Historic building digitization method and system based on feature description

Abstract

The invention discloses an ancient building digitization method and system based on feature description, and relates to the technical field of ancient building digitization, and the method comprises the steps: abstract description of line components, extraction of an ancient building skeleton, weighting of feature points of the line components, and automatic generation of simple ancient paintings. The font data of the corresponding component of each ancient building is analyzed, the corresponding skeleton of the ancient building is extracted, the corresponding weighting is performed on each feature point, and finally the components in various shapes are generated, so that the digitization efficiency of the ancient building is further improved, the ancient building is abstracted to the maximum extent, and the digitization efficiency of the ancient building is improved. Various schemes and measures are provided for research and application of historic building digitization, the problems of complexity and time consumption in traditional digitized historic buildings are solved, dynamic manufacturing of various line components is achieved, the defect of traditional methods for historic building artistic feature expression is overcome, and diversified historic building appearance display is presented.

Description

Technical Field

[0001] The present invention relates to the technical field of ancient building digitization, and in particular to an ancient building digitization method and system based on feature description. Background Art

[0002] With the continuous maturity of digital technology, it has been widely used in the field of ancient architecture. Through feature-based description, ancient buildings can be digitized and dynamic production of various line components can be achieved, making up for the shortcomings of traditional methods in expressing the artistic characteristics of ancient buildings. At the same time, ancient buildings can be intelligently displayed digitally to better preserve the integrity of ancient buildings, making the display of ancient buildings more intelligent and diversified, reducing potential damage to the actual ancient buildings, and promoting sustainable protection of ancient cultural heritage.

[0003] Prior art, such as the invention patent application with announcement number CN116882031A, discloses a method and system for building a building model based on point cloud, obtains a first cuboid and a second cuboid according to the point cloud data of the building; uses the points in each first cuboid and the points in each second cuboid as inputs of the PointNet model to obtain the feature values ​​of each first cuboid and the feature values ​​of each second cuboid; determines the first feature and the second feature according to the first cuboid and the second cuboid corresponding to the point in the point cloud; fuses the first feature, the second feature, the global feature of the point cloud and the feature of the point to obtain the comprehensive feature of the point; segments the point cloud according to the comprehensive feature of each point in the point cloud; performs surface reconstruction on the segmented point cloud, and obtains the three-dimensional model of the building after texture mapping. The present invention can effectively improve the accuracy of the reconstruction of the three-dimensional model of the building.

[0004] In view of the above scheme, the applicant of the present invention has found that the above technical problems at least exist in the following technical problems: the above problems are mainly to segment the point cloud according to the comprehensive features of each point in the point cloud; to reconstruct the surface of the segmented point cloud, and to obtain the three-dimensional model of the building after texture mapping, thereby effectively improving the accuracy of the reconstruction of the three-dimensional model of the building, but not establishing a corresponding catalog comparison table, and not analyzing the glyph data of the corresponding components of each ancient building, so that the conversion from ancient buildings to data cannot be realized, effective data information cannot be obtained, the skeleton information of the corresponding line components of each ancient building cannot be found, the corresponding ancient building skeleton cannot be extracted, and reference data cannot be provided for the subsequent generation of ancient building components. At the same time, the feature point information of the line skeleton corresponding to the ancient building skeleton cannot be obtained, and the weighting of important feature points of the ancient building cannot be performed, and the digital generation of ancient buildings cannot be realized, thereby reducing the diversity of the display of ancient buildings. Summary of the invention

[0005] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a method and system for digitizing ancient buildings based on feature description.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a method for digitizing ancient buildings based on feature description, including: Step 1, abstract description of line components: by copying the appearance of the graphic elements of each ancient building, the feature points and topological structure information of the corresponding components of each ancient building are obtained, and the feature points and topological structure information of the corresponding components of each ancient building are saved in a description library file, a corresponding catalog comparison table is established, and the glyph data of the corresponding components of each ancient building are obtained by analysis.

