A method for cleaning and restoring a disassembled unearthed armor

Abstract

The present application relates to the technical field of cultural relic restoration, and discloses a method for cleaning and restoring unearthed armor in block form, which comprises the following steps: determining the boundary of the armor remains, extracting a standard rope passing hole distance constant and a rope passing path; laying an orthogonal reference coordinate axis to divide the remains into independent matrix blocks and assign coordinate identifiers; performing orthogonal projection two-dimensional dimension reduction drawing on the blocks, extracting the contour and hole coordinates, and calibrating the upper and lower level order vector direction; generating a physical compensation entity and mapping the hole positions, combining the rope passing path to perform simulation rope passing interference detection; calculating the projection area ratio as the upper limit of the global deformation compensation area expansion, establishing a microscopic deformation gradient field according to the hole spacing reduction, and performing nonlinear stretching compensation on the two-dimensional dimension reduction drawing according to the field, and finally performing combination splicing according to the corrected drawing and the coordinate identifiers. The present application can limit the accumulation of deformation errors, eliminate the interference of spatial deformation on the connection logic, and improve the accuracy of armor restoration.

Description

Technical Field

[0001] This invention relates to the field of cultural relic restoration technology, specifically a method for modular cleaning and restoration of unearthed armor. Background Technology

[0002] During the burial process, the unearthed armor was subjected to subsidence and soil compression, resulting in nonlinear spatial deformation, layered folding, and local fractures and missing parts of its physical structure. Current armor restoration work mostly adopts whole extraction combined with manual mapping or conventional 3D scanning to obtain data.

[0003] In the above-mentioned conventional data extraction and restoration operations, the continuously distributed armor remains are treated as a whole. Local geological deformation errors accumulate continuously within the global measurement range, resulting in serious distortion of the overall dimensional tolerance of the final extraction. At the same time, due to the overlapping and occlusion of the extruded and deformed armor plates, directly mapping the hole positions and contour data in the three-dimensional folded state cannot eliminate the dimensional distortion caused by spatial deformation, which in turn leads to errors in the determination of the overlapping sequence and connection topology between the armor plates. In addition, the existing restoration process lacks reverse compensation calculations for deformation data and interference verification mechanisms before physical assembly. When operators directly assemble the physical components based on drawings containing deformation errors and faulty connection logic, spatial collisions and misalignment of rope holes occur between the armor plates, leading to physical assembly interference and restoration failure. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a modular cleaning and restoration method for unearthed armor, which solves the problems of accumulated restoration errors and mismatched connection relationships caused by geological compression and deformation of unearthed armor.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for modular cleaning and restoration of unearthed armor, comprising the following steps: The distribution boundary of the armor remains was determined, and the types of armor plates contained in the armor remains were identified. Based on the types of armor plates, corresponding standard rope hole distance constants were extracted to construct a standard parameter feature set, and physical threading paths were extracted. Orthogonal reference coordinate axes were set up, and the armor remains were divided into independent matrix blocks and assigned grid coordinate labels. Orthogonal projection was performed on the independent matrix blocks to generate a two-dimensional reduced-dimensional drawing. The two-dimensional projection contours of the armor plates and the two-dimensional plane coordinates of the rope holes in the two-dimensional reduced-dimensional drawing were extracted, and the hierarchical order vector directions between adjacent armor plates were marked in the two-dimensional reduced-dimensional drawing. A physical compensation entity was generated based on the two-dimensional projection contours. Based on the standard rope hole distance constants corresponding to the current armor plate type in the standard parameter feature set, mapping holes were generated on the physical compensation entity. Based on the physical threading paths and the hierarchical order vector directions, simulated threading flexible lines were used in conjunction with the mapping holes to perform simulated threading interference detection. The ratio of the theoretical to the measured projected area of ​​the independent matrix blocks is calculated as the upper limit of the global deformation compensation area expansion. A micro-deformation gradient field is established based on the two-dimensional plane coordinates of the rope hole and the actual reduction distance of the standard rope hole distance constant of the corresponding type. Nonlinear stretching compensation is performed on the two-dimensional reduced drawing based on the micro-deformation gradient field to generate a corrected two-dimensional reduced drawing. Combination and splicing are performed based on the corrected two-dimensional reduced drawing and the grid coordinate identifier.

[0006] Furthermore, extracting the physical connection path specifically includes: extracting the remaining traces of the connecting elements of the card plates, tracing the connection direction of the connecting traces between adjacent card plates; using the two-dimensional plane coordinates of the rope-threading holes as the basic nodes of the network, and using the physical connection path between the two-dimensional plane coordinates of adjacent rope-threading holes and the directed connection relationship formed by the upper and lower level order vector directions as the network path; constructing a typical connection logic topology graph based on the basic nodes of the network and the network path; and using the typical connection logic topology graph to determine the connection path between the physical compensation entities; this method transforms the card plate connection pattern into a graph theory network model composed of nodes and directed edges, providing a basis for determining the connection path for subsequent interference detection.

[0007] Furthermore, dividing the armor remains into independent matrix blocks and assigning grid coordinate identifiers specifically includes: dynamically determining the adaptive orthogonal grid spacing formed by the intersection of the orthogonal reference coordinate axes based on the shape, size, and spatial distribution of armor plates of different types; using the adaptive orthogonal grid formed by the intersection to define the armor remains as a discrete array set; binding the horizontal column coordinate parameters and the vertical row coordinate parameters to establish a two-dimensional relative grid coordinate system; using the spatial aggregation range of the independent matrix blocks with two-dimensional relative grid coordinates to characterize the total domain within the distribution boundary; and using the two-dimensional relative grid coordinates as the grid coordinate identifier. This method establishes a spatial distribution system of matrix blocks through grid segmentation, discretizes and calibrates irregular armor remains, and provides coordinate references for the spatial positioning of local drawings in global splicing.

[0008] Furthermore, when calibrating the hierarchical order vector direction between adjacent pieces in the two-dimensional reduced drawing, discrete feature state additional coding is further performed: a first state scalar is assigned to pieces that are inverted or flipped to trigger flip correction; a second state scalar is assigned to pieces with broken or incomplete features to trigger incomplete contour completion; a third state scalar is assigned to pieces that are cut off by the orthogonal reference coordinate axis and cross over to the adjacent independent matrix block, and the third state scalar is used as an identifier to trigger the boundary splicing alignment of the adjacent independent matrix block; this method uses discrete state scalars to record the physical anomaly features of the pieces, so that different deformation states can directly trigger the corresponding flip correction, completion and cross-boundary splicing operations of the system.

