A controlled vectorization method, device and equipment for converting a cultural relic photo into a cultural relic line drawing

Abstract

This application discloses a controlled vectorization method, apparatus, and device for converting artifact photographs into artifact line drawings, relating to the field of artifact photograph processing technology. The method includes: acquiring artifact photographs and performing structural enhancement processing on the photographs to obtain structurally enhanced images; inputting the structurally enhanced images into a line drawing generation model to generate corresponding line drawing probability images; performing binarization and line thinning processing on the line drawing probability images to obtain single-pixel line drawing images; constructing a line drawing topology model based on the single-pixel line drawing images; performing constraint optimization on the topology model based on preset archaeological drawing specifications; and vectorizing the constrained and optimized topology model while maintaining the connection relationships of the topology model to generate vector line drawings. This solution can effectively reduce the interference of texture noise on line drawing generation, solve the problems of broken lines and structural chaos, and improve the editability and applicability of line drawings.

Description

Technical Field

[0001] This invention relates to the field of cultural relic photograph processing technology, and more specifically to a controlled vectorization method, apparatus, and device for converting cultural relic photographs into cultural relic line drawings. Background Technology

[0002] In the fields of archaeological research and cultural relic preservation, line drawings of cultural relics are used to express the shape, structure, boundary relationships, and decorative features of artifacts, and are an important technical achievement in archaeological reports and academic publications. Traditional line drawings of cultural relics mainly rely on professional draftsmen to manually depict them based on photographs or actual artifacts. This method is highly dependent on the experience of the draftsmen, has a long drawing cycle, and there are significant stylistic differences among different draftsmen, making it difficult to ensure the consistency and standardization of the results.

[0003] With the development of digitization technology for cultural relics, technical solutions for automatically generating line drawings using computer vision methods have gradually emerged. Among the existing technologies, one type of method processes images based on traditional edge detection operators and extracts image edges through gradient changes. However, this type of method is easily affected by texture noise, resulting in broken lines and difficulty in distinguishing between structural edges and texture details, which makes it difficult to meet the needs of expressing the structure of cultural relics.

[0004] Another type of method is based on deep learning edge detection models, which learn image features to obtain relatively continuous edge information. However, the generated results still contain a lot of unstructured texture information and lack effective constraints on the closure of object contours and the topological structure of lines, making it difficult to meet the requirements of standardized drawing.

[0005] In addition, in recent years, there have been technical solutions that use generative models to generate line drawings. These solutions generate line drawing images by learning the mapping relationship between images and line drawings. However, such methods usually output raster images, which are difficult to apply directly to archaeological surveying and publishing scenarios. Furthermore, redundant nodes are easily generated in the subsequent vectorization process, and the line structure relationship is difficult to keep consistent.

[0006] Therefore, there is an urgent need for a technical solution that can ensure the accuracy of structural representation while achieving standardized constraints on line drawings and outputting high-quality vector results. Summary of the Invention

[0007] The purpose of this invention is to provide a controlled vectorization method, apparatus, and device for converting artifact photographs into artifact line drawings. This method involves structural enhancement processing of the artifact photograph, combined with a line drawing generation model incorporating multi-scale structural feature encoding and line continuity constraints to generate a probabilistic line drawing image. Based on this, binarization and thinning processes are performed to obtain a single-pixel line drawing image. Furthermore, a line drawing topology model is constructed, and preset archaeological drawing standards are introduced at the topology level for constraint optimization. Finally, vectorization processing is completed while maintaining the topological relationships, thereby achieving automatic conversion from artifact photographs to standardized vector line drawings. Compared to existing technologies, this solution effectively reduces the interference of texture noise on line drawing generation, solves the problems of broken lines and structural chaos, and makes the generated results more in line with the requirements of archaeological drawing in terms of structural expression and standard consistency, while also improving the editability and applicability of the line drawings.

