System and method for digital restoration of ancient armor based on three-dimensional modeling

Abstract

The application discloses an ancient armor digital restoration system and method based on three-dimensional modeling, relates to the technical field of digital restoration, and comprises the following steps: extracting local thickness and defining a neutral point, and forming a residual part neutral surface by all the neutral points; screening the nearest neighbor set of reliable areas for each damaged area, and defining a target curvature field; defining a reliable connection boundary according to a displacement field, performing rigid body registration on the reliable connection boundaries of different residual parts, screening connection relationship candidates, introducing a binary selection variable to establish a global constraint, defining a global assembly topology set for the connection relationship candidates, and performing boundary closure and difference set correction to form a restoration model. The application combines the use of the geodetic distance nearest neighbor, local coordinate system principal curvature solving and target curvature propagation, can keep the curvature transmission along the residual part body expansion path when there are fractures, curling and edge missing, and avoids the cross-layer misfit caused by the Euclidean proximity.

Description

Technical Field

[0001] This invention relates to the field of digital restoration technology, and in particular to a digital restoration system and method for ancient armor based on three-dimensional modeling. Background Technology

[0002] For ancient armor and other surviving objects composed of numerous armor plates, connecting components, and layered structures, traditional methods of manual surveying, photographic recording, and physical assembly are increasingly difficult to meet the needs of digital display and subsequent restoration and utilization due to the meticulous nature and enormous workload involved. In order to improve efficiency, current related technologies have evolved from simple morphological data collection to a comprehensive processing stage that combines geometric analysis, structural deduction, and digital reconstruction. However, existing technologies often rely on direct modeling of the surface of ancient armor fragments, lacking the ability to identify defects such as rust, bending, and dents. This leads to instability in the subsequent splicing and repair, making it difficult to distinguish between the actual shape and the damaged deformation. This increases the difficulty of restoring the original joints and overlaps between the fragments, resulting in insufficient interpretability and reliability of the restoration results. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention provides a digital restoration system and method for ancient armor based on three-dimensional modeling, which solves the problem that the existing technology lacks the ability to identify conditions such as rust, bending, and dents, which leads to the instability of the subsequent splicing and repair foundation when these conditions are applied to the restoration results. It also makes it difficult to distinguish between the real structure and the damaged deformation, increases the difficulty of restoring the original joints and overlaps between the fragments, and consequently leads to insufficient interpretability and reliability of the restoration results.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a digital restoration system for ancient armor based on three-dimensional modeling, which includes a neutral surface construction module for fragments, which defines the set of vertices and the set of triangular faces in the fragment data of ancient armor in three-dimensional modeling as the three-dimensional surface of the fragments, extracts local thickness and defines neutral points, and assembles all neutral points into a neutral surface of the fragments. The target curvature field construction module calculates the average curvature scalar based on the neutral surface of the residual part and extracts the curvature deviation and absolute thickness deviation. It then determines the damaged area and the reliable area. Based on the determination of the reliable area, it establishes a local coordinate system and calculates the local principal curvature using the least squares method. For each damaged area, it selects the nearest neighbor set of the reliable area and defines the target curvature field. The assembly topology inversion module adjusts the point positions based on the reliable area point set and the nearest neighbor points of the damaged area points, and obtains the corrected center surface by curvature recovery according to the target curvature field. It constructs the three-dimensional surface of the corrected residual part, and forms a displacement field according to the adjusted displacement amount. It extracts the set of free boundary curves, defines reliable connection boundaries according to the displacement field, performs rigid body registration on the reliable connection boundaries of different residual parts, filters connection relationship candidates, introduces binary selection variables to establish global constraints, and defines a global assembly topology set for connection relationship candidates. The restoration model generation module identifies missing boundaries in the global assembly topology set, defines them as regions to be supplemented, generates a supplementary parent body, and performs boundary closure and difference correction to form a restoration model.

[0006] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of extracting local thickness and defining neutral points, and forming a neutral surface of the fragment by combining all neutral points, includes: extracting the front vertex and the unit normal along the front vertex according to the front and back data of the three-dimensional surface of the fragment; querying the corresponding vertex of the back data; calculating the local thickness of the point pair and defining neutral points according to the front vertex and the queried corresponding vertex of the back; and forming a neutral surface of the fragment by combining all neutral points. The calculated local thickness is corrected using the neighborhood median, the median of the local thickness of the entire wreck is extracted, and the thickness deviation is calculated.

