Intangible cultural heritage project material suitable area prediction method and device and storage medium

Abstract

The application discloses a kind of non-heritage project material suitable area prediction method, device and storage medium.It relates to non-heritage project protection technical field, mainly in that it can improve the prediction accuracy of non-heritage project material suitable area, to realize the distribution prediction and targeted protection of non-heritage project. Including: in response to the suitable area prediction instruction of non-heritage project material, obtain the production raw material of the material carrier needed to form target non-heritage project, and determine the candidate area image of the candidate suitable area corresponding to the production raw material, and the environmental information under the candidate suitable area that raw material produces influence;Based on environmental information, determine the suitable probability of production raw material under each grid image in candidate area image;Based on suitable probability, select suitable grid image in each grid image, and determine at least one suitable area corresponding to non-heritage project material from suitable grid image.The application is suitable for the scene of predicting non-heritage project material suitable area and protecting non-heritage project.

Description

Technical Field

[0001] This invention relates to the field of intangible cultural heritage protection technology, and in particular to a method, apparatus and storage medium for predicting suitable habitats for intangible cultural heritage materials. Background Technology

[0002] my country's representative intangible cultural heritage items (ICH items) are representatives of China's outstanding traditional culture and the crystallization of human cultural diversity. ICH cannot be separated from the material materials associated with it. Based on this, predicting the suitable habitat of ICH item materials is particularly important for the protection of ICH items.

[0003] Currently, the region where an intangible cultural heritage item is declared is usually identified as the suitable growing area for its corresponding production materials. However, this method of determining the suitable growing area only focuses on the intangible cultural heritage item as a whole, resulting in low accuracy in determining the suitable growing area for the materials of the intangible cultural heritage item. Summary of the Invention

[0004] This invention provides a method, device, and storage medium for predicting the suitable habitat of materials for intangible cultural heritage projects. The main purpose is to improve the accuracy of predicting the suitable habitat of materials for intangible cultural heritage projects, thereby enabling the prediction of potential distribution areas of intangible cultural heritage and promoting the overall protection of intangible cultural heritage.

[0005] According to a first aspect of the present invention, a method for predicting suitable habitats for materials of intangible cultural heritage items is provided, comprising:

[0006] In response to the instruction to predict the suitable habitat of materials for intangible cultural heritage projects, the raw materials for making the material carrier required to form the target intangible cultural heritage project are obtained, and the candidate area image of the candidate suitable habitat corresponding to the raw materials is determined, as well as the environmental information under the candidate suitable habitat that affects the raw materials.

[0007] Based on the environmental information, the suitability probability of the raw materials for production is determined under each raster image in the candidate region image;

[0008] Based on the suitability probability, a suitable raster image is selected from each of the raster images, and at least one suitable area corresponding to the intangible cultural heritage material is determined from each of the suitable raster images.

[0009] Optionally, the raw material used in the production is at least one type;

[0010] The step of selecting a suitable raster image from each raster image based on the suitability probability, and determining at least one suitable area corresponding to the intangible cultural heritage material from each suitable raster image, includes:

[0011] Each of the aforementioned raw materials is taken as a target raw material. Based on the survival probability of the target raw material in each grid image of the candidate region image, a survival grid image with a survival probability greater than a preset threshold is determined in each grid image. Spatial clustering is performed on each of the survival grid images. Based on the clustering results, at least one material survival zone is determined for the target raw material.

[0012] Based on the suitable growing areas of each of the aforementioned raw materials, an overlapping suitable growing area is determined that is common to each of the aforementioned raw materials, and the overlapping suitable growing area is determined as the suitable growing area of ​​the intangible cultural heritage project materials.

[0013] Optionally, the suitable growing area of ​​the overlapping materials is defined as the suitable growing area of ​​materials for intangible cultural heritage projects, including:

[0014] Obtain a preset cross structure element template and determine the corrosion binarized image corresponding to the image of the suitable area of ​​overlapping materials. The preset cross structure element template is constructed with a central pixel as the origin and extending one pixel in both the horizontal and vertical directions of the origin. The pixel values ​​of the central pixel and the extended pixels are both set to 1.

[0015] Take any pixel in the eroded binary image as a pixel to be eroded, and align the center element of the preset cross structure element template with the pixel to be eroded.

[0016] Determine whether the pixel values ​​of the pixels in the eroded binary image covered by the position with a value of 1 in the aligned preset cross structure element template are all 1. If so, the pixel to be eroded is determined as the first foreground image pixel; otherwise, the pixel to be eroded is determined as the first background image pixel. Each first foreground image pixel and each first background image pixel constitute the overlapping material suitable area after erosion.