[0008] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics, the method comprising: Acquire photographs of cultural relics and perform structural enhancement processing on them to obtain structurally enhanced images. The structural enhancement processing includes illumination normalization, texture noise suppression, and structural edge enhancement. The structure-enhanced image is input into the line drawing generation model to generate the corresponding line drawing probability image. The line drawing generation model introduces multi-scale structural feature encoding and line continuity constraints during the training process. Binarization and line thinning are performed on the line drawing probability image to obtain a single-pixel line drawing image; A line drawing topology model is constructed based on a single-pixel line drawing image. The line drawing topology model includes nodes and edges connecting the nodes. The topological model is constrained and optimized based on pre-defined archaeological drawing standards. While maintaining the connection relationships of the topological structure model, the constrained and optimized topological structure model is vectorized to generate vector line drawings.

[0009] In some embodiments, structural enhancement processing is applied to photographs of cultural relics, including: Brightness and contrast were normalized in the photographs of cultural relics. A filtering algorithm is used to suppress texture noise in the image; Strengthen the structural boundaries based on gradient or edge enhancement operators.

[0010] In some embodiments, the training process of the line drawing generation model includes: Construct training sample pairs of cultural relic photographs and corresponding line drawings; Noise is gradually added to the line art image to construct a diffusion process; Train a neural network to predict noise or residuals in order to learn the reverse denoising process of recovering line art from noise; Line continuity constraints are introduced during the training process.

[0011] In some embodiments, the line drawing generation model further includes a structure-first generation mechanism, which includes: The contour and main structural features of the artifact are encoded through multi-scale feature extraction; During the generation process, priority generation control is applied to contour lines and main structure lines.

[0012] In some embodiments, constructing a line drawing topology model based on a single-pixel line drawing image includes: Nodes are detected based on endpoints, intersections, and inflection points in single-pixel line art images; Construct edges between nodes based on pixel connectivity in a single-pixel line drawing image; A graph structure representation is established based on nodes and edges, resulting in a line drawing topology model.

[0013] In some embodiments, constraint optimization of the topological structure model is performed based on preset archaeological drawing specifications, including: Transform pre-defined archaeological drawing standards into structural constraint rules; Rule detection is performed on the topology model based on structural constraint rules; The topology that does not meet the structural constraint rules is repaired to obtain the optimized topology model.

[0014] Secondly, the present invention also provides a controlled vectorization device for converting photographs of cultural relics into line drawings of cultural relics, the device comprising: The structure enhancement module is used to acquire photos of cultural relics and perform structure enhancement processing on the photos to obtain structure-enhanced images. The structure enhancement processing includes illumination normalization, texture noise suppression, and structure edge enhancement. The image generation module is used to input the structure-enhanced image into the line drawing generation model to generate the corresponding line drawing probability image. The line drawing generation model introduces multi-scale structural feature encoding and line continuity constraints during the training process. The image processing module is used to perform binarization and line thinning on the line drawing probability image to obtain a single-pixel line drawing image; The model building module is used to build a line drawing topology model based on a single-pixel line drawing image. The line drawing topology model includes nodes and edges connecting the nodes. The model optimization module is used to perform constraint optimization on the topological structure model based on preset archaeological drawing specifications. The line drawing generation module is used to vectorize the constrained and optimized topological model while maintaining the connection relationships of the topological model, and generate vector line drawings.

[0015] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the controlled vectorization method for converting cultural relic photographs into cultural relic line drawings provided in the first aspect.

[0016] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics provided in the first aspect.

[0017] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics provided in the first aspect.

[0018] The beneficial effects of this invention are as follows: This invention enhances the structure of artifact photographs and generates probabilistic line drawing images by incorporating multi-scale structural feature encoding and line continuity constraints into a line drawing generation model. Based on this, single-pixel line drawing images are obtained through binarization and thinning processes. Furthermore, a line drawing topology model is constructed, and preset archaeological drawing standards are introduced for constraint optimization at the topology level. Finally, vectorization is completed while maintaining the topological relationships, thus achieving automatic conversion from artifact photographs to standardized vector line drawings. Compared to existing technologies, this solution effectively reduces the interference of texture noise on line drawing generation, solves the problems of broken lines and structural chaos, and makes the generated results more in line with the requirements of archaeological drawing in terms of structural expression and standard consistency, while also improving the editability and applicability of the line drawings.