[0009] Furthermore, during the simulated threading interference detection, a closed-loop verification operation is performed: a local three-dimensional coordinate system is established, and the three-axis coordinate components of the first and second associated mapped holes in the stacked state are extracted in the local three-dimensional coordinate system; the three-dimensional straight-line spatial distance is calculated based on the difference between the three-axis coordinate components of the first and second associated mapped holes, and the three-dimensional straight-line spatial distance is defined as the actual threading chord length of the simulated threading flexible line passing between the associated holes; this method uses the local spatial coordinate system and the Euclidean distance algorithm to quantify the spatial span between holes and establish a data standard for determining whether mechanical interference occurs between entities.

[0010] Furthermore, the closed-loop verification specifically includes: when the calculated actual threaded chord length exceeds a preset standard extreme value constraint threshold, or when the simulated threaded flexible line collides and interferes with the physical compensation entity, the current region's topology is determined to be invalid; a reorganization and error correction mechanism is triggered to perform reverse flipping and reorganization on the upper and lower level order vector directions of the physical compensation entity. The reorganization and error correction mechanism is executed cyclically until the actual threaded chord length is within the preset standard extreme value constraint threshold range and no interference occurs. This method uses the calculated chord length and entity collision feedback to reversely investigate logical errors caused by previous projection occlusion, ensuring the accuracy of the plate stacking order.

[0011] Furthermore, the simulated threading interference detection process also includes a reverse filling operation: extracting the two-dimensional planar coordinates of the threading holes that have completed the simulated threading interference detection and are spatially adjacent to the missing area as the starting anchor point; obtaining the standard stitching step length and physical connection direction in the physical threading path to construct a step length vector; translating the spatial coordinates of the starting anchor point along the direction of the step length vector by a standard stitching step length to derive and generate the spatial coordinates of the virtual mounting node to be filled; this method uses the positions of adjacent verification anchor points and the standard stitching step length to derive the node coordinates of the missing hole, thereby achieving the position filling of the missing area.

[0012] Furthermore, the ratio of the theoretical to the measured projected area of ​​the independent matrix block is calculated as the upper limit of the global deformation compensation area expansion. Specifically, the sum of the theoretical projected areas of the armor plates within the independent matrix block in the non-curled state is obtained as the theoretical projected area, and the value of the projected area loss due to compression is obtained as the measured projected area. The theoretical projected area and the measured projected area are divided by a ratio calculation, and the result of the ratio calculation is used as the constraint extreme value of the upper limit of the global deformation compensation area expansion. This method, by calculating the regional area loss ratio, clarifies the overall shrinkage degree of local locations, thus defining the global compensation boundary for nonlinear stretching compensation.

[0013] Furthermore, a micro-deformation gradient field is established based on the actual reduction distance between the two-dimensional plane coordinates of the rope-threading holes and the standard rope-threading hole spacing constant of the corresponding type. Specifically, this involves: identifying the actual reduction distance between the two-dimensional plane coordinates of adjacent rope-threading holes in the two-dimensional reduced drawing; performing a division ratio operation between the actual reduction distance and the standard rope-threading hole spacing constant of the current armor plate type; calculating the local strain differential weighting coefficient of the corresponding point in the continuous coordinate system of the drawing plane using the constant minus the ratio operation result; and constructing the micro-deformation gradient field covering the independent matrix block using the local strain differential weighting coefficient values ​​of specific points. This method quantifies the non-uniform compression state of each local region by comparing the changes in hole spacing, providing differential numerical basis for mesh deformation.

[0014] Furthermore, nonlinear stretching compensation is performed on the two-dimensional reduced drawing to generate a corrected two-dimensional reduced drawing. Specifically, this involves: extracting the local strain differential weighting coefficients in the microscopic deformation gradient field; locking the coordinates of flat areas below a preset deformation threshold; and performing mesh stretching transformation using a bilinear interpolation function for abnormal strain areas above the preset deformation threshold. The mesh stretching transformation is performed iteratively until the difference between the two-dimensional plane coordinate spacing of adjacent rope holes and the standard rope hole spacing constant of the corresponding type is within a preset tolerance range, and the total area increment of the drawing reaches the upper limit of the global deformation compensation area expansion, thus completing the generation of the corrected two-dimensional reduced drawing. This method performs local stretching on abnormal areas based on deformation gradient field data, eliminates dimensional deviations caused by geological compression at the level of the two-dimensional reduced drawing, and outputs a splicing template that conforms to the original parameters.

[0015] This invention provides a method for the modular cleaning and restoration of unearthed armor. It has the following beneficial effects: 1. This invention divides continuous large-area armor remains into independent matrix blocks by setting up orthogonal reference coordinate axes, which changes the situation where deformation errors accumulate globally during traditional whole extraction and restoration; it uses discretized grid blocks for independent data extraction and mapping, which restricts irregular physical deformation caused by geological subsidence to local coordinate domain, thereby reducing the computational load of subsequent data processing and providing a stable two-dimensional relative grid coordinate reference for global combination and splicing.

[0016] 2. This invention generates physical compensation entities by two-dimensional projection contours and performs simulated cross-linking interference detection. The calculated actual cross-linking chord length and physical interference phenomenon are used as the triggering conditions for reconstructive error correction. This closed-loop verification mechanism uses the physical spatial constraints of entity connection to reverse-check and correct the misjudgment of plate occlusion and stacking order caused by dimensional reduction orthogonal projection, ensuring the feasibility of the reconstructed connection topology in the actual physical assembly process.

[0017] 3. This invention constructs a microscopic deformation gradient field by actually reducing the spacing of the projected holes, and limits the upper limit of stretching expansion by combining the ratio of measured and theoretical areas, and performs nonlinear stretching compensation on the two-dimensional reduced drawing. This mechanism can perform differentiated bilinear interpolation mesh stretching according to the specific degree of compression of each local area of ​​the drawing, quantitatively offsetting the dimensional deviation caused by the compression of the armor on the two-dimensional plane, and providing a flat splicing template that conforms to the original hole spacing tolerance for the final restoration work. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the modular cleaning and restoration system for unearthed armor according to the present invention; Figure 2 This is a schematic diagram of the main process of the modular cleaning and restoration method for unearthed armor according to the present invention; Figure 3 This is a schematic diagram of the global measurement and grid discretization process of the present invention; Figure 4 This is a schematic diagram of the attached vector constraint two-dimensional topology dimensionality reduction mapping process of the present invention; Figure 5 This is a schematic diagram of the entity compensation closed-loop verification and reverse deduction steps of the present invention; Figure 6 This is a schematic diagram of the tolerance decoupling and nonlinear tensile compensation steps of the present invention; Figure 7 This is a schematic diagram comparing the core assessment parameters of the traditional manual method and the modular restoration system of the present invention.