[0007] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of calculating the average curvature scalar based on the neutral surface of the fragment and extracting the curvature deviation and absolute thickness deviation to determine the damaged area and the reliable area includes calculating the local average curvature for each vertex of the neutral surface of the fragment using the cotangent form, and calculating the average curvature scalar. Extract the median of the average curvature scalar of the entire wrecked part and calculate the curvature deviation. Simultaneously, for the thickness deviation, extract the median of the thickness deviation of the entire wrecked part and calculate the absolute thickness deviation. The curvature and thickness standard difference constants are calculated for each vertex using curvature deviation and absolute thickness deviation to determine the damaged area and the reliable area.

[0008] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of selecting the nearest neighbor set of reliable areas for each damaged area and defining the target curvature field includes: establishing a local coordinate system in the neighborhood of reliable points determined by the reliable areas, fitting a quadratic surface to the neighborhood points, calculating the surface parameters by the least squares method, and calculating the local principal curvature. The reference average curvature is defined based on the local principal curvature. At the same time, for each damaged point in the damaged area, the nearest neighbor set of the damaged point in the reliable area is selected, and the reference curvature of the nearest neighbor point is defined as the target curvature field after propagation to the boundary of the damaged area.

[0009] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the construction of the three-dimensional surface of the corrected fragment and the formation of a displacement field according to the adjusted displacement include: adjusting the position of the reliable area points set and the nearest neighbor points of the damaged area points and the points of the damaged area points set together, based on the edges of different vertex combinations in the neutral surface, and recording the edge length before correction for each edge; making the difference between the adjusted edge length and the corresponding edge length before correction minimize; and forming an intermediate neutral surface from the adjusted edges. The current local average curvature of the damaged area point set is recalculated based on the intermediate neutral surface, and compared with the difference between the corresponding reference curvature stationary points to obtain the curvature recovery amount. Curvature recovery is performed based on the curvature recovery amount to obtain the corrected center surface. Determine the unit correction normal vector of each vertex of the corrected center face, and determine the vertices of the corrected positive surface and the corrected negative surface respectively to construct the three-dimensional surface of the corrected part, and define the displacement of each vertex of the corrected center face to form a displacement field.

[0010] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of extracting the set of free boundary curves and defining reliable connection boundaries according to the displacement field includes: extracting the set of free boundary curves for each three-dimensional surface of the correction fragment, defining the boundary points of each curve boundary as boundary points belonging to the reliable zone, calculating the median displacement of the curve boundary to which the boundary points belong, and setting a reliable threshold based on the scanning resolution and historical experience to define it as a reliable connection boundary. For two different residuals and any two reliable connection boundaries, define relationship assumptions, including the parallel connection assumption and the overlapping assumption. The parallel connection assumption means that the two boundaries are connected at the end, and the overlapping assumption means that one residual part partially covers another residual. Perform equal arc length sampling on each set of boundaries, pair them according to the parallel connection assumption and the overlapping assumption, and perform rigid body registration. Filter the local connection candidates of each set of boundaries and define them as specific connection relationship candidates.

[0011] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the following steps are taken: introducing binary selection variables to establish global constraints and defining a global assembly topology set for candidate connection relationships, including introducing binary selection variables to establish global constraints for local connection candidates and candidate connection relationships of each group of boundaries, so that a unified boundary can participate in at most one final connection relationship, thereby realizing hierarchical optimization and defining a global assembly topology set by obtaining a binary solution.

[0012] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of identifying missing boundaries as regions to be supplemented and generating a supplementary parent body for the global assembly topology set includes reading all connected boundaries and unclosed boundaries therein. Boundaries that are not occupied by any connection relationships and are located on the outer contour of the assembly are identified as missing boundaries. The area enclosed by the missing boundaries is defined as the area to be filled. For each area to be filled, local configuration rules are defined to generate the filling parent body.

[0013] As a preferred embodiment of the ancient armor digital restoration system based on three-dimensional modeling described in this invention, the step of performing boundary closure and difference set correction to form a restoration model includes performing boundary closure and difference set correction based on the replacement matrix, so that the missing boundary between the replacement matrix and the area to be replaced is continuously closed, while setting constraint correction to obtain the corrected replacement parts corresponding to each area to be replaced, and can be incorporated into the current global assembly topology set to form a restoration model.