[0019] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a controlled vectorization device for converting photographs of cultural relics into line drawings of cultural relics, according to an embodiment of the present invention. Figure 3 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0021] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0022] It should be noted that references to "an embodiment," "embodiment," "example embodiment," etc., in this specification refer to the described embodiment including specific features, structures, or characteristics; however, not every embodiment must include these specific features, structures, or characteristics. Furthermore, such expressions do not refer to the same embodiment. Moreover, when describing specific features, structures, or characteristics in conjunction with embodiments, whether or not explicitly described, it is indicated that incorporating such features, structures, or characteristics into other embodiments is within the knowledge of those skilled in the art.

[0023] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0024] In some embodiments, such as Figure 1 As shown, Figure 1 This is a flowchart illustrating a controlled vectorization method for converting photographs of cultural relics into line drawings. The specific method includes: S101: Obtain a photograph of the cultural relic and perform structural enhancement processing on the photograph to obtain a structurally enhanced image.

[0025] The structural enhancement process includes illumination normalization, texture noise suppression, and structural edge enhancement.

[0026] Optionally, structural enhancement processing can be performed on the artifact photographs, including: normalizing the brightness and contrast of the artifact photographs; using filtering algorithms to suppress texture noise in the images; and enhancing the structural boundaries based on gradient or edge enhancement operators.

[0027] Specifically, the first step is to acquire photographs of the artifacts. These photographs are digital images containing images of artifacts or relics, captured by cameras, scanning devices, or mobile terminals. They can be color or grayscale images, and the resolution can be set according to the actual application scenario; for example, images with a resolution of 300 dpi or higher are typically used in archaeological report mapping. After acquisition, the artifact photographs undergo structural enhancement processing. This process involves using image processing techniques to enhance the structural boundary information related to the artifact's shape and suppress irrelevant texture information. Specifically, this is achieved first by normalizing the illumination, for example, using histogram equalization or adaptive contrast enhancement algorithms to make the brightness distribution in different areas of the image more uniform. Then, filtering algorithms are used to suppress texture noise, such as bilateral filtering or median filtering to remove high-frequency texture details. Based on this, gradient operators or edge enhancement operators (such as the Sobel operator or Laplacian operator) are used to strengthen the structural boundaries, resulting in a structurally enhanced image. For example, for a photograph of a bronze artifact, the above processing can make the artifact's outline clearer while suppressing surface rust texture. This step makes the structural information in the input image more prominent, providing a stable foundation for subsequent line drawing generation, thereby reducing the interference of texture noise on the line drawing result.

[0028] S102, Input the structure-enhanced image into the line drawing generation model to generate the corresponding line drawing probability image.

[0029] The line drawing generation model incorporates multi-scale structural feature encoding and line continuity constraints during training. The training process includes: constructing training sample pairs between artifact photographs and corresponding line drawings; progressively adding noise to the line drawing images to construct a diffusion process; training a neural network to predict noise or residuals to learn the reverse denoising process of recovering the line drawing from noise; and introducing line continuity constraints during training. The model also includes a structure-priority generation mechanism, which involves: encoding the artifact's outline and main structural features through multi-scale feature extraction; and prioritizing the generation of outline lines and main structural lines during the generation process.

[0030] Specifically, after obtaining the structure-enhanced image, it is input into the line drawing generation model to generate a probabilistic line drawing image. The line drawing generation model is a deep learning model that generates data by progressively adding noise to the data and learning a reverse denoising process. It can be based on the UNet structure or other neural network structures. During training, training sample pairs of artifact photographs and corresponding line drawings are first constructed. The line drawings can originate from manually drawn results or existing standardized drawings. Then, Gaussian noise is progressively added to the line drawing image to form a diffusion process, and the neural network is trained to predict the noise or residuals to learn the reverse denoising process of recovering the line drawing from the noise. Multi-scale structural feature encoding is introduced during this training process. This involves extracting the contour and detailed structural features of the artifact through convolutional layers of different scales. Simultaneously, a line continuity constraint term is added to the loss function to constrain the connectivity of the generated lines and avoid breaks. Furthermore, a structure-priority generation mechanism can be introduced, assigning higher weights to contour lines and main structural lines during the model inference stage, prioritizing their generation. For example, when generating the contour of pottery, the outer contour curve is recovered first, while small decorative patterns are generated later. This step allows us to obtain a line drawing probability image that represents the likelihood of a line's existence in pixel probability form, thus distinguishing structural lines from non-structural textures and improving the problem of unclear structural representation in traditional methods.