[0007] Step 2: Extracting the skeleton of ancient buildings: Find out the skeleton information of line components from the glyph data of the corresponding components of each ancient building, and extract the corresponding skeleton of the ancient building.

[0008] Step 3: Weighting of feature points of line components: Obtain feature point information of the line skeleton corresponding to the skeleton of the ancient building, and analyze and obtain the weight corresponding to each feature point.

[0009] Step 4: Automatically generate simple ancient paintings: Receive weighted feature point information of each line component and generate components of various shapes.

[0010] In a second aspect, the present invention provides an ancient building digitization system based on feature description, including: a line component abstract description module, which is used to copy the graphic appearance of each ancient building, and then obtain the feature points and topological structure information of the corresponding components of each ancient building, and save the feature points and topological structure information of the corresponding components of each ancient building in a description library file, establish a corresponding directory comparison table, and analyze to obtain the glyph data of the corresponding components of each ancient building.

[0011] The ancient building skeleton extraction module is used to find out the skeleton information of the line components from the glyph data of the corresponding components of each ancient building, and extract the corresponding ancient building skeleton.

[0012] The line component feature point weighting module is used to obtain the feature point information of the line skeleton corresponding to the ancient building skeleton, and analyze and obtain the weight corresponding to each feature point.

[0013] The automatic generation module of simple ancient paintings is used to receive the weighted feature point information of each line component and generate components of various shapes.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method and system for digitizing ancient buildings based on feature description, by establishing a corresponding catalog comparison table, analyzing the glyph data of the corresponding components of each ancient building, extracting the corresponding ancient building skeleton, performing corresponding weighting on each feature point, and finally generating components of various shapes, further improving the efficiency of digitizing ancient buildings, fully abstracting the ancient buildings to the greatest extent, providing a variety of solutions and measures for the research and application of digitizing ancient buildings, solving the deficiencies in the current technology, realizing the automatic generation of the skeleton of the ancient building through a unique feature description method, obtaining the corresponding simple ancient painting, and realizing the maximum degree of abstraction of the ancient building, solving the complexity and time-consuming problems existing in the traditional digitization of ancient buildings, greatly simplifying the digitization process, improving the efficiency of digitizing ancient buildings, simplifying the complexity of data processing, realizing the dynamic production of a variety of line components, making up for the deficiencies of the traditional method in expressing the artistic characteristics of ancient buildings, and permanently preserving the cultural value of the ancient buildings in a digital form, leaving behind a valuable cultural heritage, creating innovative and modern works of art, injecting new vitality into traditional culture, and realizing a diverse display of ancient building appearances. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 The present invention is a schematic flow chart of the steps for implementing the method.

[0017] Figure 2 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1As shown, a method for digitizing ancient buildings based on feature description includes: step one, abstract description of line components: by copying the appearance of the graphic elements of each ancient building, the feature points and topological structure information of the corresponding components of each ancient building are obtained, and the feature points and topological structure information of the corresponding components of each ancient building are saved in a description library file, a corresponding catalog comparison table is established, and the glyph data of the corresponding components of each ancient building are obtained by analysis.

[0020] It should be noted that geometric feature points include corners, intersections, tangent points, etc., which are usually used to represent the boundaries and shapes of components. Material features include different characteristics of materials such as walls, roofs, and columns. Decorative features include specific positions and forms such as carvings, patterns, and ornaments.

[0021] It should be noted that the connection relationship refers to the connection method between components, such as the connection between beams and columns, the relationship between walls and roofs, etc. The hierarchical structure refers to the levels of the building, such as the relationship between different levels such as foundation, wall, roof, etc. The spatial relationship refers to the position, orientation, and direction of components in three-dimensional space.