[0014] Secondly, the present invention provides a method for digitally restoring ancient armor based on three-dimensional modeling, including: based on the fragment data of ancient armor in three-dimensional modeling, defining the vertex set and triangle face set in the fragment data as the three-dimensional surface of the fragment, extracting the local thickness and defining neutral points, and forming a neutral surface of the fragment by combining all the neutral points. Calculate the average curvature scalar based on the neutral surface of the wreck and extract the curvature deviation and absolute thickness deviation. Determine the damaged area and the reliable area. Establish a local coordinate system based on the determination of the reliable area and calculate the local principal curvature using the least squares method. Select the nearest neighbor set of the reliable area for each damaged area and define the target curvature field. The point positions of the nearest neighbors of the reliable area point set and the damaged area point are adjusted, and the corrected center surface is obtained by curvature recovery according to the target curvature field. The three-dimensional surface of the corrected residual part is constructed, and the displacement field is formed according to the adjusted displacement amount. The set of free boundary curves is extracted, and the reliable connection boundary is defined according to the displacement field. Rigid body registration is performed on the reliable connection boundary of different residual parts, and candidate connection relationships are screened. Binary selection variables are introduced to establish global constraints, and a global assembly topology set is defined for the candidate connection relationships. For the global assembly topology set, missing boundaries are identified and defined as regions to be filled, and a filling parent body is generated. Boundary closure and difference correction are performed to form a restoration model.

[0015] The beneficial effects of this invention are as follows: By combining geodesic nearest neighbor, local coordinate system principal curvature solution and target curvature propagation, curvature transmission can be maintained along the unfolding path of the residual part when there are fractures, curling and edge missing, avoiding cross-layer mismatch caused by Euclidean proximity, and thus enabling the morphological restoration within the same residual part and subsequent cross-residual part assembly to have a unified geometric basis. By combining reliable connection boundary screening, two types of relationship assumptions of parallel and overlapping, rigid body registration and binary global constraints, it is possible to further eliminate the overall contradiction caused by multiple connections on one side and local optimal splicing when the local boundaries seem to be able to match, so that the obtained assembly topology has both local geometric docking capability and global structural consistency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a structural diagram of the ancient armor digital restoration system based on 3D modeling in Example 1.

[0018] Figure 2 This is a flowchart of the method for digitally restoring ancient armor based on 3D modeling in Example 1.

[0019] Figure 3 This is a flowchart illustrating the assembly topology inversion process of the ancient armor digital restoration method based on 3D modeling in Example 1. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] 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 phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, referring to Figures 1-3 This is the first embodiment of the present invention, which provides a digital restoration system for ancient armor based on three-dimensional modeling, including the following steps: S1, the Neutral Surface Construction Module for Residual Parts, is based on the residual data of ancient armor in 3D modeling. It defines the set of vertices and the set of triangles in the residual data as the 3D surface of the residual part, extracts the local thickness and defines neutral points, and combines all the neutral points into the neutral surface of the residual part. S1.1 Based on the front and back data of the three-dimensional surface of the wreck, extract the front vertex and the unit normal along the front vertex respectively, query the corresponding vertex of the back data, calculate the local thickness of the point pair based on the front vertex and the queried back vertex, define the neutral point, and form the neutral surface of the wreck by combining all the neutral points. Since local corrosion or scanning noise can cause burrs in the thickness, the calculated local thickness is corrected using the neighborhood median to obtain the corrected thickness field. The median of the local thickness of the entire part is then extracted to calculate the thickness deviation.

[0024] S2, the target curvature field construction module, calculates the average curvature scalar based on the neutral surface of the residual part and extracts the curvature deviation and absolute thickness deviation, determines the damaged area and the reliable area, establishes a local coordinate system based on the determination of the reliable area and calculates the local principal curvature through the least squares method, selects the nearest neighbor set of the reliable area for each damaged area, and defines the target curvature field; S2.1, For each vertex of the neutral surface of the residual part, the local average curvature is calculated using the cotangent form, and the average curvature scalar is calculated; Extract the median of the average curvature scalar of the entire wrecked part and calculate the curvature deviation. Simultaneously, for the thickness deviation, extract the median of the thickness deviation of the entire wrecked part and calculate the absolute thickness deviation. The curvature and thickness standard deviation constants are calculated for each vertex using curvature deviation and absolute thickness deviation. The curvature and thickness standard deviation constants are calculated based on the calibrated residual data. Curvature anomaly thresholds and thickness anomaly thresholds are set based on the sum of the mean and twice the standard deviation. If the curvature deviation is greater than or equal to the curvature anomaly threshold or the absolute thickness deviation is greater than or equal to the thickness anomaly threshold, the corresponding vertex set is determined to belong to the damaged area. If the vertex does not belong to the damaged area, it is defined as a reliable area. Damaged area and reliable area determination are then performed. Specifically, the calculation of the local mean curvature can be expressed as:

[0025]

[0026] in, This represents the local mean curvature. Let represent the mixed area of ​​a ring around a vertex, and let i represent the vertex index. A ring neighborhood representing a vertex. and Let i and j represent the sides formed by vertices i and j, respectively, and let j represent the included angles between adjacent triangles opposite to those sides. and Let i and j represent vertices respectively. Represents the scalar value of mean curvature; Furthermore, the calculation of the curvature and thickness standard difference constants can be expressed as follows:

[0027]

[0028]

[0029]

[0030]

[0031] in, Indicates thickness deviation. Indicates the corrected thickness field. This represents the median thickness of the local area of ​​the entire remaining part. Indicates absolute thickness deviation. This represents the median of the thickness deviation of the entire defective part. Indicates curvature deviation. This represents the median of the average curvature scalar of the entire remaining component. and These represent curvature deviation and absolute thickness deviation, respectively. This represents the median; By placing the front and back surfaces of the wrecked part, local thickness, neutral surface, curvature anomaly and thickness anomaly into the same judgment link, the geometric disturbance caused by the superposition of corrosion, noise, wrinkles and real configuration can be decomposed into two types of information: correctable deformation and reliable structure. This avoids misjudgment caused by relying solely on surface shape or thickness features, thereby improving the stability of the subsequent restoration benchmark. By using the modified thickness field and the average curvature scalar to jointly constrain the identification of the damaged area, and then using the local quadratic surface fitting result of the reliable area to generate the target curvature field in reverse, it is possible to achieve a directional backtracking from the current state of the damaged part to the original service form. This makes the restoration of the damaged area no longer dependent on overall smoothness or empirical shape repair, but controlled by the continuity of the local real configuration.

[0032] S2.2, for reliable points determined by the reliable region, establish a local coordinate system in the neighborhood, fit a quadratic surface to the neighborhood points, calculate the surface parameters by the least squares method, and calculate the local principal curvature; The reference average curvature is defined based on the local principal curvature. At the same time, for each damaged point in the damaged area, the nearest neighbor set of the damaged point in the reliable area is selected (the nearest neighbor is determined by the geodesic distance theorem), and the reference curvature of the nearest neighbor is defined as the target curvature field after propagation to the boundary of the damaged area. Specifically, for the local coordinate system, for each reliable point, with the point as the origin and the normal of the point as the normal axis of the local coordinate system, two mutually orthogonal tangential axes are constructed in the tangential plane; then the point and its neighboring points are projected into the local coordinate system to obtain the local coordinates of the neighboring points. In the process of establishing a local coordinate system, the expression for the local surface is defined as follows:

[0033] in, Represents a local curved surface. , , , , as well as Here are the parameters of the local quadratic surface, where... This indicates that within the neighborhood of a reliable point, the surface... The strength of the secondary bending in the direction. Represents the surface within the neighborhood of a reliable point. and Coupling bending between two directions This indicates that within the neighborhood of a reliable point, the surface... The strength of the secondary bending in the direction. This indicates that within the neighborhood of a reliable point, the surface... The amount of tilt in the direction. This represents the first tilt of the surface in the v direction within the neighborhood of the reliable point. This represents the constant term of the local surface at the reliable point. and Represents the coordinates in two directions of the local coordinate system; Based on the expression of the local surface, all neighborhood points are substituted to form an overdetermined system of equations. The parameters are then solved using the least squares method based on the sample points, as follows:

[0034]

[0035] in, This represents the coefficient matrix composed of the coordinates of the sample points. This represents the column vector of parameters to be determined. This represents the column vector of normal direction coordinates of the sample points. Indicates transpose calculation; Furthermore, the reference curvature for the nearest neighbor point can be expressed as:

[0036]

[0037]

[0038]

[0039] in, and Let these represent the first and second approximate principal curvatures, respectively. Indicates the reference mean curvature. Represents the nearest neighbor set. Represents the reliable region. represents the target curvature field after propagation to the boundary of the damaged region, and j represents the neighbor point index; By combining geodesic nearest neighbor, local coordinate system principal curvature solution with target curvature propagation, curvature transfer can be maintained along the unfolding path of the fragment body when there are fractures, curling, and edge missing parts. This avoids cross-layer mismatch caused by Euclidean proximity, and thus provides a unified geometric basis for morphological restoration within the same fragment and subsequent cross-fragment assembly. Through the continuous coupling of the above technology chain, a restoration foundation with noise-resistant identification, local reliable modeling, and directional restoration capabilities for damaged areas is finally formed, obtaining original structural information that cannot be directly revealed from the original scan data. This is a combined technology effect that cannot be achieved by using thickness correction, anomaly detection, or surface fitting alone.