[0031] S103 performs binarization and line thinning processing on the line drawing probability image to obtain a single-pixel line drawing image.

[0032] Specifically, after obtaining the line drawing probability image, it undergoes binarization and line thinning processing to obtain a single-pixel line drawing image. The line drawing probability image is a grayscale image where each pixel has a line probability value. Binarization involves dividing pixels into line and non-line categories based on a threshold; for example, an adaptive thresholding method can be used to set pixels with probability values ​​higher than the threshold as foreground lines. Line thinning involves compressing lines of a certain width into a skeleton structure close to a single pixel width, which can be achieved using morphological thinning algorithms. During the thinning process, connectivity constraints are introduced to avoid disrupting the original structure; for example, when deleting pixels, it is determined whether it will cause line breakage. For instance, for wide line areas on the edge of an object, thinning can yield continuous single-pixel curves. Through this step, a single-pixel line drawing image with a clear structure and uniform width is obtained, providing standard input for subsequent topology modeling and thus improving the consistency of structural representation.

[0033] S104, Constructing a line drawing topology model based on a single-pixel line drawing image.

[0034] The line drawing topology model includes nodes and edges connecting the nodes.

[0035] Optionally, a line art topology model is constructed based on a single-pixel line art image, including: detecting endpoints, intersections, and turning points as nodes based on the single-pixel line art image; constructing edges between nodes based on pixel connectivity in the single-pixel line art image; and establishing a graph structure representation based on nodes and edges to obtain the line art topology model.

[0036] Specifically, a topological model of the line art is constructed based on the aforementioned single-pixel line art image. A single-pixel line art image refers to a binary image where the line width is a single pixel. The topological model is a data model that describes the connection relationships between lines in the form of a graph structure, where nodes represent key points and edges represent the connection paths between nodes. In practice, firstly, pixel connectivity analysis is performed on the single-pixel line art image to detect endpoints, intersections, and turning points as nodes. For example, an endpoint is a pixel connected to only one adjacent pixel, and an intersection is a pixel with multiple branches. Then, edges between nodes are constructed based on the connectivity relationships between pixels, abstracting continuous pixel paths into line segments. Finally, a graph structure representation composed of nodes and edges is established. For example, the outline of a closed object can be represented as a closed loop structure formed by connecting several nodes. Through this step, the pixel-level line art is transformed into a structured representation, clearly expressing the connection relationships between lines, thus facilitating subsequent rule constraints and structural adjustments.

[0037] S105, constrain and optimize the topological structure model based on the preset archaeological drawing specifications.

[0038] Optionally, the topology model is constrained and optimized based on preset archaeological drawing specifications, including: converting the preset archaeological drawing specifications into structural constraint rules; performing rule detection on the topology model based on the structural constraint rules; and repairing the topology that does not meet the structural constraint rules to obtain the optimized topology model.

[0039] Specifically, after obtaining the topological model, it is constrained and optimized based on preset archaeological drawing standards. These standards refer to structural expression rules developed in archaeological drawing practice, such as the requirement that artifact outlines be closed curves, that relic boundaries should not contain numerous small closed loops, and that decorative patterns should not cross structural boundaries. To facilitate computation, these standards are transformed into structural constraint rules; for example, "outline closure" is transformed into node connectivity constraints, and "small closed loop restrictions" are transformed into minimum loop length constraints. In practice, the topological model is first checked against these structural constraint rules to identify structures that do not conform to the rules, such as unclosed outlines or abnormal small closed loops. Then, structures that do not meet the constraint rules are repaired, for example, by connecting broken nodes to achieve outline closure or deleting redundant edges to eliminate abnormal closed loops. For instance, when a break in the artifact outline is detected, it can be automatically completed to form a closed curve. This step ensures that the topological structure conforms to archaeological drawing standards, thereby improving the standardization and structural rationality of the line drawing.

[0040] S106. While maintaining the connection relationship of the topology model, the constrained optimization topology model is vectorized to generate vector line drawings.