[0022] As an optional implementation method, the feature points and topological structure information of the components corresponding to each ancient building are obtained, and the specific acquisition process is as follows: first, the corresponding camera equipment is set up using an unmanned aerial vehicle, and then corresponding shooting is performed on each ancient building to obtain the corresponding primitive image of each ancient building. Secondly, the human-computer interactive component description technology is used to copy the photographed primitive image corresponding to each ancient building to obtain the corresponding primitive appearance of each ancient building, and the feature set corresponding to each historical component stored in the database is extracted, and then the feature points and topological structure information of the corresponding component are extracted from the primitive appearance corresponding to each ancient building.

[0023] The features corresponding to the components in the corresponding primitive appearance of each ancient building are compared with the reference extraction feature set corresponding to each historical component. If a feature corresponding to the component in the corresponding primitive appearance of a certain ancient building is in the reference extraction feature set corresponding to a certain historical component, the feature is used as the extracted feature point, and the feature points of the corresponding components of each ancient building are obtained by comparison and analogy, wherein the feature points include geometric feature points, material features and decorative features, and based on the extracted feature points, the topological structure connection diagram between the corresponding components of each ancient building is obtained, and the topological structure information of the corresponding components of each ancient building is extracted, wherein the topological structure information includes connection relationship, hierarchical structure and spatial relationship, so as to obtain the feature points and topological structure information of the corresponding components of each ancient building.

[0024] As an optional implementation method, the establishment of the corresponding directory comparison table is specifically carried out as follows: the feature points of the corresponding components of each ancient building in the description library file are compared with the feature points of each historical component corresponding to each attribute stored in the database. If the feature point of the corresponding component of a certain ancient building in the description library file is the same as the feature point of a certain historical component corresponding to a certain attribute stored in the database, then the attribute is used as the feature point attribute of the feature point of the corresponding component of the ancient building, and so on, the feature point attributes of the feature points of the corresponding components of each ancient building are obtained. Similarly, the topological relationship of the topological structure of the corresponding components of each ancient building can be obtained, and a directory comparison table of the corresponding components of each ancient building is established based on the feature point attributes and topological relationships, and the attributes and relationships of the corresponding components of each ancient building are displayed one by one in the form of a table, and the attributes and relationships of the corresponding components of each ancient building are given identifiers to establish the corresponding directory comparison table.

[0025] It should be noted that the basic attributes include shape, length, direction and color, the line types include straight lines and curves, and the primitive vectors include lines, rectangles, circles, etc.

[0026] It should be noted that model building software includes 3dmax and CAD.

[0027] As an optional implementation, the analysis obtains the glyph data of the components corresponding to each ancient building. The specific analysis process is as follows: the comprehensive attributes of the components corresponding to each ancient building are obtained from the catalog comparison table, and the appearance attributes in the comprehensive attributes corresponding to the components are removed, the basic attributes corresponding to the components are extracted, and the corresponding basic models are obtained by integrating the model building software, the corresponding constituent lines in the basic model are regarded as primitives, the primitive vectors corresponding to each line component are obtained, and the common features of the primitive vectors corresponding to each line component are extracted, wherein the common features include the starting point, end point, line type and width, and the above are analyzed to obtain the glyph data of the components corresponding to each ancient building.

[0028] Step 2: Extracting the skeleton of ancient buildings: Find out the skeleton information of line components from the glyph data of the corresponding components of each ancient building, and extract the corresponding skeleton of the ancient building.

[0029] As an optional implementation method, the skeleton information of the line components is found from the glyph data of the corresponding components of each ancient building. The specific search process is as follows: the corresponding skeleton keywords in the directory comparison table are input, and the directory in the component library is searched. Since the ancient building data is collected in the order of component drawing when executing information collection, and the lines are marked, the data of each component includes the type of each line, the number of lines and the topological relationship of the skeleton information. By performing corresponding reverse processing on the ancient building data, when the data is automatically processed, the corresponding keyword input matching search is continued to be performed to find the skeleton information corresponding to the line components, wherein the skeleton information includes geometric curvature, vertical and horizontal quantities of node connections, material thickness and position, and the skeleton information obtained is subjected to comprehensive data normalization processing to obtain the comprehensive skeleton data value of each line component.