[0019] Among them: 100, benchmark determination module; 200, mesh generation module; 300, two-dimensional mapping module; 400, solid compensation module; 500, tolerance decoupling module. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] See attached document Figure 1 This invention provides a modular cleaning and restoration system for unearthed armor. The system is implemented on a restoration operation platform, which includes: The benchmark measurement module 100 includes a micro-distance measuring caliper and a morphological calibration rod, which is used to measure the overall boundary of the unearthed armor remains, extract the solid geometric boundary dimensions of the local armor plate units that have not undergone nonlinear deformation, and the standard rope threading hole spacing set composed of at least one of the transverse hole spacing, longitudinal hole spacing and oblique hole spacing. The grid division module 200, including physical marker lines and endpoint anchors, is used to suspend orthogonal reference coordinate axes above the unearthed armor remains and establish micro-matrix block areas that block the cumulative error of geological deformation. The two-dimensional mapping module 300 includes a rigid transparent medium plate and a trajectory recording carrier, which is used to lay on the surface of the micro matrix block, extract the two-dimensional projection contour and hole plane coordinates under the vertical field of view, and record the hierarchical stacking relationship. The entity compensation module 400 includes a cuttable rigid sheet and a simulated threading flexible line, which is used to establish a physical operation entity based on the corresponding contour extracted by the two-dimensional mapping module 300, and to perform three-dimensional spatial stacking and mechanical tensile self-consistency verification. The tolerance decoupling module 500 includes an image photoelectric acquisition peripheral and a data processing unit with a logic operation processor. It is used to scan the recording surface of the two-dimensional mapping module 300, calculate the physical micro-deformation gradient field, and execute local two-dimensional coordinate grid stretching compensation instructions on the two-dimensional reduced-dimensional drawing data output by the two-dimensional mapping module 300 and after scanning vectorization.

[0022] Based on the hardware support of the restoration operation platform disclosed above, this invention enables reverse analysis and physical restoration of unearthed armor with complex burial distortion and local loss characteristics.

[0023] See attached document Figure 2 This invention provides a method for modular cleaning and restoration of unearthed armor, comprising the following steps: S100 uses the benchmark measurement module 100 to delineate the global two-dimensional distribution boundary of the armor remains, measure the standard outer dimensions, hole topology and standard rope hole spacing set of locally intact armor plates, and map the physical connection path and stacking relationship of the in-situ armor plate array. S200: Using the mesh division module 200, cross orthogonal lines are drawn based on the original row and column directions of the armor plate entities to divide the global armor remains into independent matrix blocks distributed in an array, and assign corresponding mesh coordinate labels to each matrix block. S300, the two-dimensional mapping module 300 is covered on the surface of a single matrix block, and the two-dimensional outline of the single piece of armor and the two-dimensional coordinates of the hole in the current projection state are depicted by orthographic projection. The upper and lower hierarchical order vector directions of adjacent solid armor pieces are marked on the surface of the two-dimensional mapping module 300. S400, the entity compensation module 400 generates the corresponding physical compensation entity by cutting the two-dimensional projection contour output by the two-dimensional mapping module 300. The two-dimensional coordinates of the hole in the current projection state extracted by the two-dimensional mapping module 300 are used as the initial hole position. The standard rope hole spacing set extracted by the benchmark measurement module 100 is used as the hole position correction constraint. The hole is mapped and drilled on the physical compensation entity. The entity threading step is simulated to detect the hole position interference blockage and fill the missing unit in reverse. S500, the tolerance decoupling module 500 obtains the theoretical projected standard area and the total two-dimensional measured projected area including extrusion deformation, calculates the shrinkage rate of the local hole projection spacing to establish the deformation compensation gradient, applies nonlinear stretching compensation to the two-dimensional reduced drawing data output by the two-dimensional mapping module 300 and after scanning vectorization, and performs overall physical compensation entity splicing based on the two-dimensional coordinates after stretching compensation.

[0024] In the initial stage of this method, the operator calls the benchmark measurement module 100 to define the extent of the remains in the complex original buried soil and extract the original armor plate parameters with intact geometric characteristics. This step solidifies the armor plate data that has not been damaged by burial stress, providing benchmark parameters under ideal conditions for subsequent deduction steps.

[0025] Subsequently, a grid division module 200 is introduced to deploy a mandatory orthogonal grid above the armor remains. The presence of grid lines changes the traditional continuous progress mode in clearing, blocking large areas of remains containing geological subsidence and deformation into matrix blocks of controllable size. This operation limits the measurement errors, projection errors, and compensation errors caused by geological subsidence and burial compression to independent matrix blocks, reducing their cumulative diffusion during the global restoration and splicing process.

[0026] After obtaining discrete matrix blocks, the in-situ topographic data is extracted using the 2D mapping module 300. The edge position of the plate is projected vertically onto the rigid transparent medium plate, and the compression and folding state in three-dimensional space is compressed and transformed into two-dimensional planar drawing data. The operator records the layering relationship in the spatial thickness direction by drawing directional vector symbols, thus completing the transformation of physical entities into two-dimensional information with attribute constraints.

[0027] To address the issues of information occlusion and logical judgment errors during the two-dimensional projection process, the method flows into the operation domain of the entity compensation module 400. A physical verification entity is established using a proportionally cut planar thin sheet to eliminate physical interference from the in-situ environment. The operator uses a flexible line to pass through the holes in the rigid thin sheet, based on the recorded connection path, to create a rigid dynamic detection environment. If the overlapping layers recorded in the previous projection cause the flexible line to be unable to pass through the holes smoothly or deviate from the normal connectivity limit, then a topological mismatch is determined at that point and a reorganization correction is performed.