[0017] Take any pixel in the dilated binarized image corresponding to the suitable area of ​​the overlapping material after corrosion treatment as a pixel to be dilated, and align the center element of the preset cross structure element template with the pixel to be dilated.

[0018] Determine whether there is at least one 1 in the pixel value of the pixel in the dilated binarized image covered by the position with a value of 1 in the pre-aligned preset cross structure element template. If so, the pixel to be dilated is determined as a second foreground image pixel; otherwise, the pixel to be dilated is determined as a second background image pixel. Each second foreground image pixel and each second background image pixel constitute the dilated material suitable area.

[0019] The suitable growing area of ​​the overlapping material after expansion treatment is identified as the suitable growing area of ​​the intangible cultural heritage project material.

[0020] Optionally, determining the suitability probability of the raw material for production under each raster image in the candidate region image based on the environmental information includes:

[0021] The raster environment information f corresponding to each raster image z is determined from the environment information. i (z), and determine the environment information f for each of the grids respectively. i The weighting coefficient λ corresponding to (z) i ;

[0022] Based on each of the grid environment information f i (z) and its corresponding weighting coefficient λ i Determine the fitness probability p(z) corresponding to each of the raster images z. Where n is the number of raster environment information, i is the identifier of each raster environment information, and a is a normalization constant.

[0023] Optionally, after determining at least one suitable habitat for the intangible cultural heritage material from each of the suitable habitat grid images, the method further includes:

[0024] Obtain the predicted environmental information of each suitable area after a preset time, and based on the predicted environmental information, determine the predicted suitable probability corresponding to each grid image in the suitable area image of each suitable area.

[0025] Based on the predicted suitable habitat probability, the predicted suitable habitat area of ​​the intangible cultural heritage project materials after the preset time is determined in each of the suitable habitat areas;

[0026] Based on each suitable habitat and its corresponding predicted suitable habitat, determine the degree of area reduction of each suitable habitat after the preset time.

[0027] Based on the degree of reduction in the area of ​​the current suitable habitat of the intangible cultural heritage material, it is determined whether it is necessary to recommend suitable habitats for the intangible cultural heritage material. If so, a recommended suitable habitat is determined in each suitable habitat, and a recommended path is determined for the intangible cultural heritage material from the current suitable habitat to the recommended suitable habitat. Based on the recommended path, the intangible cultural heritage material is recommended from the current suitable habitat to the recommended suitable habitat. Otherwise, it is prohibited to recommend suitable habitats for the intangible cultural heritage material.

[0028] Optionally, determining a recommended path from the current suitable habitat to the recommended suitable habitat for the intangible cultural heritage project materials includes:

[0029] Multiple candidate recommended paths are determined between the current suitable habitat area and the recommended suitable habitat area. Based on the terrain data of the area where each candidate recommended path is located, the terrain slope and land use type of each candidate recommended path are determined, as well as the slope resistance corresponding to the terrain slope and the land resistance corresponding to the land use type.

[0030] The relative importance ratio of the terrain slope and the land use type to the recommended resistance is determined, and based on the relative importance ratio, a judgment matrix of the terrain slope and the land use type to the recommended resistance is constructed. Based on the judgment matrix, the resistance weight coefficients of the terrain slope and the land use type are determined respectively.

[0031] Based on the resistance weighting coefficient, the slope resistance and the land resistance are weighted and summed to obtain the comprehensive recommendation coefficient corresponding to each candidate recommended path. Based on the comprehensive recommendation coefficient, the recommended path for the intangible cultural heritage project materials is determined in each candidate recommended path.

[0032] Optionally, the raw materials for obtaining the material carrier required to form the target intangible cultural heritage item include:

[0033] Obtain the descriptive text information of the target intangible cultural heritage project, search for supplementary corpus information in a preset corpus that has a similarity greater than a preset similarity threshold with the descriptive text information, and input the descriptive text information and the supplementary corpus information into a large model with added sequence label heads for carrier prediction to obtain the material carrier required to form the target intangible cultural heritage project;

[0034] Obtain material prediction prompts, input the material prediction prompts and the material carrier into a preset material prediction model to perform material prediction, and obtain the raw materials for making the material carrier. The preset material prediction model is pre-constructed based on a sample material carrier dataset with manufacturing material labels.

