and Data Processing

[0099] 1.21 Lighting compensation model:

[0100] Expressed as:

[0101] L ′ =L·k 光照 ,a′=a+Δa环境 ,b′=b+Δb 环境

[0102] Where L, a, b are the original measured Lab color space values; k 光照 is the light intensity attenuation coefficient (e.g., set to 0.95-0.98 in a cloudy environment), which corrects the brightness deviation caused by insufficient light or overexposure; L is the brightness (0-100, 0 is black, 100 is white); a is the red and green channel (-128 to +127, negative values are greener, positive values are redder); b is the yellow and blue channel (-128 to +127, negative values are bluer, positive values are yellower); a 环境 , b 环境 is the color shift caused by ambient light.

[0103] 1.22 Color gradient model construction:

[0104] Radial Basis Function (RBF) interpolation:

[0105] C(x,y,z)=∑w i (||(x,y,z)-(x i ,y i ,z i )||)

[0106] Where (r) = r 2 logr is the thin plate spline basis function, which smoothly transitions the color difference; w i is the weight coefficient, which is solved by the least square method to ensure that the interpolation surface fits the sampling data; C(x,y,z) is the color value after interpolation; (x,y,z) is the three-dimensional coordinate of the target point; (x i ,y i ,z i ) is the three-dimensional coordinate of the i-th sampling point;

[0107] Output result: Generate a gradient map in Lab color space with an accuracy of ΔE≤2.0;

[0108] By constructing a continuous color distribution through sparse sampling points, the color gradient on complex textures can be accurately restored, giving digital models high-fidelity visual effects.

[0109] 2. Material composition analysis

[0110] 2.1 Sample collection and processing

[0111] Collection principles:

[0112] Location selection: Prioritize non-display surfaces, damaged or hidden areas to avoid damaging the structural integrity;

[0113] Sample size: small samples (wood: 10-20 mg; masonry: 5-10 mm3 ), to meet the analysis needs.

[0114] Preprocessing:

[0115] Wood samples: washed with ethanol and then dried (60°C, 2 hours) to remove contaminants;

[0116] Masonry samples: Cut into flat specimens to avoid interference from the surface oxide layer.

[0117] 2.2 SEM-EDS analysis process

[0118] 2.21 Scanning Electron Microscope (SEM) Imaging:

[0119] Parameter settings: acceleration voltage 15-20 kV, resolution 1 nm, backscattered electron mode.

[0120] Output:

[0121] Wood: Displays microstructures such as tracheids and wood rays, and identifies tree species (such as oak and redwood);

[0122] Brick and stone: Observe the mineral crystal morphology (such as quartz and calcite) and pore distribution.

[0123] 2.22 Energy Dispersive Spectrometer (EDS) Composition Analysis:

[0124] Element detection: quantitative analysis of the content of elements such as C, O, Ca, Si, etc.

[0125] Material determination: Compare to a database (such as NIST standards) to determine the material type (such as travertine stone, clay brick);

[0126] Provide material composition, microstructure and physical property data to guide the selection of repair materials and process design (such as moisture content control and mortar ratio).

[0127] 3. Digital Reconstruction

[0128] 1. 3D modeling and topology optimization

[0129] Point cloud data processing

[0130] Software platform: Rhino 7, Maya, supports Poisson surface reconstruction and mesh editing.

[0131] Modeling process:

[0132] a. Point cloud import: Import the complete stitched point cloud data into the software, with the data volume ranging from tens to hundreds of millions of points.

[0133] b. Surface reconstruction: Use the Poisson algorithm to generate the initial triangular mesh model;

[0134]

[0135] Among them, χ is an implicit function that describes the surface geometry; is the point cloud normal field, which drives the surface generation. By solving the partial differential equation, an implicit surface χ is generated, and its normal vector is aligned with the input point cloud, and finally a triangular mesh model is extracted.

[0136] c. Mesh simplification:

[0137] Manual optimization: For carving details, edge loop reduction is used to reduce the number of faces while retaining 0.1mm-level features.

[0138] Automatic simplification: Quadric Edge Collapse algorithm, cost function:

[0139]

[0140] Among them, Q(v) is the simplification cost of vertex v, which represents the error impact on the model geometry after merging the vertex; f∈F is all the facets associated with vertex v; n f is the patch normal vector, ensuring that the geometric features of the model remain unchanged after simplification; p f is the center point of the patch, guiding the vertex merging priority; v is the vertex to be merged.