[0028] After completing the single-block error correction, the hierarchical stacking sequence, hole connection path, and virtual missing unit coordinates verified by the entity compensation module 400 are written back to the two-dimensional reduced drawing data. Then, the two-dimensional reduced drawing data is transferred to the tolerance decoupling module 500 to perform the final tolerance elimination. The tolerance decoupling module 500 compares the theoretical ideal area with the actual projected area reduced due to the curling of the plate. The data processing unit identifies the non-uniform reduction ratio of the distance between two holes in the grid and constructs the deformation gradient value accordingly. The data processing unit calculates the required two-dimensional tensor expansion ratio of each local area based on the gradient value, generates a non-uniform deformation grid, and unfolds the squeezed two-dimensional reduced drawing into the original flat state at the digital level. The entity component assembly is completed based on the reconstructed flat coordinate domain.

[0029] See attached document Figure 3 In specific implementations, steps S100 and S200 provided by the present invention may include the following steps: S101, the operator uses the measuring tools in the benchmark measurement module 100 to probe along the boundary area between the armor remains and the buried soil in the field or laboratory environment. After the boundary is clearly defined by the probe, a two-dimensional physical boundary bounding box is set in the outer space containing the scattered armor pieces. The area within the bounding box is defined as the global two-dimensional distribution boundary. For the soil surface cleaning and boundary demarcation operation, those skilled in the art can use conventional soft brushes and bamboo sticks to remove the boundary. The above-mentioned boundary cleaning method is a well-known technology in the field.

[0030] S102, within the global two-dimensional distribution boundary, select armor plates and armor plate arrays of different types without obvious folding deformation as standard reference samples. The operator uses measuring tools to classify and extract the outer contour geometric boundary dimensions of the standard reference samples of each type, and measures the straight-line physical distance between the centers of adjacent rope holes on different types of samples, recording them as the standard rope hole distances corresponding to each type of armor plate, and jointly constructing a standard parameter feature set. This standard rope hole distance characterizes the original spatial physical parameters of the armor before it is damaged by burial stress. Its numerical range depends on the objective measurement specifications of the actual armor plates. In one embodiment of the present invention, the value range of the standard rope hole distance is set between 2.0 mm and 12.0 mm. This parameter serves as the calculation constant condition for generating physical operation cards and establishing deformation difference mathematical models in subsequent stages.

[0031] S103, the reference measurement module 100 extracts the original remaining traces of the stitching or the wear marks of the perforation in the local armor plate array, uses the morphological calibration rod to track the physical connection direction of the stitching between adjacent armor plates, and for areas where the organic material decays and the stitching is lost, it calls the empirical topological rules of the same type of armor plate for comparison, infers the hidden back connection direction, and records the upper and lower layer stacking order of the armor plate entities due to the original stitching relationship. By combining the acquired data on the connection direction and stacking order, a typical connection logic topology diagram is constructed. The principle of the topology diagram is essentially to abstract the physical laws of entity connection into a directed network of nodes and paths. This typical connection logic topology diagram is used to determine the connection order between entity compensating sheets, the path of flexible lines, the generation direction of virtual lost units, and the standard connection relationship between adjacent hole nodes. This connection logic topology diagram is digitally defined based on a graph theory mathematical model, and its construction formula is set as follows: ; In the formula: This represents a typical chain logic topology diagram. This represents the set of nodes containing the rope-threading holes for the armor plates extracted from the standard reference sample. This represents the set of directed edges formed by the physical connectivity paths and stacking order directions between adjacent rope-threading hole nodes.

[0032] S201, after the reference parameters are extracted, the mesh generation module 200 is introduced to perform physical discretization preprocessing on the armor entity within the global two-dimensional distribution boundary. Since the original stitching state of the armor pieces presents a longitudinal and transverse arrangement, the operator sets two perpendicular orthogonal reference coordinate axes based on the original row and column directions shown by the armor pieces. The endpoint anchors in the mesh generation module 200 are used to fix the armor remains around them, and physical marker lines with physical tension are pulled to make them suspend on the surface of the armor entity.

[0033] S202, multiple horizontal and vertical physical marker lines intersect to form an orthogonal grid. The system dynamically adjusts the grid density based on the shape and size of the mainstream armor plate types in the current division area to form an adaptive grid spacing, so that a single independent matrix block contains a preset number of armor plate units. When an independent matrix block crosses the boundary of armor plates of different types, it automatically accommodates the extraction of multiple types and sizes. It is preferred that a single independent matrix block contains 8 to 70 armor plate units. The grid divides the large-area continuous global armor remains into array-arranged independent matrix blocks. Continuous remains are blocked within their respective independent matrix blocks to limit the measurement error, projection error, and compensation error caused by excavation deformation within the independent matrix blocks, reducing their accumulation and diffusion to the surrounding areas. Then, combined with the grid division mechanism, a spatial distribution mathematical model of the matrix blocks is constructed. Through mathematical expression, irregular physical remains are forcibly normalized into discrete array sets.

[0034] S203: Assign an identification code to each independent matrix block formed by grid segmentation, and establish corresponding two-dimensional relative grid coordinates based on the orthogonal grid system. The identification code and relative grid coordinates are bound together to form the spatial positioning attribute of the matrix block. In the subsequent global stitching step, the local dimensionality reduction drawing performs a spatial reset operation based on the relative grid coordinates. Its matrix distribution mapping formula is set as follows: ; In the formula: This represents the total domain of armor remains within the global two-dimensional distribution boundary; Represents an independent matrix block cell with specific two-dimensional relative grid coordinates within a matrix array; This represents the lateral column coordinate parameter in an orthogonal mesh system; This represents the longitudinal row coordinate parameter in an orthogonal grid system.

[0035] See attached document Figure 4 In specific implementations, step S300 provided by the present invention may include the following steps: S301, the operator uses a micro cleaning tool to remove non-body attached soil or rust from the surface and pores of the independent matrix block, peeling off the attached material layer by layer to expose the physical outline of the individual armor plate in its in-situ buried state and the edge morphology of the rope hole. After the cleaning operation is completed, the rigid transparent medium plate in the two-dimensional mapping module 300 is combined with a level measuring tool and suspended in a horizontal reference posture by a support in the physical space directly above the matrix block. A preset distance is maintained between the rigid transparent medium plate and the armor plate to avoid contact and pressure on the remains. For the surface soil removal and rust removal methods involved in the archaeological cleaning operation, those skilled in the art can use conventional ultrasonic bone cleaning machines or physical removal tools to carry out the operation. The micro-surface cleaning method is a well-known technology in the field.