[0040] S3, the assembly topology inversion module, adjusts the point positions based on the reliable area point set and the nearest neighbor points of the damaged area points, and obtains the corrected center surface by curvature recovery according to the target curvature field, constructs the three-dimensional surface of the corrected residual part, and forms a displacement field according to the adjusted displacement amount, extracts the set of free boundary curves, defines reliable connection boundaries according to the displacement field, performs rigid body registration on the reliable connection boundaries of different residual parts, filters connection relationship candidates, introduces binary selection variables to establish global constraints, and defines a global assembly topology set for connection relationship candidates; S3.1, for the edges based on different vertex combinations in the neutral surface, and record the edge length before correction for each edge, adjust the position of the points of the damaged area and the nearest neighbor of the damaged area points in the reliable area point set, so that the difference between the adjusted edge length and the corresponding edge length before correction is minimized, and form the intermediate neutral surface with the adjusted edges. The current local average curvature of the damaged area point set is recalculated based on the intermediate neutral surface, and compared with the difference between the corresponding reference curvature stationary point to obtain the curvature recovery amount. Curvature recovery is performed based on the curvature recovery amount, and the position of the damaged area points is adjusted to obtain the corrected center surface. Determine the unit correction normal vector of each vertex of the corrected center face, and determine the vertices of the corrected positive surface and the corrected negative surface respectively to construct the three-dimensional surface of the corrected part, and define the displacement of each vertex of the corrected center face to form a displacement field.

[0041] S3.2 For each three-dimensional surface of the correction residue, extract the set of free boundary curves, and for each curve boundary, define the boundary points that belong to the reliable zone. Calculate the median displacement of the curve boundary to which the boundary points belong. Based on the scanning resolution, set a reliability threshold according to historical experience. If the median displacement is less than or equal to twice the scanning resolution, it is defined as a reliable connection boundary, where twice the scanning resolution can be used as the reliability threshold. For two different residuals and any two reliable connection boundaries, define relationship assumptions, including the parallel connection assumption and the overlapping assumption. The parallel connection assumption means that the two boundaries are connected at the end, and the overlapping assumption means that one residual part partially covers another residual. Perform equal arc length sampling on each set of boundaries, pair them according to the parallel connection assumption and the overlapping assumption, and perform rigid body registration. Filter the local connection candidates of each set of boundaries and define them as specific connection relationship candidates. Specifically, for rigid body registration, it is expressed as:

[0042] in, Indicates the fragment Candidate boundaries mapped to remnants Rigid body transformation on candidate boundaries Indicates the fragment To the fragments The rotation matrix, z represents any point of the boundary sampling point. Indicates the fragment To the fragments Translation vector; Based on the rigid body transformation, the squared distance between the registered point pairs is calculated and minimized as the registration residual, expressed as:

[0043] in, This represents the registration residual, where m represents the total number of sampling points on the current boundary. Indicates the fragment The boundary of the first One sampling point, Indicates the fragment On the boundary with The corresponding sampling points, among which Then the corresponding assumption is followed. This corresponds to the overlapping assumption; Based on minimizing the registration residual, several local connection candidates for each group of boundaries are selected, and specific relationship types are marked as connection relationship candidates, including parallel connection and overlapping connection. S3.3, for each group of boundary local connection candidates and connection relationship candidates, binary selection variables are also introduced to establish global constraints, so that a unified boundary can participate in at most one final connection relationship, thereby realizing hierarchical optimization and defining a global assembly topology set of the obtained binary solution; Furthermore, the rules for performing hierarchical optimization include first maximizing the number of selected valid connections, then minimizing the total boundary gap error under the premise of maximizing the number of connections, and finally obtaining a globally consistent assembly topology set. It should be noted that, in the process of defining the global assembly topology set for binary solutions, the binary variable is defined as follows: if the corresponding connection candidate is selected, it represents 1; otherwise, it represents 0. By continuously coupling the adjustment of damaged area locations, restoration of target curvature, reconstruction of corrected surfaces, and generation of displacement fields, the residual parts can be restored from a disturbed geometric state to a stable state that can be used for connection judgment. Then, the displacement information formed during the restoration process is used as the basis for boundary reliability, thereby unifying morphological correction and connection judgment into the same data link and avoiding the disconnect between correction results and assembly basis. By combining reliable connection boundary screening, parallel and overlapping relationship assumptions, rigid body registration, and binary global constraints, it is possible to further eliminate the overall contradictions caused by multiple connections on one side and local optimal splicing when local boundaries seem to be able to match. This allows the obtained assembly topology to have both local geometric dockability and global structural consistency.