[0035] According to a second aspect of the present invention, a device for predicting the suitable habitat of materials for intangible cultural heritage projects is provided, comprising:

[0036] The acquisition unit is used to respond to the suitable habitat prediction instruction of the intangible cultural heritage project materials, acquire the raw materials for making the material carrier required to form the target intangible cultural heritage project, and determine the candidate area image of the candidate suitable habitat corresponding to the raw materials, as well as the environmental information under the candidate suitable habitat that affects the raw materials.

[0037] The first determining unit is used to determine the suitability probability of the raw material for production under each grid image in the candidate region image based on the environmental information.

[0038] The second determining unit is used to select a suitable raster image from each of the raster images based on the suitability probability, and to determine at least one suitable area corresponding to the intangible cultural heritage material from each of the suitable raster images.

[0039] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described method for predicting suitable habitats for materials of intangible cultural heritage items.

[0040] According to a fourth aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned method for predicting suitable habitats for intangible cultural heritage materials.

[0041] According to the present invention, a method, apparatus, and storage medium for predicting suitable habitats for intangible cultural heritage (ICH) materials are provided. Compared with the current method of determining the region where an ICH project is inherited as the suitable habitat for its corresponding materials, the present invention determines the raw materials used in the production of the target ICH project, selects candidate suitable habitats, and then, based on the environmental information of the candidate suitable habitats, determines the suitability probability of the raw materials in each grid image of the candidate region. Based on the suitability probability, suitable grid images are selected from each grid image, and at least one suitable habitat corresponding to the ICH material is determined from each suitable grid image. In other words, the present invention automatically determines the suitable habitats for ICH materials by considering environmental information, which improves the accuracy of determining suitable habitats for ICH materials. Through suitable habitats, the effective propagation of ICH materials can be achieved, thereby improving the effective protection of ICH projects. Attached Figure Description

[0042] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0043] Figure 1 This invention provides a flowchart of a method for predicting suitable growing areas for materials used in intangible cultural heritage projects, according to an embodiment of the present invention.

[0044] Figure 2 This invention provides a flowchart of another method for predicting suitable habitats for intangible cultural heritage materials, according to an embodiment of the present invention.

[0045] Figure 3 This invention provides a schematic diagram of the structure of a device for predicting the suitable growing area of ​​materials for intangible cultural heritage projects, according to an embodiment of the present invention.

[0046] Figure 4 This invention provides a schematic diagram of the structure of another device for predicting the suitable growing area of ​​materials for intangible cultural heritage projects, according to an embodiment of the present invention.

[0047] Figure 5 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention is shown. Detailed Implementation

[0048] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0049] Currently, the method of determining the suitable habitat for the materials used in the application of intangible cultural heritage projects ignores the connection between the intangible cultural heritage itself and the environment. It only facilitates the administrative management of intangible cultural heritage projects, resulting in low accuracy in determining the suitable habitat for the materials. At the same time, if the existing habitat of the materials of intangible cultural heritage projects shrinks, it will lead to the endangerment of the materials and thus affect the protection of intangible cultural heritage projects.

[0050] To address the aforementioned problems, embodiments of the present invention provide a method for predicting suitable growing areas for materials used in intangible cultural heritage projects, such as... Figure 1 As shown, the method includes:

[0051] 101. In response to the instruction to predict the suitable habitat of materials for intangible cultural heritage projects, obtain the raw materials for the material carrier required to form the target intangible cultural heritage project, and determine the candidate area image of the candidate suitable habitat corresponding to the raw materials, as well as the environmental information of the candidate suitable habitat that affects the raw materials.

[0052] The target intangible cultural heritage item can be any intangible cultural heritage item. For example, if the target intangible cultural heritage item is "Shalang", its corresponding material carriers are "Qiang flute" and "hemp clothing". The raw material required to make the Qiang flute is "oil bamboo", and the raw material required to make hemp clothing is "ramie". If the target intangible cultural heritage item is "Pingyao lacquerware", its corresponding material carrier is "natural lacquer", and the raw material required to make natural lacquer is "lacquer tree".

[0053] In this embodiment of the invention, in order to predict the suitable habitat for the materials of the target intangible cultural heritage project, it is first necessary to determine the raw materials required to form the target intangible cultural heritage project. Based on this, step 101 specifically includes: obtaining the descriptive text information of the target intangible cultural heritage project; searching for supplementary corpus information in a preset corpus that has a similarity greater than a preset similarity threshold with the descriptive text information; inputting the descriptive text information and the supplementary corpus information into a large model with added sequence label heads for carrier prediction to obtain the material carrier required to form the target intangible cultural heritage project; obtaining material prediction prompt information; inputting the material prediction prompt information and the material carrier into a preset material prediction model for material prediction to obtain the raw materials for making the material carrier; wherein the preset material prediction model is pre-constructed based on a sample material carrier dataset with production material labels.