[0036] S302, the observer performs orthographic projection observation along the field of view perpendicular to the rigid transparent medium plate, and draws the two-dimensional outer edge contour of the single armor piece in the current physical posture on the surface of the rigid transparent medium plate. The two-dimensional planar coordinate parameters of the rope hole in the projection state are extracted. This operation process is essentially to reduce the geometric information of the physical armor piece with three-dimensional spatial undulation and different tilt angles into planar drawing data. The two-dimensional projection contour and hole coordinates obtained by dimensional reduction are directly used as the boundary input parameters for subsequent generation of entity compensation card size cutting and physical punching.

[0037] S303, on the two-dimensional reduced drawing surface of the rigid transparent medium plate, a directed graph of the overlapping relationship between adjacent armor plate areas that overlap and occlude in physical space is established. The operator selects the extracted two-dimensional plane coordinates of the rope hole as the reference point for topological connection, and uses directional vector arrow symbols to connect the reference points of adjacent armor plates. According to the direction of the directional vector arrow, the layer overlapping order of the corresponding armor plates in the thickness direction of physical space is defined. The armor plate projection area represented by the beginning node of the directional vector arrow is set to overlap the armor plate projection area represented by the end node on the vertical axis of physical space. The above steps convert the three-dimensional layering relationship obtained by vision into objective two-dimensional marking instructions.

[0038] S304, combining the generation mechanism of the overlay relationship directed graph, constructs the corresponding graph theory mathematical model expression. At the principle level, it completes digital abstraction by binding the extracted drawing orientation vector direction with the existing hierarchical state of the entity through numerical attributes. This overlay relationship directed graph is digitally logically defined based on a mathematical graph theory network model, and its calculation formula is set as follows: ; In the formula: To construct a two-dimensional directed graph of overlapping relationships; This is the set of A-plate entity unit nodes generated by orthographic projection within the current matrix block; This is a set of directed edges constructed using directional vector arrow symbols to indicate the order of stacking of the entity armor plates.

[0039] S305, independent discrete feature state scalars are introduced into the drawing surface generated by the two-dimensional mapping module 300 to perform physical state additional coding. A first state scalar is assigned to the armor plate unit that has been upside down or flipped and misaligned. A second state scalar is assigned to the armor plate unit with mechanical fracture or severe mineralization defects. A third state scalar is assigned to the edge armor plate that is cut off by the physical grid and whose main spatial area crosses the adjacent block. Various feature state scalars are saved synchronously with the two-dimensional projection contour data and are used as execution conditions for triggering the inference algorithm and splicing alignment in the subsequent reverse inference of missing areas and splicing of adjacent drawing space. Among them, the first state scalar is used to trigger the verification of the armor plate flipping or upside down. The second state scalar is used to trigger the incomplete contour completion and virtual missing unit inference. The third state scalar is used to trigger the splicing of adjacent matrix block boundaries and the alignment of cross-boundary armor plate contours.

[0040] See attached document Figure 5 In specific implementations, step S400 provided by the present invention may include the following steps: S401, based on the two-dimensional reduction drawing output by the two-dimensional mapping module 300, the outer edge contour is extracted, and the cuttable rigid sheet is cut at a 1:1 scale to generate the corresponding physical compensation entity. The two-dimensional coordinates of the hole in the current projection state extracted from the two-dimensional reduction drawing are used as the initial hole position, and the standard rope hole distance set extracted by the benchmark measurement module 100 is used as the hole position correction constraint. Spatial position mapping and penetration drilling operations are performed on the surface of the physical compensation entity. For the specific processing implementation of the sheet shape cutting and mechanical drilling, those skilled in the art can use conventional laser cutting machines or CNC micro punching machines. The physical cutting method of the two-dimensional contour is a well-known technology in this field.

[0041] S402, extract the directed graph of the superposition relationship attached to the surface of the two-dimensional reduced drawing, and call the typical connection logic topology graph constructed by the benchmark measurement module 100 to determine the connection path between adjacent hole nodes. According to the hierarchical superposition order defined by the directional vector arrows on the drawing, the previously generated physical compensation entities are stacked layer by layer on the three-dimensional physical operation table. The simulated connection flexible line penetrates the corresponding mapped hole position of the physical compensation entity under superposition. The operation process reproduces the connection action of the armor remnant entity in the physical space and constructs a physical dynamic detection environment with mechanical tension constraint characteristics.

[0042] S403 introduces a hole-position spatial connectivity distance determination model to perform dynamic closed-loop verification. The determination principle essentially quantifies the mechanical tension span between physical hole positions by calculating the three-dimensional Euclidean distance between two spatial nodes. Specifically, it calculates the actual spatial span of the simulated connecting flexible line traversing between two related physical compensation physical hole positions and defines it as the actual connecting chord length. A local three-dimensional coordinate system is established using the reference plane of the three-dimensional physical operating platform. coordinates and The coordinates are read from the placement position of the physical compensation entity in the three-dimensional physical manipulation table plane positioning mesh. The coordinates are determined by the stacking levels of the physical compensation entities, the thickness of a single rigid sheet, and the height of the shims, or read by a depth measuring ruler, laser rangefinder, or 3D scanning equipment. The spatial span verification formula is set as follows: ; In the formula: This represents the actual chord length of the simulated threaded flexible wire; This represents the abscissa component parameter of the first associated mapped aperture in the local three-dimensional coordinate system under stacked conditions. This represents the ordinate component parameter of the first associated mapped aperture in the local three-dimensional coordinate system under stacked conditions. This represents the vertical height component parameter of the first associated mapped aperture in the local three-dimensional coordinate system under stacked conditions. This represents the abscissa component parameter of the second associated mapped aperture in the local three-dimensional coordinate system under stacked conditions; This represents the ordinate component parameter of the second associated mapped aperture in the local three-dimensional coordinate system under stacked conditions. This represents the vertical height component parameter of the second associated mapped hole position in the local three-dimensional coordinate system under stacked conditions.

[0043] S404, the system calculates and generates a standard extreme value constraint threshold based on the standard lacing hole spacing parameters extracted by the benchmark measurement module 100 and the preset allowable stacking thickness of the armor plate entity in physical space; when the calculated actual lacing chord length variable... When the value exceeds the standard extreme value constraint threshold or when the simulated flexible line collides with the surface of the solid sheet, triggering spatial interference occlusion, the topological failure of the two-dimensional stacking order in the current region is determined. The error correction mechanism is triggered to reverse and reorganize the hierarchical stacking order of the local physical compensation entity. The reorganization step is repeated until the connection state of the simulated flexible line is restored to within the constraint threshold range and no spatial interference physical phenomenon occurs.