[0141] d. Result output: After optimization, the number of model faces is reduced to 10%-20% of the original data, and the geometric error is ≤0.1mm;

[0142] During the mesh simplification process, the edges with the lowest cost are merged first to minimize geometric distortion, reduce model complexity while ensuring accuracy, and meet the needs of real-time rendering and engineering applications.

[0143] 2. Material rendering and information library construction

[0144] 2.1 Color and Material Mapping

[0145] Lab to RGB conversion process:

[0146] a. Color space conversion:

[0147]

[0148] Where X, Y, and Z are the tristimulus values in the CIE XYZ color space;

[0149] b. RGB generation: Calculate the final color value through the CIE XYZ to sRGB conversion matrix;

[0150] Conversion steps:

[0151] Lab→XYZ: Calculate the XYZ value using the above formula;

[0152] XYZ → RGB: Use the conversion matrix (sRGB standard):

[0153]

[0154] Gamma correction: Performs nonlinear mapping of RGB values to adapt to display devices.

[0155] Weathering effect simulation:

[0156] Expressed as:

[0157] S ′ =S·(1-k·w)

[0158] Among them, S is the original saturation; S ′ is the adjusted saturation; w is the weathering degree weight (0≤w≤1, 0 means no weathering, 1 means complete weathering); k is the attenuation coefficient (usually 0.3-0.5), which controls the saturation decrease rate.

[0159] Operation example:

[0160] Slightly weathered: w = 0.3, k = 0.3 → S ′ =S·0.91;

[0161] Severe weathering: w = 0.8, k = 0.5 → S ′ =S·0.6.

[0162] 2.2 Material color information library design

[0163] 2.21 Database Architecture and Data Association

[0164] Core goal: Build a structured database in professional 3D modeling software (such as Rhino, Maya, and Blender) to multi-dimensionally correlate material composition data obtained by electron microscope analyzers (SEM-EDS), color data collected by colorimeter analyzers (such as the X-Rite Ci7800), and the geometric coordinate information of 3D models to achieve precise matching and rapid access to materials and colors.

[0165] Database field design (pseudo code):

[0166]

[0167]

[0168] 2.22 Data Association and Storage Implementation

[0169] Association method:

[0170] Coordinate mapping: Bind each Lab value collected by the colorimeter to its spatial coordinates (x, y, z) on the 3D model surface.

[0171] Unique identifier (MaterialID): MaterialID is used to associate electron microscopy analysis data, color data, texture maps and physical parameters into the same material entry.

[0172] Storage format:

[0173] Database type: lightweight SQLite or NoSQL (such as MongoDB) to support fast query and expansion;

[0174] 3D software integration: Implement database read and write interfaces in modeling software through plug-ins or scripting languages (such as RhinoPython and Maya MEL).

[0175] Operation process:

[0176] Data input:

[0177] Electron Microscope Data: Parse the composition table (CSV / Excel) exported by SEM-EDS into JSON format and automatically fill it into the Components and Microstructure fields;

[0178] Color data: directly read Lab values through the colorimeter interface and associate them with corresponding coordinates;

[0179] Texture Mapping: Scan or photograph a high-resolution surface image, store it as a PNG / TIFF file, and write the path to the TextureMap.

[0180] Data verification:

[0181] Check the uniqueness of MaterialID to prevent duplicate entry;

[0182] Verify the matching accuracy of coordinates and Lab values (error ≤ 0.1 mm).

[0183] 2.23 Calling and matching mechanism

[0184] Repair scene call:

[0185] Search by material: Enter the material type (such as "redwood") to return all matching items, displaying parameters such as composition and moisture content to guide restoration material selection;

[0186] By color matching: Enter the target Lab value (such as L=65, a=12, b=8), and the database returns the closest MaterialID and corresponding texture map.

[0187] Virtual Showcase Call:

[0188] Automatic mapping: When loading a 3D model, the associated Lab value and texture are called according to the model surface coordinates to achieve "one-click rendering";

[0189] Dynamic adjustment: Supports modifying saturation (S) and brightness (L) through sliders, and previewing weathering effects in real time.