[0030] As an optional implementation, the extraction obtains the corresponding ancient building skeleton, and the specific extraction process is as follows: the comprehensive skeleton data value of each line component is compared with the reference skeleton data value corresponding to each ancient building skeleton stored in the database. If the comprehensive skeleton data value of a certain line component is the same as the reference skeleton data value corresponding to a certain ancient building skeleton stored in the database, then the ancient building skeleton is taken as the ancient building skeleton corresponding to the line component, and the corresponding ancient building skeleton is extracted by analogy with this comparison.

[0031] It should be noted that the skeleton data values ​​corresponding to the skeletons of each ancient building are obtained, and the skeleton data values ​​corresponding to the skeletons of each ancient building are calculated based on the average value to obtain the average skeleton data value, and the average skeleton data value is used as the reference skeleton data value.

[0032] Step 3: Weighting of feature points of line components: Obtain feature point information of the line skeleton corresponding to the skeleton of the ancient building, and analyze and obtain the weight corresponding to each feature point.

[0033] It should be noted that edge detection algorithms include Canny edge detection, etc., applied corner detection algorithms include Harris corner detection and Shi-Tomasi detection, etc., morphological skeleton extraction algorithms include Zhang-Suen thinning algorithm, etc., and deep learning technologies include the use of convolutional neural networks, etc.

[0034] As an optional implementation, the feature point information of the line skeleton corresponding to the skeleton of the ancient building is obtained, and the specific acquisition process is as follows: first, edge detection is performed on the ancient building image, and the contour edge corresponding to the ancient building is identified by using an edge detection algorithm, so as to obtain the main structural lines of the skeleton corresponding to the ancient building; secondly, the corner points in the ancient building image are extracted based on the application of a corner point detection algorithm; finally, the shape of the ancient building image is skeletonized by using a morphological skeleton extraction algorithm, so as to extract the simplified shape and structural features corresponding to the ancient building, and deep learning technology is used to perform automated feature point extraction to obtain the feature point information of the line skeleton corresponding to the skeleton of the ancient building, wherein the feature point information includes the main structural lines, corner points, simplified shapes and structural features, and data normalization is performed on the feature point information to obtain the feature points of the line skeleton corresponding to the skeleton of the ancient building.

[0035] As an optional implementation method, the analysis obtains the weight corresponding to each feature point, and the specific analysis process is as follows: compare each feature point of the line skeleton corresponding to the ancient building skeleton with the feature point type of each weighted type corresponding to the historical ancient buildings stored in the database. If a feature point of the line skeleton corresponding to the ancient building skeleton is different from the feature point type of a weighted type corresponding to the historical ancient building stored in the database, then the feature point of the line skeleton corresponding to the ancient building skeleton is not weighted. If a feature point of the line skeleton corresponding to the ancient building skeleton is the same as the feature point type of a weighted type corresponding to the historical ancient building stored in the database, then the feature point of the line skeleton corresponding to the ancient building skeleton is weighted. By comparing and analogizing, the weights corresponding to each feature point are obtained.

[0036] Step 4: Automatically generate simple ancient paintings: Receive weighted feature point information of each line component and generate components of various shapes.

[0037] It should be noted that the attributes corresponding to each feature point include position and direction, etc., and the weighting methods include TFIDF and Gaussian weighting, etc.

[0038] It should be noted that weighted processing is performed on each feature point: the attributes corresponding to each feature are compared with the attribute sets corresponding to each weighting method. If the attribute corresponding to a feature is within the attribute set corresponding to a weighting method, then the weighting method is used as the weighting method for the feature point, and so on to obtain the weighting method for each feature point. Weighted processing is automatically performed on each feature point based on the loading technology corresponding to the screened weighting method.

[0039] It should be noted that the line component can be drawn and presented in any form of a straight line, a curve or a curve segment.