[0044] S405, for large-area entity loss areas in a severely mineralized and incomplete state, extracts the verified and spatially adjacent transition boundary A-piece entities from the entity compensation module 400 as starting anchor points. Using the principle of spatial vector translation, the positions of unknown nodes are derived based on known node coordinates and standard spacing. The generation of virtual mounted nodes terminates at the matrix block boundary, adjacent verified A-piece boundary, global two-dimensional distribution boundary, or adjacent block splicing boundary. Its derivation formula is set as follows: ; In the formula: Represents the spatial coordinates of the virtual mount node to be filled; Represents the verified spatial coordinates extracted from the strongly correlated starting anchor point; This represents the standard bracing step length vector extracted from a typical linked logic topology diagram. This standard bracing step length vector comprehensively includes the physical connection path direction between adjacent armor plate entities, the standard rope hole spacing span, and the layer stacking thickness parameter, so as to mathematically represent the three-dimensional spatial bracing pattern of the physical armor plates.

[0045] See attached document Figure 6 In specific implementations, step S500 provided by the present invention may include the following steps: S501, the operator calls the image photoelectric acquisition peripheral in the tolerance decoupling module 500 to perform orthogonal surface optical scanning on the previously generated single-block two-dimensional reduced drawing. During scanning, the grid coordinate identifiers, scale marks, and positioning marks corresponding to at least two endpoint anchors of the matrix block are acquired simultaneously to establish the conversion relationship between pixel coordinates and actual physical coordinates. The scanned raster image data is transmitted to the data processing unit with a logic operation processor inside the tolerance decoupling module 500 for vectorization feature extraction. For the optical scanning and line vectorization feature extraction operations on the drawing surface, those skilled in the art can use conventional CCD industrial cameras and edge detection operators to perform the processing. This optical-digital conversion method is a well-known technology in the field.

[0046] S502, the data processing unit extracts the measured total projected area parameter of the independent matrix block and retrieves the theoretical projected standard area. The system compares the above two area parameters to calculate the macroscopic decoupling coefficient, which serves as the upper limit constraint condition for the global deformation compensation area expansion of a single block. The theoretical projected standard area value is objectively established by the sum of the theoretical planar unfolded areas of the armor plates contained in the block. It is obtained by multiplying the geometric boundary dimension of the outer contour of the standard reference armor plate extracted by the benchmark measurement module 100 with the statistical number of armor plate units contained in the independent matrix block, or by retrieving it from the same type of armor plate sample database. The calculation formula for its macroscopic decoupling coefficient is set as follows: ; In the formula: The macroscopic decoupling coefficient of a single block is represented. In one embodiment of the present invention, the value range of the macroscopic decoupling coefficient of a single block is set to be between 1.05 and 1.35. This represents the theoretical standard total area of ​​the solid plates within the block in their uncurled state; It represents the total two-dimensional measured projected area, including the loss of projected area, due to geological compression and curling.

[0047] S503, the data processing unit further identifies the actual reduction distance between the centers of any two adjacent projected holes within the two-dimensional reduced drawing. The system compares this actual reduction distance with the standard rope hole distance extracted from the previous benchmark. The deformation gradient field principle essentially quantifies the degree of compression in each region of space through the displacement difference of discrete nodes. Based on the comparison results, the nonlinear strain difference weighting coefficient of the local spatial node pair is calculated, and a microscopic deformation gradient field covering the entire block drawing area is constructed accordingly. The mathematical model of its strain difference weighting coefficient is set as follows: ; In the formula: The local strain differential weighting coefficient represents a specific point located in a two-dimensional continuous coordinate system on the drawing plane. In one embodiment of the present invention, the value range of the local strain differential weighting coefficient is limited to a closed interval between 0 and 1. This represents the actual reduced spacing between adjacent projected holes measured at the corresponding coordinate points; This represents the standard rope hole spacing constant, which is objectively determined by the standard reference sample and matches the armor plate type of the current coordinate point. The horizontal axis component parameter represents the horizontal axis component in the two-dimensional continuous coordinate system of the drawing plane; This refers to the vertical axis component parameter in the two-dimensional continuous coordinate system of the drawing plane.

[0048] S504, the data processing unit applies a nonlinear stretching compensation command to the planar coordinate system of the two-dimensional reduced drawing based on the constructed micro-deformation gradient field distribution values. The system locks the coordinates of flat areas in the micro-deformation gradient field where the weight values ​​are lower than the preset deformation judgment threshold. In one embodiment of the present invention, the preset deformation judgment threshold is set to a value between 0.05 and 0.10. For abnormal strain areas where the weight values ​​are higher than the threshold, a local mesh expansion stretching transformation positively correlated with the weight coefficient is performed. At the algorithm level, a bilinear interpolation function is used to process the stretching process of the primitive array. The stretching compensation operation continues to be executed until the difference between the local hole spacing and the standard rope hole spacing is within the preset tolerance range. In one embodiment of the present invention, the preset tolerance range is set to within ±2% of the standard rope hole spacing value, and the total area increment reaches the upper limit of the macroscopic decoupling coefficient.

[0049] S505: The system uses the two-dimensional reduced-dimensional drawings of each block after nonlinear stretching compensation as the underlying physical alignment reference template. Based on the absolute grid coordinates of the transition boundary armor plates at the junction of adjacent matrix blocks, which are converted from relative grid coordinates and matrix block row and column numbers on the compensated drawings, the system performs global spatial splicing of local drawings of multiple blocks. When adjacent matrix blocks contain part of the outline of the same cross-boundary armor plate, the third state scalar of the cross-boundary armor plate is used as the splicing trigger mark, and its compensated outline junction point, hole center point or outer edge feature point is used as the registration point. The operator refers to the coordinate system of the distortion-free two-dimensional reduced-dimensional drawings generated by splicing to physically assemble and connect all local physical compensation entities. At the junction boundary of each block, the simulated through-link flexible line is used to perform mechanical tensile coupling verification to complete the physical combination and restoration of the whole excavated armor entity.

[0050] To aid in understanding the technical solution of this invention, the following provides an application example and test example of a modular cleaning and restoration method for unearthed armor.