[0044] S4, the restoration model generation module, identifies missing boundaries in the global assembly topology set, defines them as regions to be filled, generates a parent body for filling, and performs boundary closure and difference correction to form a restoration model; S4.1, for the global assembly topology set, read all connected boundaries and unclosed boundaries; Boundaries that are not occupied by any connection relationships and are located on the outer contour of the assembly are identified as missing boundaries. The area enclosed by the missing boundaries is defined as the area to be filled. For each area to be filled, local configuration rules are defined to generate the filling parent body. Specifically, the local configuration rules include the first priority: if there are continuous boundary trends, curvature trends or thickness trends on both sides of the area to be supplemented, the extension result of the adjacent residual boundary is used as the supplementation parent body. Set a second priority: if there are similar armor plates, repeated arrangement units, left-right symmetrical units, or front-back similar units in the current assembly topology, extract the corresponding configuration from these units, perform rigid body mapping, scaling-restricted mapping, or mirror mapping to form a replacement parent body; Set the third priority: If there is no unit that can be directly copied, then use the boundary, normal, curvature and stacking direction of the residual parts around the area to be replaced to construct a local interpolation matrix and generate a continuous configuration of the missing area. S4.2, based on the replacement parent body, perform boundary closure and difference set correction to make the missing boundary between the replacement parent body and the area to be replaced continuously closed. At the same time, set constraint correction to obtain the corrected replacement parts corresponding to each area to be replaced, and can incorporate them into the current global assembly topology set to form a restoration model; Specifically, constraint corrections include ensuring that the replacement parent does not intrude into the solid domain of existing residuals, that the replacement parent is consistent with the current assembly topology in the stacking direction, and that the replacement parent transitions continuously with adjacent residuals in terms of thickness, normal, and local curvature.

[0045] By directly using the assembly topology output for missing boundary identification, hierarchical generation of the replacement parent body, boundary closure, difference set correction, and stacking continuity constraint, the replacement region can be generated by reverse inference from the restored real assembly relationship. During the replacement process, the boundary trend, curvature trend, thickness distribution, and stacking direction of adjacent residual parts are inherited simultaneously, thus obtaining a restoration model that can be incorporated into the existing residual part system. After combining the above technologies, a closed-loop restoration from residual part correction, connection inversion to missing replacement is achieved. This ensures that the final restoration result is derived from the geometric evidence of real residual parts and maintains the continuity of the overall assembly. This chain-like technical effect, driven by displacement credibility to select boundaries and further constrain the generation of replacement parts, cannot be naturally obtained by using each step in parallel.

[0046] This embodiment also provides a method for digitally restoring ancient armor based on 3D modeling, including: Based on the fragment data of ancient armor in 3D modeling, the set of vertices and the set of triangles in the fragment data are defined as the 3D surface of the fragment. Local thickness is extracted and neutral points are defined. All neutral points are combined to form the neutral surface of the fragment. Calculate the average curvature scalar based on the neutral surface of the wreck and extract the curvature deviation and absolute thickness deviation. Determine the damaged area and the reliable area. Establish a local coordinate system based on the determination of the reliable area and calculate the local principal curvature using the least squares method. Select the nearest neighbor set of the reliable area for each damaged area and define the target curvature field. The point positions of the nearest neighbors of the reliable area point set and the damaged area point are adjusted, and the corrected center surface is obtained by curvature recovery according to the target curvature field. The three-dimensional surface of the corrected residual part is constructed, and the displacement field is formed according to the adjusted displacement amount. The set of free boundary curves is extracted, and the reliable connection boundary is defined according to the displacement field. Rigid body registration is performed on the reliable connection boundary of different residual parts, and candidate connection relationships are screened. Binary selection variables are introduced to establish global constraints, and a global assembly topology set is defined for the candidate connection relationships. For the global assembly topology set, missing boundaries are identified and defined as regions to be filled, and a filling parent body is generated. Boundary closure and difference correction are performed to form a restoration model.