[0040] As an optional implementation, the components of various shapes are generated, and the specific generation process is as follows: based on the weighted selection corresponding to each feature point, and according to the attributes corresponding to each feature point, each feature point is weighted. After the weighted processing corresponding to each feature point is completed, an initial contour is constructed based on the weighted feature point information, and the initial contour is refined by connecting adjacent feature points or using an interpolation method, and the corresponding redundancy is removed. At the same time, the contour is smoothed using an optimization algorithm to generate corresponding line components, and then the generated line components are combined and spliced ​​accordingly to form a corresponding component shape. In this way, analogy is performed to generate components of various shapes.

[0041] See also Figure 2 As shown, a digitization system of ancient buildings based on feature description includes a line component abstract description module, an ancient building skeleton extraction module, a line component feature point weighting module, a simple ancient painting automatic generation module and a database.

[0042] The ancient building skeleton extraction module is respectively connected to the line component abstract description module, the line component feature point weighting module and the database, and the line component feature point weighting module is respectively connected to the simple ancient painting automatic generation module and the database.

[0043] The line component abstract description module is used to copy the appearance of the graphic elements of each ancient building, and then obtain the characteristic points and topological structure information of the corresponding components of each ancient building. The characteristic points and topological structure information of the corresponding components of each ancient building are saved in the description library file, and the corresponding directory comparison table is established to analyze and obtain the glyph data of the corresponding components of each ancient building.

[0044] The ancient building skeleton extraction module is used to find out the skeleton information of the line components from the glyph data of the corresponding components of each ancient building, and extract the corresponding ancient building skeleton.

[0045] The line component feature point weighting module is used to obtain the feature point information of the line skeleton corresponding to the ancient building skeleton, and analyze and obtain the weight corresponding to each feature point.

[0046] The automatic generation module of simple ancient paintings is used to receive the weighted feature point information of each line component and generate components of various shapes.

[0047] The database is used to store feature points, topological structure information, reference extraction feature sets, basic attributes, line types, common features, skeleton information and feature point information.

[0048] The embodiment of the present invention establishes a corresponding catalog comparison table, analyzes the glyph data of the corresponding components of each ancient building, extracts the corresponding ancient building skeleton, performs corresponding weighting on each feature point, and finally generates components of various shapes, thereby further improving the efficiency of ancient building digitization, fully abstracting the ancient building to the greatest extent, and providing a variety of solutions and measures for the research and application of ancient building digitization, solving the deficiencies in the current technology, and realizing the automatic generation of the skeleton of the ancient building through a unique feature description method, and obtaining the corresponding simple ancient painting, and realizing the maximum degree of abstraction of the ancient building, solving the complexity and time-consuming problems existing in the traditional digitization of ancient buildings, greatly simplifying the digitization process, improving the efficiency of ancient building digitization, simplifying the complexity of data processing, and realizing the dynamic production of multiple line components, making up for the deficiencies of traditional methods in expressing the artistic characteristics of ancient buildings, and permanently preserving the cultural value of ancient buildings in a digital form, leaving behind valuable cultural heritage, creating innovative and modern works of art, injecting new vitality into traditional culture, and realizing a diversified display of ancient building appearances.

[0049] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A method for digitizing ancient buildings based on feature description, characterized in that: include: Step 1: Abstract description of line components: By copying the primitive appearance of each ancient building, the characteristic points and topological structure information of the corresponding components of each ancient building are obtained, and the characteristic points and topological structure information of the corresponding components of each ancient building are saved in the description library file, and the corresponding catalog comparison table is established to analyze and obtain the glyph data of the corresponding components of each ancient building; Step 2: Extracting the skeleton of ancient buildings: Finding the skeleton information of line components from the glyph data of the corresponding components of each ancient building, and extracting the corresponding skeleton of the ancient building; Step 3: Weighting of feature points of line components: obtaining feature point information of the line skeleton corresponding to the skeleton of the ancient building, and analyzing and obtaining the weight corresponding to each feature point; Step 4: Automatically generate simple ancient paintings: Receive weighted feature point information of each line component and generate components of various shapes.

2. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The characteristic points and topological structure information of the corresponding components of each ancient building are obtained, and the specific acquisition process is as follows: Firstly, drones are used to set up corresponding camera equipment, and then corresponding photography is performed on each ancient building to obtain the corresponding primitive images of each ancient building. Secondly, human-computer interactive component description technology is used to copy the corresponding primitive images of each ancient building to obtain the corresponding primitive appearance of each ancient building. Based on the reference extraction of each historical component stored in the database, the feature points and topological structure information of the corresponding component are extracted from the corresponding primitive appearance of each ancient building. The features corresponding to the components in the corresponding primitive appearance of each ancient building are compared with the reference extraction feature set corresponding to each historical component. If a feature corresponding to the component in the corresponding primitive appearance of a certain ancient building is in the reference extraction feature set corresponding to a certain historical component, the feature is used as the extracted feature point, and the feature points of the corresponding components of each ancient building are obtained by comparison and analogy, wherein the feature points include geometric feature points, material features and decorative features, and based on the extracted feature points, the topological structure connection diagram between the corresponding components of each ancient building is obtained, and the topological structure information of the corresponding components of each ancient building is extracted, wherein the topological structure information includes connection relationship, hierarchical structure and spatial relationship, so as to obtain the feature points and topological structure information of the corresponding components of each ancient building.

3. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The specific establishment process of the corresponding directory comparison table is as follows: the feature points of the corresponding components of each ancient building in the description library file are compared with the feature points of each historical component corresponding to each attribute stored in the database. If the feature point of the corresponding component of a certain ancient building in the description library file is the same as the feature point of a certain historical component corresponding to a certain attribute stored in the database, then the attribute is used as the feature point attribute of the feature point of the corresponding component of the ancient building, and so on, the feature point attributes of the feature points of the corresponding components of each ancient building are obtained. Similarly, the topological relationship of the topological structure of the corresponding components of each ancient building can be obtained, and a directory comparison table of the corresponding components of each ancient building is established based on the feature point attributes and topological relationships, and the attributes and relationships of the corresponding components of each ancient building are displayed one by one in the form of a table, and the attributes and relationships of the corresponding components of each ancient building are prefixed with identifiers, so as to establish the corresponding directory comparison table.

4. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The analysis obtains the glyph data of the corresponding components of each ancient building, and the specific analysis process is as follows: The comprehensive attributes of the components corresponding to each ancient building are obtained from the catalog comparison table, and the appearance attributes in the comprehensive attributes corresponding to the components are removed to extract the basic attributes corresponding to the components. The corresponding basic models are obtained by integrating the model building software, and the corresponding constituent lines in the basic model are regarded as primitives to obtain the primitive vectors corresponding to each line component, and the common features of the primitive vectors corresponding to each line component are extracted, wherein the common features include the starting point, end point, line type and width. Based on the above, the glyph data of the components corresponding to each ancient building are obtained through analysis.

5. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The skeleton information of the line components is found from the glyph data of the corresponding components of each ancient building. The specific search process is as follows: Through the corresponding skeleton keywords in the catalog comparison table, the skeleton keywords are input to search the catalog in the component library. Since the ancient building data is collected in the order of component depiction during information collection, and the lines are marked, the data of each component includes the type of each line, the number of lines and the topological relationship of the skeleton information. By performing corresponding reverse processing on the ancient building data, when the data is automatically processed, the corresponding keyword input and matching search are continued to find the skeleton information corresponding to the line component, wherein the skeleton information includes geometric curvature, longitudinal and transverse amount of node connection, material thickness and position, and the skeleton information obtained is subjected to comprehensive data normalization processing to obtain the comprehensive skeleton data value of each line component.

6. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The extraction obtains the corresponding ancient building skeleton, and the specific extraction process is as follows: The comprehensive skeleton data value of each line component is compared with the reference skeleton data value corresponding to each ancient building skeleton stored in the database. If the comprehensive skeleton data value of a line component is the same as the reference skeleton data value corresponding to a certain ancient building skeleton stored in the database, then the ancient building skeleton is taken as the ancient building skeleton corresponding to the line component. By analogy, the corresponding ancient building skeleton is extracted.

7. The ancient building digitization method and system based on feature description as claimed in claim 1, characterized in that: The feature point information of the line skeleton corresponding to the skeleton of the ancient building is obtained, and the specific acquisition process is as follows: First, edge detection is performed on the ancient building image, and the edge detection algorithm is used to identify the contour edges corresponding to the ancient building, so as to obtain the main structural lines of the corresponding skeleton of the ancient building. Secondly, the corner points in the ancient building image are extracted based on the application of the corner point detection algorithm. Finally, the shape of the ancient building image is skeletonized using the morphological skeleton extraction algorithm, so as to extract the simplified shape and structural features corresponding to the ancient building. The deep learning technology is used to perform automated feature point extraction to obtain the feature point information of the line skeleton corresponding to the skeleton of the ancient building, where the feature point information includes the main structural lines, corner points, simplified shapes and structural features. The feature point information is normalized and data processing is performed on the feature point information to obtain the feature points of the line skeleton corresponding to the skeleton of the ancient building.

8. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The analysis obtains the weight corresponding to each feature point, and the specific analysis process is as follows: The characteristic points of the line skeleton corresponding to the skeleton of the ancient building are compared with the characteristic point types of each weighted type corresponding to the historical ancient buildings stored in the database. If a characteristic point of the line skeleton corresponding to the skeleton of the ancient building is different from the characteristic point type of a certain weighted type corresponding to the historical ancient buildings stored in the database, the characteristic point of the line skeleton corresponding to the skeleton of the ancient building is not weighted. If a characteristic point of the line skeleton corresponding to the skeleton of the ancient building is the same as the characteristic point type of a certain weighted type corresponding to the historical ancient buildings stored in the database, the characteristic point of the line skeleton corresponding to the skeleton of the ancient building is weighted. The weights corresponding to each characteristic point are obtained by comparison and analogy.

9. The ancient building digitization method based on feature description as claimed in claim 1, characterized in that: The components of various shapes are generated, and the specific generation process is as follows: After obtaining the weighted selection corresponding to each feature point and according to the corresponding attributes of each feature point, each feature point is weighted. After the weighted processing corresponding to each feature point is completed, the initial contour is constructed based on the weighted feature point information, and the initial contour is refined by connecting adjacent feature points or using an interpolation method, and the corresponding redundancy is removed. At the same time, the contour is smoothed using an optimization algorithm to generate corresponding line components, and then the generated line components are combined and spliced ​​accordingly to form corresponding component shapes. In this way, analogy is performed to generate components of various shapes.

10. An ancient building digitization system for executing the ancient building digitization method based on feature description according to any one of claims 1 to 9, characterized in that: include: The line component abstract description module is used to copy the appearance of the graphic elements of each ancient building, and then obtain the characteristic points and topological structure information of the corresponding components of each ancient building, and save the characteristic points and topological structure information of the corresponding components of each ancient building in the description library file, establish a corresponding catalog comparison table, and analyze and obtain the glyph data of the corresponding components of each ancient building; The ancient building skeleton extraction module is used to find out the skeleton information of the line components from the glyph data of the corresponding components of each ancient building, and extract the corresponding ancient building skeleton; The line component feature point weighting module is used to obtain the feature point information of the line skeleton corresponding to the ancient building skeleton, and analyze and obtain the weight corresponding to each feature point; The automatic generation module of simple ancient paintings is used to receive the weighted feature point information of each line component and generate components of various shapes.

Images