[0054] The descriptive text information includes the name, introduction, inheritance area, technique description, and species distribution data of the target intangible cultural heritage project; the preset corpus stores descriptive text information of various intangible cultural heritage projects; and the preset similarity threshold is set according to actual needs.

[0055] Specifically, to improve the prediction accuracy of the large model, it is first necessary to train and construct a large model with added sequence label headers. The specific construction method includes: obtaining an initial large model with added sequence label headers and acquiring a sample dataset, which contains descriptive text information of sample intangible cultural heritage items with carrier labels; dividing the sample dataset into training and testing data using a random or predetermined strategy; training the initial large model using the training data; monitoring metrics such as loss value and mAP during training to evaluate model performance; and adjusting training parameters such as learning rate, optimizer, and regularization as needed to optimize training results. During training, a cross-entropy loss function with class weights is used to add sequence label headers to the initial large model, as shown below:

[0056]

[0057] Among them, w c For class weights, y i,c For the real label of the i-th token, p i,cLet N be the predicted probability of the initial large model, N be the total number of tokens, and C be the total number of classes. An optimizer such as AdamW (Adaptive Moment Estimation Weight Decay) is used to optimize the initial model parameters to minimize the loss function, training the initial large model to recognize carrier entities. Then, the trained initial large model is tested using test data to evaluate its performance on unseen data. Metrics such as mAP, precision, and recall on the test set are calculated and recorded. If the performance of the large model does not meet the requirements, it can return to the training phase for more iterations or adjustments until the model performance meets the requirements, thus obtaining a satisfactory large model.

[0058] Furthermore, the descriptive text information is cleaned and standardized, such as removing HTML tags and special symbols from the intangible cultural heritage description text; using regular expressions to extract key paragraphs (such as sentences containing "materials," "making," or "using"); and performing OCR text recognition and sentence segmentation on the descriptive text information. For example, if the target intangible cultural heritage item is "Shalang," the preprocessing procedure for the descriptive text information of Shalang can be as follows:

[0059]

[0060]

[0061] Based on the above preprocessing method, the processed descriptive text information can be obtained. Further, to improve the prediction accuracy of the material carrier, it is necessary to supplement the processed descriptive text information. Based on this, the similarity between the descriptive text information and each corpus in the preset corpus can be calculated separately. Then, corpora with similarity greater than a preset similarity threshold are selected from the preset corpus as supplementary corpora. Finally, the supplementary corpora and the descriptive text information are input together into a large model with added sequence label headers for carrier prediction, obtaining the material carrier required to form the target intangible cultural heritage project. Specifically, to utilize the latest information, improve carrier prediction accuracy, and make the predicted carrier more interpretable and adaptable, this invention adds a RAG (Retrieval Augmented Generation) module to the large model. This module uses information from private or proprietary data sources to assist in text generation, thereby compensating for the limitations of the descriptive text information, helping to solve the illusion problem and improve timeliness. Text segmentation: Files of different formats input from external sources are converted into plain text, and sentences are segmented by periods (“.”) and newlines, making them more suitable for embedding search and adapting to the large model, thereby improving the accuracy of fragment retrieval. Text embedding: The text content is transformed into a multi-dimensional vector (i.e., outputting digital language understood by artificial intelligence) through word embedding. This embodiment of the invention can use the bge-large-en-v1.5 embedding model for text vectorization. Index creation: The original descriptive text information and supplementary corpus information are stored as information pairs for quick and frequent future searches. Data retrieval: Based on the descriptive text information, the most relevant supplementary corpus information is quickly retrieved and integrated into the prompt. Based on the descriptive text information, the same encoding model used for index creation is used to convert the descriptive text information into a vector; the similarity between this vector and various corpus vectors in the preset corpus is calculated, and the top K most relevant corpora are selected as supplementary corpus information for the current descriptive text information based on the similarity level. This embodiment of the invention, by determining the supplementary corpus information for the descriptive text information, can improve the prediction accuracy and comprehensiveness of the material carriers corresponding to the target intangible cultural heritage project.