[0190] Algorithm implementation (taking color matching as an example):

[0191] Nearest neighbor search: Calculate the Euclidean distance between the target Lab value and all entries in the database, and select the MaterialID corresponding to the minimum value:

[0192]

[0193] Among them, the judgment criteria are:

[0194] ΔE≤1.0: The human eye can hardly detect the difference (industrial-grade accuracy);

[0195] ΔE≤2.0: Generally regarded as “imperceptible difference” (applicable to the restoration of historical buildings);

[0196] ΔE>3.0: color difference is clearly visible;

[0197] L i 、a i and b i The Lab value of the i-th material in the database; used to compare with the target color (L, a, bL, a, b) one by one to find the closest match.

[0198] Example:

[0199] Suppose you need to match restoration materials for the red area of a carved part of a historic building. The steps are as follows:

[0200] Target color measurement: L = 50.0, a = 60.0, b = 30.0 measured using a colorimeter;

[0201] Database Search: Calculate the ΔE values for all materials in the database:

[0202] Material A: L i =50.5,a i =59.8,b i =29.5→ΔЕ=

[0203]

[0204] Material B: L i =48.0,a i=62.0,b i =28.0→ΔЕ≈2.83;

[0205] Result judgment: Select material A with the smallest ΔE (ΔE≤1.0) to ensure highly consistent color reproduction.

[0206] Example 2:

[0207] Specific Experiment 1: Reconstruction of Small Wooden Carved Special-Shaped Parts

[0208] For a small wooden carved window lattice component, a portable 3D laser scanner was selected. 4-6 scanning stations were set up around the component at a distance of 0.5-1 meter, and scanning was performed with a 50% overlap to obtain point cloud data.

[0209] In the data processing software, median filtering is used to remove noise points caused by the rough surface of the wood and dust particles, and the feature-based ICP algorithm is used to stitch the point cloud to restore the fine geometric shapes of the carved window lattices.

[0210] Using a colorimeter, 50-100 sampling points were selected from different textured areas of the carved flower, such as petals, stamens, and leaf veins, to measure the colors and construct a color model.

[0211] Tiny sawdust samples are collected from the corners of window frames and placed in an electron microscope analyzer to determine the type of wood (such as redwood, camphor wood, etc.) and moisture content and other information. The samples are then imported into 3D modeling software to construct a 3D model, give it color and texture, and store it in the information database.

[0212] Example 3:

[0213] Specific Experiment 2: Reconstruction of Special-Shaped Components of Large Masonry Domes

[0214] For the domes of large brick and stone ancient buildings, 8-12 scanning stations are arranged on the ground below the dome, on the surrounding second-floor building platforms, etc. Long-distance, high-precision 3D laser scanners are used to scan from different elevation angles to ensure 30%-40% overlap and obtain the dome point cloud.

[0215] A robust statistical denoising algorithm is used to remove noise caused by sunlight reflection and bird occlusion, and a marker sphere-assisted stitching method is used to fuse the point cloud to accurately present the complex curves of the dome and the shape of the masonry joints.

[0216] With the help of a colorimeter, 200-300 points were densely sampled along the radial and circumferential directions of the dome in areas with different degrees of weathering on the masonry surface to measure the color. The data was corrected considering the influence of light to form a color gradient map.

[0217] Brick and stone samples were taken from the maintenance holes in the dome, and the brick and stone composition (such as clay bricks, travertine stone, etc.) and mineral content were determined through electron microscopy and energy spectrum analysis. A digital model of the dome was constructed in 3D software, and the materials were rendered based on the color map and material information. The color information library was improved to achieve digital reconstruction.

[0218] Summarize:

[0219] This method uses 3D laser scanning technology to comprehensively collect geometric data of irregularly shaped components of historical buildings. It then combines high-precision colorimeter and electron microscopy to obtain material composition and color information. After denoising, splicing, and modeling, specialized software is used to generate high-fidelity 3D models. A material color information library is then constructed for data association and storage. This library supports rapid matching and recall, providing a scientific basis for material selection for restoration. Furthermore, through color gradient algorithms and weathering simulation, the original appearance of the components is accurately restored. Ultimately, this method achieves a complete digital reconstruction of complex irregularly shaped components of historical buildings, from data collection to virtual display, balancing the high precision of cultural heritage preservation with the efficiency of engineering applications.