[0041] Specifically, after constraint optimization, while maintaining the connectivity of the topological model, the optimized topological model is vectorized to generate vector line art. Vectorization refers to the process of converting discrete pixel or graph structure representations into editable geometric curve representations, with the output being either a Bezier curve or a polyline. In practice, line segments are first segmented according to node type, for example, by cutting at nodes; then, Bezier curves or polylines are used to fit each line segment. Bezier curves are suitable for smooth curves, while polylines are suitable for approximately straight-line structures. Curvature constraints are introduced during the fitting process to control the smoothness of the curve, and redundant control points are removed to reduce data complexity. For example, for the outer contour of an artifact, a low-order Bezier curve can be used for fitting to obtain a smooth and editable curve representation. Through this step, vector line art is generated while maintaining the original topological connectivity, allowing the results to be directly used in archaeological surveying, publishing, and digital editing scenarios, thereby improving the editability and application adaptability of the line art.

[0042] This embodiment enhances the structure of artifact photographs, generates probabilistic line drawing images by incorporating multi-scale structural feature encoding and line continuity constraints into a line drawing generation model, and obtains single-pixel line drawing images through binarization and thinning. Based on this, a line drawing topology model is constructed, and archaeological drawing standards are transformed into structural constraint rules to optimize the topology. Finally, vectorization is completed while maintaining the topological connectivity, achieving automatic conversion from artifact photographs to standardized vector line drawings. Compared to existing technologies, this solution effectively reduces the interference of texture noise in line drawing generation, improves issues of broken lines and chaotic structural relationships, and makes the line drawings more consistent with archaeological drawing requirements in terms of structural expression and standardization. Furthermore, the generated vector line drawings have more concise control points and clearer structural relationships, facilitating subsequent editing and application, thereby improving the efficiency and quality of artifact digitization.

[0043] Based on the same inventive concept, this application also provides a controlled vectorization device for converting cultural relic photographs into line drawings, which implements the controlled vectorization method for converting cultural relic photographs into line drawings as described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more controlled vectorization device embodiments for converting cultural relic photographs into line drawings provided below can be found in the limitations of the controlled vectorization method for converting cultural relic photographs into line drawings described above, and will not be repeated here.

[0044] In one embodiment, such as Figure 2 As shown, a controlled vectorization device for converting photographs of cultural relics into line drawings of cultural relics is provided. The device includes: The structure enhancement module 30 is used to acquire photos of cultural relics and perform structure enhancement processing on the photos of cultural relics to obtain structure-enhanced images. The structure enhancement processing includes illumination normalization, texture noise suppression and structure edge enhancement. Image generation module 31 is used to input the structure-enhanced image into the line drawing generation model to generate the corresponding line drawing probability image. The line drawing generation model introduces multi-scale structural feature encoding and line continuity constraints during the training process. Image processing module 32 is used to perform binarization and line thinning processing on the line drawing probability image to obtain a single-pixel line drawing image; Model building module 33 is used to build a line drawing topology model based on a single-pixel line drawing image. The line drawing topology model includes nodes and edges connecting the nodes. Model optimization module 34 is used to perform constraint optimization on the topological structure model based on preset archaeological drawing specifications; The line drawing generation module 35 is used to vectorize the constrained optimized topology model while maintaining the connection relationship of the topology model, and generate vector line drawings.

[0045] This application also provides an electronic device, in some embodiments, referring to... Figure 3 As shown, the electronic device 700 includes an input unit 710, a memory 720, a processor 730, and an output unit 740. The memory 720 stores program instructions that can be executed on the processor 730. The processor 730 can execute the controlled vectorization method and / or technical solution for converting cultural relic photographs into cultural relic line drawings based on the foregoing embodiments by calling the program instructions. The electronic device 700 can be a mobile terminal device such as a mobile phone or a computer.

[0046] Furthermore, embodiments of this application also provide a computer-readable storage medium for storing a computer program that performs a controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics. For example, computer program instructions, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. The program instructions that invoke the methods of this application may be stored in a fixed or removable storage medium, and / or transmitted via data streams in broadcast or other signal carrying media, and / or stored in a storage medium that operates according to the program instructions.