[0062] Furthermore, after predicting the material carriers required to form the target intangible cultural heritage project, it is also necessary to predict the materials used to make these carriers. Specifically, a pre-defined material prediction model can be used to predict the materials used to make the carriers. Before this, to improve the prediction accuracy of the pre-defined material prediction model, it is necessary to first train and construct the model. Based on this, the method includes: obtaining a pre-defined initial material prediction model and a sample dataset, wherein the sample dataset contains sample material carriers labeled with the raw materials used in the production; dividing the sample dataset into training data and test data using a random or predetermined strategy; training the pre-defined initial material prediction model using the training data; monitoring indicators such as loss value and mAP during training to evaluate model performance; and adjusting training parameters such as learning rate, optimizer, and regularization as needed to optimize training effects. Then, testing the trained pre-defined initial material prediction model using test data to evaluate its performance on unseen data; calculating and recording indicators such as mAP, precision, and recall on the test set. If the performance of the pre-defined initial material prediction model does not meet the requirements, it can return to the training phase for further iterations or adjustments. The process continues until the performance of the preset initial material prediction model meets the requirements, thus obtaining a preset material prediction model that meets the requirements.

[0063] Furthermore, material prediction prompts are determined, which are based on requirements. For example, if the material carrier is a "Qiang flute," the material prediction prompts could be "The material must be a natural raw material, excluding synthetic materials" or "Prioritize materials related to the input text." The material prediction prompts and the material carrier are input into a pre-constructed material prediction model for material prediction to obtain the raw materials for manufacturing the material carrier. For example, if the material carrier is a "Qiang flute," the prediction process for its manufacturing material is as follows: Given the material carrier: {Qiang flute}, infer its manufacturing material based on the following knowledge:

[0064] The material prediction prompts are: materials must be natural raw materials, excluding synthetic materials; prioritize materials related to the input text. Output format: material name (type), such as: oil bamboo (biological). Answer: Reference context: {context}#RAG knowledge base retrieved text block, chain-of-thought fine-tuning: facilitates explicit output of the reasoning process. Input: carrier = Qiang flute, output: reasoning steps: the Qiang flute is a traditional musical instrument of the Qiang ethnic group. According to literature, its production materials must meet the following conditions: - hollow bamboo tube, easy to produce sound → requires a specific bamboo species; - resistant to dry climate, adapted to the environment of the western Sichuan plateau → exclude bamboo from humid areas; "Chinese Bamboo Classification" points out that oil bamboo (Fargesia angustissima) distributed in the Minjiang River basin meets the above characteristics; conclusion: oil bamboo (biological). After predicting the production raw materials, you can check whether the production raw materials have distribution records in intangible cultural heritage areas. At the same time, you can add a manual review interface to check whether the production raw materials are predicted accurately. Furthermore, the species distribution database and the mineral resource distribution database can be used to check whether the raw material is a biological material or a non-biological material. The species distribution database records various biological materials and their corresponding distribution information, while the mineral resource distribution database records various non-biological materials and their corresponding distribution data.

[0065] Furthermore, after determining the raw materials corresponding to the target intangible cultural heritage project, candidate suitable growing areas are identified based on actual needs. These candidate suitable growing areas can be any region. Simultaneously, based on the environmental information of the existing areas containing the raw materials, environmental information affecting the growth of the raw materials is determined. This environmental information may include annual average temperature, monthly average diurnal temperature range, isotherm, standard deviation of seasonal temperature variation, highest temperature of the warmest month, lowest temperature of the coldest month, annual average temperature range, average temperature of the wettest quarter, average temperature of the driest quarter, average temperature of the warmest quarter, average temperature of the coldest quarter, annual average precipitation, precipitation of the wettest month, precipitation of the driest month, coefficient of variation of precipitation, precipitation of the wettest quarter, precipitation of the driest quarter, precipitation of the warmest quarter, precipitation of the coldest quarter, altitude, soil type, soil pH, soil organic carbon content, and average sunshine duration.

[0066] 102. Based on environmental information, determine the suitability probability of raw materials for production in each raster image of the candidate region.

[0067] This process involves determining the candidate region images corresponding to the candidate suitable growing areas, and then dividing these candidate region images into raster images according to actual needs, resulting in each raster image within the candidate region images. For example, when dividing the raster images, the raster size (e.g., 100x100 pixels), starting coordinates (e.g., starting from the top left corner of the image by default), and overlap ratio (e.g., 5% overlap to avoid edge effects) are first determined. Then, the candidate region images are divided into raster images according to the raster size, starting coordinates, and overlap ratio, thus obtaining each raster image.