[0047] This embodiment also provides a computer device applicable to the case of a digital restoration system for ancient armor based on 3D modeling, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the digital restoration system for ancient armor based on 3D modeling as proposed in the above embodiment. The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0048] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the ancient armor digital restoration system and method based on 3D modeling as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0049] In summary, this invention combines geodesic nearest neighbor, local coordinate system principal curvature solution, and target curvature propagation to maintain curvature transfer along the unfolding path of the remaining part even when there are fractures, curling, and missing edges. This avoids cross-layer mismatch caused by Euclidean proximity, thus providing a unified geometric basis for morphological restoration within the same remaining part and subsequent cross-part assembly. By combining reliable connection boundary screening, parallel and overlapping relationship assumptions, rigid body registration, and binary global constraints, this invention can further eliminate overall contradictions caused by multiple connections on one side and local optimal splicing, even when local boundaries seem to be matchable. This ensures that the obtained assembly topology has both local geometric compatibility and global structural consistency.

[0050] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A digital restoration system for ancient armor based on 3D modeling, characterized in that: include, The Neutral Surface Construction Module for Remains is based on the data of ancient armor remnants in 3D modeling. It defines the set of vertices and the set of triangles in the data of the remnants as the 3D surface of the remnants, extracts the local thickness and defines neutral points, and combines all the neutral points into the neutral surface of the remnants. The target curvature field construction module calculates the average curvature scalar based on the neutral surface of the residual part and extracts the curvature deviation and absolute thickness deviation. It then determines the damaged area and the reliable area. Based on the determination of the reliable area, it establishes a local coordinate system and calculates the local principal curvature using the least squares method. For each damaged area, it selects the nearest neighbor set of the reliable area and defines the target curvature field. The assembly topology inversion module adjusts the point positions based on the reliable area point set and the nearest neighbor points of the damaged area points, and obtains the corrected center surface by curvature recovery according to the target curvature field. It constructs the three-dimensional surface of the corrected residual part, and forms a displacement field according to the adjusted displacement amount. It extracts the set of free boundary curves, defines reliable connection boundaries according to the displacement field, performs rigid body registration on the reliable connection boundaries of different residual parts, filters connection relationship candidates, introduces binary selection variables to establish global constraints, and defines a global assembly topology set for connection relationship candidates. The restoration model generation module identifies missing boundaries in the global assembly topology set, defines them as regions to be supplemented, generates a supplementary parent body, and performs boundary closure and difference correction to form a restoration model.

2. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 1, characterized in that: The step of extracting local thickness and defining neutral points, and forming a neutral surface of the residual part by combining all neutral points, includes: extracting the front vertex and the unit normal along the front vertex according to the front and back data of the three-dimensional surface of the residual part; querying the corresponding vertex of the back data; calculating the local thickness of the point pair and defining neutral points according to the front vertex and the queried corresponding vertex of the back; and forming a neutral surface of the residual part. The calculated local thickness is corrected using the neighborhood median, the median of the local thickness of the entire wreck is extracted, and the thickness deviation is calculated.

3. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 2, characterized in that: The step of calculating the average curvature scalar based on the neutral surface of the residual part and extracting the curvature deviation and absolute thickness deviation to determine the damaged area and the reliable area includes calculating the local average curvature for each vertex of the neutral surface of the residual part using the cotangent form, and calculating the average curvature scalar. Extract the median of the average curvature scalar of the entire wrecked part and calculate the curvature deviation. Simultaneously, for the thickness deviation, extract the median of the thickness deviation of the entire wrecked part and calculate the absolute thickness deviation. The curvature and thickness standard difference constants are calculated for each vertex using curvature deviation and absolute thickness deviation to determine the damaged area and the reliable area.

4. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 3, characterized in that: The process of selecting the nearest neighbor set of reliable regions for each damaged region and defining the target curvature field includes: establishing a local coordinate system in the neighborhood of reliable points determined by the reliable regions, fitting a quadratic surface to the neighborhood points, calculating the surface parameters using the least squares method, and calculating the local principal curvature. The reference average curvature is defined based on the local principal curvature. At the same time, for each damaged point in the damaged area, the nearest neighbor set of the damaged point in the reliable area is selected, and the reference curvature of the nearest neighbor point is defined as the target curvature field after propagation to the boundary of the damaged area.

5. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 4, characterized in that: The construction of the three-dimensional surface of the correction fragment and the formation of a displacement field based on the adjusted displacement include: adjusting the position of the reliable area points set, the damaged area points set, and the nearest neighbor points of the damaged area points in the reliable area points set, so that the difference between the adjusted edge length and the corresponding edge length before correction is minimized, and forming an intermediate neutral surface by adjusting the edges. The current local average curvature of the damaged area point set is recalculated based on the intermediate neutral surface, and compared with the difference between the corresponding reference curvature stationary points to obtain the curvature recovery amount. Curvature recovery is performed based on the curvature recovery amount to obtain the corrected center surface. Determine the unit correction normal vector of each vertex of the corrected center face, and determine the vertices of the corrected positive surface and the corrected negative surface respectively to construct the three-dimensional surface of the corrected part, and define the displacement of each vertex of the corrected center face to form a displacement field.

6. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 5, characterized in that: The extraction of the free boundary curve set and the definition of reliable connection boundaries based on the displacement field include: extracting the free boundary curve set for each three-dimensional surface of the correction residual part; defining the boundary points of each curve boundary as boundary points belonging to the reliable zone; calculating the median displacement of the curve boundary to which the boundary points belong; and setting a reliable threshold based on the scanning resolution and historical experience to define it as a reliable connection boundary. For two different residuals and any two reliable connection boundaries, define relationship assumptions, including the parallel connection assumption and the overlapping assumption. The parallel connection assumption means that the two boundaries are connected at the end, and the overlapping assumption means that one residual part partially covers another residual. Perform equal arc length sampling on each set of boundaries, pair them according to the parallel connection assumption and the overlapping assumption, and perform rigid body registration. Filter the local connection candidates of each set of boundaries and define them as specific connection relationship candidates.

7. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 6, characterized in that: The introduction of binary selection variables to establish global constraints and the definition of a global assembly topology set for candidate connectivity relationships include the introduction of binary selection variables to establish global constraints for local connectivity candidates and candidate connectivity relationships for each group of boundaries, so that a unified boundary can participate in at most one final connectivity relationship, thereby realizing hierarchical optimization and defining a global assembly topology set for the obtained binary solution.

8. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 7, characterized in that: The process of identifying missing boundaries in the global assembly topology set and defining them as regions to be supplemented and generating a supplementation parent body includes reading all connected and unclosed boundaries within it. Boundaries that are not occupied by any connection relationships and are located on the outer contour of the assembly are identified as missing boundaries. The area enclosed by the missing boundaries is defined as the area to be filled. For each area to be filled, local configuration rules are defined to generate the filling parent body.

9. The ancient armor digital restoration system based on three-dimensional modeling as described in claim 8, characterized in that: The process of performing boundary closure and difference set correction to form a restoration model includes performing boundary closure and difference set correction based on the replacement parent body to achieve continuous closure of the missing boundaries between the replacement parent body and the area to be replaced, while setting constraint correction to obtain the corrected replacement parts corresponding to each area to be replaced, and incorporating them into the current global assembly topology set to form a restoration model.

10. A method for digitally restoring ancient armor based on three-dimensional modeling, based on the digital restoration system for ancient armor based on three-dimensional modeling as described in any one of claims 1 to 9, characterized in that: include, Based on the fragment data of ancient armor in 3D modeling, the set of vertices and the set of triangles in the fragment data are defined as the 3D surface of the fragment. Local thickness is extracted and neutral points are defined. All neutral points are combined to form the neutral surface of the fragment. Calculate the average curvature scalar based on the neutral surface of the wreck and extract the curvature deviation and absolute thickness deviation. Determine the damaged area and the reliable area. Establish a local coordinate system based on the determination of the reliable area and calculate the local principal curvature using the least squares method. Select the nearest neighbor set of the reliable area for each damaged area and define the target curvature field. The point positions of the nearest neighbors of the reliable area point set and the damaged area point are adjusted, and the corrected center surface is obtained by curvature recovery according to the target curvature field. The three-dimensional surface of the corrected residual part is constructed, and the displacement field is formed according to the adjusted displacement amount. The set of free boundary curves is extracted, and the reliable connection boundary is defined according to the displacement field. Rigid body registration is performed on the reliable connection boundary of different residual parts, and candidate connection relationships are screened. Binary selection variables are introduced to establish global constraints, and a global assembly topology set is defined for the candidate connection relationships. For the global assembly topology set, missing boundaries are identified and defined as regions to be filled, and a filling parent body is generated. Boundary closure and difference correction are performed to form a restoration model.

Images