Claims

1. A method for digital reconstruction of complex and irregularly shaped parts of historical buildings, characterized by: The following steps are involved: Step 1: Data collection: Use 3D laser scanning technology to perform multiple superimposed data scans to obtain point cloud information of special-shaped components of historical buildings; De-noising and splicing of 3D laser scanning data to accurately obtain the shape of complex and special-shaped components; Step 2: Material and color analysis: Use a colorimeter to quickly identify the surface texture and color of special-shaped parts; Use electron microscope analyzer to accurately identify material components and obtain material information; Step 3: Digital reconstruction: The complex and irregularly shaped parts of historical buildings are realistically digitally reconstructed through 3D modeling software, and a color information library of the materials is established.

2. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 1, characterized in that: The method for obtaining point cloud information of special-shaped building components in step 1 specifically includes: Use 3D laser scanners to rationally arrange scanning stations based on the size, location, and surrounding environment of special-shaped parts to ensure there are no blind spots in the scanning. Perform multi-angle and multi-level scanning on special-shaped parts, ensuring a certain degree of overlap between adjacent scanning stations to obtain dense and complete point cloud data; During the scanning process, laser parameters, including but not limited to wavelength and pulse frequency, are adjusted according to the surface material characteristics of the component to improve the quality of point cloud data.

3. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 2, characterized in that: The processing of the three-dimensional laser scanning data in step 1 specifically includes: Utilize professional data processing software and statistical filtering algorithms to remove isolated noise points from point cloud data, including but not limited to judging and removing noise points based on neighborhood density, distance threshold, etc. Scanning artifacts, including but not limited to those caused by ambient light and obstructions, are repaired through manual editing or geometric feature-based patching algorithms. Then, based on the common feature areas between scanning sites, the iterative closest point (ICP) algorithm is used for point cloud stitching, accurately fusing the local point clouds of each site into a complete point cloud model of the special-shaped component, restoring its complex geometric shape.

4. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 1, characterized in that: The identification of the surface texture and color of the special-shaped parts in step 2 specifically includes: Use a high-precision colorimeter with multi-angle measurement capabilities to measure different areas of the surface of special-shaped parts point by point or in sections; For textured surfaces, take into account the texture direction, including but not limited to dense sampling at key locations of texture convexity, concavity, and turning points, measure their color values, and record the corresponding spatial coordinates; By analyzing the color data of a large number of sampling points, a color gradient model is constructed to accurately capture the subtle changes in color texture, providing a basis for subsequent color rendering of digital models.

5. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 4, characterized in that: The material information acquisition in step 2 specifically includes: Collect tiny samples from inconspicuous locations on special-shaped parts or damaged repair areas, place them in a scanning electron microscope (SEM), and use secondary electrons and backscattered electron signals generated by the interaction between the electron beam and the sample surface to observe the microstructural characteristics of the material, including but not limited to crystal morphology and pore distribution; The elemental composition of the sample is analyzed using an energy dispersive spectrometer (EDS). By comparing it with a database of known materials, the type of material, including but not limited to masonry, wood, and metal alloys, as well as the relative content of each component, is determined to understand the material properties and provide a reference for material selection for restoration.

6. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 1, characterized in that: The digital reconstruction of the complex and irregularly shaped parts of the historical building in step 3 specifically includes: Import the processed point cloud data into professional 3D modeling software, build a polygon mesh model based on the point cloud, and generate a high-quality, low-polygon 3D model with accurate geometric features through manual topology optimization or automatic reconstruction algorithm; Based on the color data obtained by the colorimeter, the software's material editing function is used to give realistic color textures to each surface area of the model, achieving a high-fidelity visual effect.

7. The method for digitally reconstructing complex and irregularly shaped parts of historical buildings according to claim 6, characterized in that: The establishment of the color information library of the material in step 3 specifically includes: A material color information library is created in professional 3D modeling software, and the material information obtained by the electron microscope analyzer is associated with the corresponding color and stored for easy subsequent retrieval. When restoration or virtual display is required, materials and colors can be matched quickly and accurately to complete the digital reconstruction of complex and special-shaped parts of historical buildings.

Images