[0068] In this embodiment of the invention, after determining the environmental information affecting the raw materials under the candidate suitable growing area, it is necessary to determine the suitable growing probability of the raw materials under each grid image based on the environmental information. Therefore, step 102 specifically includes: determining the grid environment information f corresponding to each grid image z from the environmental information. i (z), and determine the environment information f for each of the grids respectively. i The weighting coefficient λ corresponding to (z) i Based on each of the grid environment information f i (z) and its corresponding weighting coefficient λ i Determine the fitness probability p(z) corresponding to each of the raster images z. Where n is the number of raster environment information, i is the identifier of each raster environment information, and a is a normalization constant.

[0069] Specifically, by substituting the raster environment information and its corresponding weight coefficients for each raster image into the above formula, the suitability probability (i.e., the probability of survival) for each raster image can be calculated. This embodiment of the invention, by determining the environmental information that affects the raw materials and combining this information to predict the suitability probability, avoids wasting resources and time analyzing environments that do not affect the raw materials. Furthermore, determining the suitable habitat for the raw materials through environmental information improves the accuracy of this determination.

[0070] 103. Based on the suitability probability, select a suitable raster image from each raster image, and determine at least one suitable area corresponding to the intangible cultural heritage material from each suitable raster image.

[0071] Specifically, a target suitability probability greater than a preset threshold is determined in each raster image, and the raster image corresponding to the target suitability probability is determined as a suitable raster image. Then, based on the suitable raster images corresponding to each raw material, at least one suitable area corresponding to the intangible cultural heritage material is determined. Thus, by predicting the suitable area for each raw material corresponding to the target intangible cultural heritage project, this embodiment of the invention can ensure that the finally determined suitable area can meet the survival needs of each raw material. That is, in the process of predicting the suitable area of ​​intangible cultural heritage material, the collaborative constraints of multiple materials are considered, thereby improving the prediction accuracy of the suitable area of ​​intangible cultural heritage material.

[0072] According to the present invention, a method for predicting suitable habitats for intangible cultural heritage (ICH) materials is provided. Compared with the current method of determining the suitable habitat for the corresponding materials of an ICH project by identifying the raw materials used in its production, this invention determines the raw materials for the target ICH project, selects candidate suitable habitats, and then, based on the environmental information of the candidate suitable habitats, determines the suitability probability of the raw materials in each grid image of the candidate region. Based on the suitability probability, suitable grid images are selected from each grid image, and at least one suitable habitat corresponding to the ICH material is determined from each suitable grid image. In other words, this invention automatically determines the suitable habitats for ICH materials by considering environmental information, which improves the accuracy of determining suitable habitats for ICH materials. Through suitable habitats, the effective propagation of ICH materials can be achieved, thereby improving the effective protection of ICH projects.

[0073] Furthermore, to better illustrate the above data classification process, and as a refinement and extension of the above embodiments, this invention provides another method for predicting the suitable habitat areas of intangible cultural heritage materials, such as... Figure 2 As shown, the method includes:

[0074] 201. In response to the instruction to predict the suitable habitat of materials for intangible cultural heritage projects, obtain the raw materials for the material carriers required to form the target intangible cultural heritage project, and determine the candidate area images of the candidate suitable habitats corresponding to the raw materials, as well as the environmental information that affects the raw materials under the candidate suitable habitats.

[0075] Specifically, environmental information affecting the survival of raw materials is analyzed by examining the environmental information of the area where the raw materials are currently located.

[0076] 202. Based on environmental information, determine the suitability probability of the raw material in each raster image of the candidate region image, wherein the raw material is at least one type.

[0077] Specifically, the environmental information affecting the raw materials in the candidate suitable growing area includes the environmental information under each grid image. Based on the environmental information under each grid image, the suitable growing probability of the raw materials under each grid is determined. Different grid images correspond to different grid areas, thereby determining the suitable growing probability corresponding to each small grid area in the candidate suitable growing area.

[0078] 203. Take any one of the production raw materials as a target production raw material. Based on the survival probability of the target production raw material in each grid image in the candidate region image, determine the survival grid images with a survival probability greater than a preset threshold in each grid image, and perform spatial clustering on each survival grid image. Based on the clustering results, determine at least one material survival area for the target production raw material.