[0051] To verify the restoration effect of the present invention, this application embodiment selects the iron armor fragments that have been geologically squeezed and deformed from archaeological excavations as the initial operation object. The operator uses the endpoint anchors of the mesh generation module 200 and the physical marker lines to set up intersecting orthogonal reference coordinate axes, and divides the continuous entity remains, which are 95 cm long and 73 cm wide, into a mesh, extracting 13 independent matrix blocks. After physical exploration, it was confirmed that the armor plate entities inside the matrix block numbered in the second row and third column exhibited extremely strong nonlinear curling spatial characteristics. The benchmark measurement module 100 calls the micro-distance measuring caliper to extract the standard rope hole distance of the local armor plate units in the area outside the specific matrix block where no nonlinear deformation has occurred. The measured standard rope hole distance value is 5.0 mm. The system directly uses the 5.0 mm standard rope hole distance as the constant calculation constant parameter benchmark for subsequent bottom local two-dimensional coordinate mesh tolerance decoupling and physical compensation entity space interference verification.

[0052] In the subsequent data processing flow for a specific independent matrix block in the second row and third column, the total measured projection area constant, including compression loss, obtained by the two-dimensional mapping module 300 under the vertical field of view through orthographic projection is 420 square centimeters. The data processing unit objectively establishes the theoretical projection standard area constant as 500 square centimeters based on the total theoretical planar unfolded area of ​​the A-plate units contained in the independent matrix block. The data processing unit with a logic operation processor divides 500 square centimeters by 420 square centimeters to calculate the macroscopic decoupling coefficient of the single block as 1.19. The actual reduction distance between the centers of adjacent projection holes extracted by the system's photoelectric scanning recognition of the local overlapping area inside the two-dimensional reduced drawing is 4.1 millimeters. The data processing unit performs a difference comparison operation between the actual reduction distance of 4.1 millimeters and the standard rope hole distance of 5.0 millimeters to calculate the nonlinear strain differential weighting coefficient of the local spatial node pair as 0.18.

[0053] The data processing unit determines that the 0.18 weight value obtained from the calculation of the local strain region is higher than the system's set deformation judgment threshold of 0.08. The data processing unit outputs a local two-dimensional coordinate grid stretching compensation instruction that is positively correlated with the 0.18 weight coefficient to the two-dimensional reduced drawing plane coordinate system. The data processing unit calls the bilinear interpolation function to iteratively process the nonlinear stretching process of the drawing array of the abnormal strain region. The actual spacing of the local holes continues to expand during the stretching compensation operation. When the actual spacing of the local holes expands to 4.95 mm, the system extracts the difference between it and the standard rope hole spacing of 5.0 mm as 0.05 mm. This 0.05 mm difference is within the preset tolerance range of ±2%, and the total area increment of the matrix block reaches the upper limit condition of the macroscopic decoupling coefficient of 1.19. The data processing unit terminates the stretching compensation instruction and outputs the corrected two-dimensional reduced drawing as an alignment template.

[0054] Test example: The experimental subject was a complete, interconnected iron armor replica with standard geometric dimensions, meticulously replicated based on the parameters of real unearthed armor plates. The operator laid the complete, interconnected iron armor replica flat at the bottom of a rigid experimental landfill box with a length of 120 cm and a width of 80 cm. The operator filled the rigid experimental landfill box with a 50 cm thick layer of mixed clay to completely cover and enclose the iron armor replica. The operator activated a heavy hydraulic mechanical device with a pressure monitoring and feedback module to apply a continuous static compression constant physical load of 50 MPa to the mixed clay layer inside the landfill box from directly above. The continuous static compression physical load forced the complete, interconnected iron armor replica at the bottom of the landfill box and its internal armor plate arrays to undergo spatial misalignment, distortion, and overlapping and adhesion of the hierarchical structure, exhibiting physical characteristics of these changes.

[0055] Operators removed the static compression constant physical load and extracted the simulated remains of the iron armor, which had undergone spatial morphological distortion, from the bottom of the landfill box. Using mechanical cutting tools, the operators divided the simulated remains into a left and right half-test area, each with an area of ​​2400 square centimeters, along the longitudinal central axis. The operators designated the left half-test area as the control group and performed the restoration operation using a combination of conventional manual mapping and direct two-dimensional planar physical splicing. The operators designated the right half-test area as the experimental group and used a modular cleaning restoration system to sequentially perform grid physical segmentation, entity compensation closed-loop verification, and tolerance decoupling stretching data compensation operations. During the two independent restoration operations, an external industrial-grade photoelectric scanning instrument collected real-time data on the hole position alignment characteristics of the entity compensation sheet and the spatial mechanical physical interference frequency data of the simulated threaded flexible line. The test results are shown in the table below.

[0056] Comparison Table of Core Assessment Parameters between Traditional Manual Methods and Modular Restoration System From the table above and appendix Figure 7 The results show that: the experimental group's global boundary maximum offset parameter value was 2.17 mm after executing the local mesh nonlinear stretching command in the tolerance decoupling module 500; the control group's global boundary maximum offset parameter value was 18.63 mm in the environment without mesh splitting mechanism and stretching intervention; the experimental group's local hole misalignment rate was 1.83% and the mechanical interference frequency was 2 times after calling the entity compensation module 400 to perform physical closed-loop verification; the control group's local hole misalignment rate was 14.27% and the mechanical interference frequency was 29 times when using direct two-dimensional reduced drawing measurement docking method; the experimental group's comprehensive physical restoration time was 13.6 hours after checking for mismatch errors using the data system; and the control group's comprehensive physical restoration time was 47.4 hours after repeated disassembly and reassembly in the later verification stage.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for modular cleaning and restoration of unearthed armor, characterized in that, The method includes the following steps: The distribution boundary of the armor remains was determined, the types of armor plates contained in the armor remains were identified, and the standard rope hole distance constants corresponding to the armor plate types were extracted to construct a standard parameter feature set, and the physical rope connection path was extracted. Orthogonal reference coordinate axes are set up to divide the armor remains into independent matrix blocks and assign grid coordinate labels; Perform orthographic projection on the independent matrix blocks to generate a two-dimensional reduced drawing, extract the two-dimensional projection outline of the armor plate and the two-dimensional plane coordinates of the rope hole in the two-dimensional reduced drawing, and mark the upper and lower hierarchical order vector directions between adjacent armor plates in the two-dimensional reduced drawing; A physical compensation entity is generated based on the two-dimensional projection contour. A mapping hole position is generated on the physical compensation entity based on the standard rope hole distance constant corresponding to the current armor plate type in the standard parameter feature set. Simulated threading interference detection is performed using a simulated threading flexible line in conjunction with the mapping hole position based on the physical threading path and the upper and lower level sequence vector direction. The ratio of the theoretical to the measured projected area of ​​the independent matrix blocks is calculated as the upper limit of the global deformation compensation area expansion. A micro-deformation gradient field is established based on the two-dimensional plane coordinates of the rope hole and the actual reduction distance of the standard rope hole distance constant of the corresponding type. Nonlinear stretching compensation is performed on the two-dimensional reduced drawing based on the micro-deformation gradient field to generate a corrected two-dimensional reduced drawing. Combination and splicing are performed based on the corrected two-dimensional reduced drawing and the grid coordinate identifier.

2. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, The steps for extracting the physical connection path include: Extract the remaining traces of the connecting elements of the card plates and trace the connection direction of the connecting traces between adjacent card plates; use the two-dimensional plane coordinates of the rope holes as the basic nodes of the network, and use the physical connection path between the two-dimensional plane coordinates of adjacent rope holes and the directed connection relationship formed by the upper and lower level order vector directions as the network path. Construct a typical connecting logic topology graph based on the basic nodes of the network and the network path, and use the typical connecting logic topology graph to determine the passage path between the physical compensation entities.

3. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, The steps of dividing the armor remains into independent matrix blocks and assigning grid coordinate identifiers include: Based on the shape, size, and spatial distribution of different types of armor plates, the adaptive orthogonal grid spacing formed by the intersection of the orthogonal reference coordinate axes is dynamically determined. The adaptive orthogonal grid formed by the intersection is used to define the armor remains as a discrete array set. The horizontal column coordinate parameters and the vertical row coordinate parameters are bound together to establish a two-dimensional relative grid coordinate system. The spatial aggregation range of the independent matrix blocks with two-dimensional relative grid coordinates is used to characterize the total domain within the distribution boundary. The two-dimensional relative grid coordinates are used as the grid coordinate identifier.

4. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, When determining the hierarchical order vector direction between adjacent A-pieces in the two-dimensional reduced-dimensional drawing, further discrete feature state additional coding is performed: For armor pieces that are inverted or flipped, a first state scalar is assigned to trigger flip correction; for armor pieces with broken or incomplete features, a second state scalar is assigned to trigger incomplete contour completion; for cross-boundary armor pieces that are cut off by the orthogonal reference coordinate axis and cross over to the adjacent independent matrix block, a third state scalar is assigned, and the third state scalar is used as an identifier to trigger the boundary splicing and alignment of the adjacent independent matrix block.

5. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, When performing the simulated paper-through interference detection, a closed-loop verification operation is performed: A local three-dimensional coordinate system is established, and the three-axis coordinate components of the first associated mapping hole and the second associated mapping hole in the stacked state are extracted in the local three-dimensional coordinate system. The three-dimensional straight-line spatial distance is calculated based on the difference of the three-axis coordinate components of the first associated mapping hole and the second associated mapping hole, and the three-dimensional straight-line spatial distance is defined as the actual chord length of the simulated connecting flexible line passing through the associated holes.

6. The method for modular cleaning and restoration of unearthed armor according to claim 5, characterized in that, The closed-loop verification steps include: When the calculated actual threaded chord length exceeds the preset standard extreme value constraint threshold, or when the simulated threaded flexible line collides and interferes with the physical compensation entity, the current region's topology is determined to be invalid; a reorganization and error correction mechanism is triggered, and the upper and lower hierarchical order vector directions of the physical compensation entity are reversed and reorganized. The reorganization and error correction mechanism is executed cyclically until the actual threaded chord length is within the preset standard extreme value constraint threshold range and no interference occurs.

7. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, The process of performing the simulated cross-linking interference detection also includes a reverse filling operation: Extract the two-dimensional plane coordinates of the rope-threading holes that have completed the simulated threading interference detection and are spatially adjacent to the lost area as the starting anchor point; obtain the standard stitching step length and physical connection direction in the physical threading path to construct the step length vector; translate the spatial coordinates of the starting anchor point along the direction of the step length vector by a standard stitching step length to derive and generate the spatial coordinates of the virtual mounting node to be filled.

8. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, The ratio of the theoretical to the measured projected area of ​​the independent matrix blocks is calculated as the upper limit of the global deformation compensation area expansion. The steps are as follows: The theoretical planar unfolded area of ​​the armor plates within the independent matrix block in the uncurled state is obtained as the theoretical projection standard area. The value of the compressed area, including the loss of projection area, is obtained as the two-dimensional measured total projection area. The theoretical projection standard area and the two-dimensional measured total projection area are divided by a ratio calculation, and the ratio calculation result is used as the macroscopic decoupling coefficient. The macroscopic decoupling coefficient is used as the upper limit of the global deformation compensation area expansion.

9. The method for modular cleaning and restoration of unearthed armor according to claim 1, characterized in that, A micro-deformation gradient field is established based on the two-dimensional planar coordinates of the rope-threading hole and the actual reduced spacing of the standard rope-threading hole distance constant of the corresponding type. The steps are as follows: Identify the actual reduction distance between the two-dimensional plane coordinates of adjacent rope holes in the two-dimensional reduced drawing, and perform a division ratio calculation between the actual reduction distance and the standard rope hole distance constant of the current armor plate type. By subtracting the ratio from the constant, the local strain differential weighting coefficient of the corresponding point in the continuous coordinate system of the drawing plane is calculated; the micro-deformation gradient field covering the independent matrix block is constructed using the local strain differential weighting coefficient of the corresponding point.

10. The method for modular cleaning and restoration of unearthed armor according to claim 9, characterized in that, The steps for generating a corrected two-dimensional reduced-dimensional drawing by performing nonlinear stretching compensation on the two-dimensional reduced-dimensional drawing are as follows: Extract the local strain differential weighting coefficients in the micro-deformation gradient field, lock the coordinates of flat areas below the preset deformation judgment threshold, and perform mesh stretching transformation using a bilinear interpolation function for abnormal strain areas above the preset deformation judgment threshold; cyclically execute the mesh stretching transformation until the difference between the two-dimensional plane coordinate spacing of adjacent rope holes and the standard rope hole spacing constant of the corresponding type is within the preset tolerance range, and the total area increment of the drawing reaches the upper limit of the global deformation compensation area expansion, thus completing the generation of the corrected two-dimensional reduction drawing.

Images