[0047] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0048] The technical features of the above embodiments can be arbitrarily integrated. For the sake of brevity, not all possible integrations of the technical features in the above embodiments are described. However, as long as the integration of these technical features does not contradict each other, they should be considered to be within the scope of this specification.

[0049] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics, characterized in that, The method includes: Acquire photographs of cultural relics and perform structural enhancement processing on the photographs to obtain structurally enhanced images. The structural enhancement processing includes illumination normalization, texture noise suppression, and structural edge enhancement. The structure-enhanced image is input into the line drawing generation model to generate a corresponding line drawing probability image. The line drawing generation model introduces multi-scale structural feature encoding and line continuity constraints during the training process. The line drawing probability image is binarized and the lines are thinned to obtain a single-pixel line drawing image; A line drawing topology model is constructed based on the single-pixel line drawing image. The line drawing topology model includes nodes and edges connecting the nodes. The topological model is constrained and optimized based on preset archaeological drawing standards. While maintaining the connection relationships of the topological structure model, the constrained and optimized topological structure model is vectorized to generate vector line drawings.

2. The controlled vectorization method for converting cultural relic photographs into line drawings as described in claim 1, characterized in that, The structural enhancement process performed on the photographs of the cultural relics includes: Brightness and contrast were normalized in the photographs of cultural relics. A filtering algorithm is used to suppress texture noise in the image; Strengthen the structural boundaries based on gradient or edge enhancement operators.

3. The controlled vectorization method for converting cultural relic photographs into line drawings as described in claim 1, characterized in that, The training process of the line drawing generation model includes: Construct training sample pairs of cultural relic photographs and corresponding line drawings; Noise is gradually added to the line art image to construct a diffusion process; Train a neural network to predict noise or residuals in order to learn the reverse denoising process of recovering line art from noise; Line continuity constraints are introduced during the training process.

4. The controlled vectorization method for converting cultural relic photographs into line drawings as described in claim 1 or 3, characterized in that, The line drawing generation model also includes a structure-first generation mechanism, which includes: The contour and main structural features of the artifact are encoded through multi-scale feature extraction; During the generation process, priority generation control is applied to contour lines and main structure lines.

5. The controlled vectorization method for converting cultural relic photographs into line drawings as described in claim 1, characterized in that, Constructing a line art topology model based on the single-pixel line art image includes: The endpoints, intersections, and turning points of the single-pixel line drawing image are detected as nodes; Construct edges between nodes based on the pixel connectivity relationships in the single-pixel line drawing image; Based on the nodes and edges, a graph structure representation is established to obtain the topological structure model of the line drawing.

6. The controlled vectorization method for converting cultural relic photographs into line drawings as described in claim 1 or 5, characterized in that, The topological model is constrained and optimized based on preset archaeological drawing standards, including: The preset archaeological drawing specifications are converted into structural constraint rules; The topology model is subjected to rule detection based on the structural constraint rules. The topology that does not meet the structural constraint rules is repaired to obtain an optimized topology model.

7. A controlled vectorization device for converting photographs of cultural relics into line drawings of cultural relics, characterized in that, The device includes: The structure enhancement module is used to acquire photos of cultural relics and perform structure enhancement processing on the photos of cultural relics to obtain structure-enhanced images. The structure enhancement processing includes illumination normalization, texture noise suppression, and structure edge enhancement. The image generation module is used to input the structure-enhanced image into the line drawing generation model to generate a corresponding line drawing probability image. The line drawing generation model introduces multi-scale structural feature encoding and line continuity constraints during the training process. The image processing module is used to perform binarization and line thinning processing on the line drawing probability image to obtain a single-pixel line drawing image; The model building module is used to build a line drawing topology model based on the single-pixel line drawing image. The line drawing topology model includes nodes and edges connecting the nodes. The model optimization module is used to perform constraint optimization on the topological structure model based on preset archaeological drawing specifications; The line drawing generation module is used to vectorize the constrained optimized topology model while maintaining the connection relationship of the topology model, and generate vector line drawings.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the controlled vectorization method for converting cultural relic photographs into cultural relic line drawings as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the controlled vectorization method for converting cultural relic photographs into cultural relic line drawings as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the controlled vectorization method for converting photographs of cultural relics into line drawings of cultural relics as described in any one of claims 1 to 6.

Images