[0079] The preset threshold is set according to actual needs. Specifically, each target intangible cultural heritage project corresponds to at least one raw material. Taking a target raw material as an example, suitable raster images with a survival probability greater than the preset threshold are identified in each raster image. Then, the regional spatial characteristics of the raster area corresponding to each raster image are determined. Regional spatial characteristics include geometric features and land use type. Regional features include geometric features (including the center coordinates of the raster area, area, perimeter, and distance from adjacent raster areas, etc.), and land use type (such as residential, green space, commercial land, etc.). Based on the above regional spatial characteristics, a preset clustering method, such as K-means clustering, is used to divide the spatial regions of each suitable raster image. For example, the raster areas corresponding to the closest suitable raster images are grouped together. At the same time, land use type can also be considered when dividing the regions. Finally, the raster areas with the closest and the same land use type are grouped together. In this way, at least one material survival area can be determined for the target raw material from the candidate survival areas.

[0080] 204. Based on the suitable growing areas of each material corresponding to each production raw material, determine the overlapping material suitable growing areas that are common to each production raw material, and determine the overlapping material suitable growing areas as the suitable growing areas of intangible cultural heritage materials.

[0081] In this embodiment of the invention, step 203 allows for the determination of at least one suitable material habitat for each raw material. Then, overlapping suitable material habitats are determined among the suitable material habitats corresponding to each raw material. For example, if the suitable material habitats for raw material a are regions A, B, C, and D; the suitable material habitats for raw material b are regions E, B, F, and G; and the suitable material habitats for raw material c are regions H, B, I, and G, then region B is an overlapping suitable material habitat corresponding to raw materials a, b, and c. Therefore, region B is ultimately determined as the suitable habitat for the intangible cultural heritage (ICH) material. By comprehensively considering each raw material corresponding to the target ICH project to determine its suitable habitat, the synergistic constraints of the target ICH project's reliance on multiple raw materials are fully considered. This improves the accuracy and comprehensiveness of determining the suitable habitats for ICH materials, ensuring that the determined suitable habitats meet the survival needs of each raw material used in the target ICH project. In another embodiment of the present invention, the raw materials required to form the target intangible cultural heritage item may include biological and non-biological materials. If non-biological materials are included, suitable growing areas for these materials can be determined through a mineral resource distribution database. Then, areas suitable for both biological and non-biological materials are identified as suitable growing areas for the intangible cultural heritage item materials.

[0082] Furthermore, when determining the suitable growing areas of materials for intangible cultural heritage projects, in order to eliminate regional noise and fill holes, the suitable growing areas of overlapping materials can be preprocessed. Based on this, the method includes: obtaining a preset cross-shaped structural element template and determining the eroded binary image corresponding to the image of the suitable growing areas of overlapping materials, wherein the preset cross-shaped structural element template is constructed with a central pixel as the origin, and extending one pixel in both the horizontal and vertical directions of the origin, and the pixel values ​​of the central pixel and the extended pixels are all set to 1; taking any pixel in the eroded binary image as a pixel to be eroded, aligning the central element of the preset cross-shaped structural element template with the pixel to be eroded; determining whether the pixel values ​​of the pixels in the eroded binary image covered by the position with a value of 1 in the aligned preset cross-shaped structural element template are all 1, if so, then determining the pixel to be eroded as the first foreground image pixel, otherwise, the pixel to be eroded is... The first background image pixel is determined, and the overlapping material suitable area after erosion is formed by each first foreground image pixel and each first background image pixel. Any pixel in the dilated binarized image corresponding to the overlapping material suitable area after erosion is taken as a pixel to be dilated. The center element of the preset cross structure element template is aligned with the pixel to be dilated. It is determined whether there is at least one 1 in the pixel value of the pixel in the dilated binarized image covered by the position with a value of 1 in the aligned preset cross structure element template. If so, the pixel to be dilated is determined as the second foreground image pixel; otherwise, the pixel to be dilated is determined as the second background image pixel. The overlapping material suitable area after dilation is formed by each second foreground image pixel and each second background image pixel. The overlapping material suitable area after dilation is determined as the suitable area for intangible cultural heritage materials.

[0083] The preset cross-shaped structural element template is shown below:

[0084]

[0085] Specifically, to determine the binarized image corresponding to the image of the suitable growing area of ​​overlapping materials, a pixel at a corner of the binarized image can be used as the starting pixel. Each pixel in the binarized image is then aligned with the center element in the preset cross structure element template. If the pixel values ​​in the binarized image covered by the position with a value of 1 in the preset cross structure element template are also all 1, then the pixel aligned with the center element is retained as the foreground; otherwise, it is set as the background. In this way, the foreground or background of each pixel in the binarized image can be set. After each pixel is set, the suitable growing area of ​​overlapping materials after erosion can be obtained. Furthermore, after eroding the suitable growing area of ​​the overlapping materials, a corner pixel in the binarized image of the eroded suitable growing area of ​​the overlapping materials is used as the starting pixel. Each pixel in the binarized image is then aligned sequentially with the center element of a preset cross-shaped structural element template. If at least one pixel in the binarized image covered by a position with a value of 1 in the preset cross-shaped structural element template is 1, the pixel aligned with the center element is retained as the foreground; otherwise, it is set as the background. In this way, foreground or background settings can be applied to each pixel in the binarized image. After each pixel is set, the dilated suitable growing area of ​​the overlapping materials is obtained. This embodiment of the invention, by eroding the suitable growing area of ​​the overlapping materials, can eliminate small objects, smooth boundaries, and break the adhesion between objects. Through dilation, it can fill holes, connect broken areas, and expand object boundaries, thereby obtaining a suitable growing area for intangible cultural heritage materials that meets the requirements.

[0086] Furthermore, after determining the suitable habitat areas for intangible cultural heritage materials under the current environmental information, in order to permanently protect the materials, it is also necessary to predict the predicted suitable habitat area after a preset time for each suitable habitat area under the current environmental information. Based on this, the method includes: acquiring the predicted environmental information for each suitable habitat area after the preset time, and based on the predicted environmental information, determining the predicted suitability probability corresponding to each raster image in the suitable habitat area image of each suitable habitat area; based on the predicted suitability probability, determining the predicted suitable habitat area for the intangible cultural heritage materials after the preset time in each suitable habitat area; For each suitable habitat area and its corresponding predicted suitable habitat area, determine the degree of area reduction of each suitable habitat area after the preset time. Based on the degree of area reduction of the current suitable habitat area where the intangible cultural heritage material is located, determine whether it is necessary to recommend suitable habitat areas for the intangible cultural heritage material. If so, determine a recommended suitable habitat area in each suitable habitat area and determine a recommended path from the current suitable habitat area to the recommended suitable habitat area for the intangible cultural heritage material. Based on the recommended path, recommend the intangible cultural heritage material from the current suitable habitat area to the recommended suitable habitat area. Otherwise, it is prohibited to recommend suitable habitat areas for the intangible cultural heritage material.

[0087] The preset time is set according to actual needs. Specifically, a large model or other arbitrary method can be used to determine the predicted environmental information of each suitable area corresponding to the intangible cultural heritage materials after the preset time (this predicted environmental information is environmental information that affects the survival of the intangible cultural heritage materials, such as environmental temperature, humidity, temperature difference, etc.). Taking any suitable area as an example, the suitable area is divided into grid regions, and based on the predicted environmental information of each grid region after the preset time, the predicted suitability probability of the corresponding grid image of each grid region is calculated in the manner described in step 102. Target grid images with a predicted suitability probability greater than a preset threshold are identified in each grid image, and spatial clustering is performed on each target grid image. Based on the spatial clustering results, the predicted suitable area corresponding to the suitable area after the preset time is obtained. Thus, the above method can be used to calculate... Calculate the predicted suitable area corresponding to each suitable area. Based on the area of ​​the suitable area and its corresponding predicted suitable area area, determine the degree of area reduction of the suitable area after a preset time. If the area reduction of the suitable area where the target intangible cultural heritage material is currently located is greater than a preset threshold (the preset threshold is set according to actual needs), then the suitable area with the smallest area reduction needs to be determined as the recommended suitable area. A recommended path is set for the target intangible cultural heritage material from the current suitable area to the recommended suitable area, so that the target intangible cultural heritage material can be migrated from the current suitable area to the recommended suitable area based on the recommended path. If the area reduction of the current suitable area is less than or equal to the preset threshold, it means that the current suitable area is suitable for the long-term stable growth of the intangible cultural heritage material, so it does not need to be migrated. In this embodiment of the invention, the method for setting recommended paths specifically includes: determining multiple candidate recommended paths from the current suitable habitat area to the recommended suitable habitat area; determining the terrain slope and land use type of each candidate recommended path based on the terrain data of the area where each candidate recommended path is located, as well as the slope resistance corresponding to the terrain slope and the land resistance corresponding to the land use type; determining the relative importance ratio of the terrain slope and the land use type to the recommended resistance; constructing a judgment matrix of the terrain slope and the land use type to the recommended resistance based on the relative importance ratio; determining the resistance weight coefficients of the terrain slope and the land use type based on the judgment matrix; weighting and summing the slope resistance and the land resistance based on the resistance weight coefficients to obtain a comprehensive recommendation coefficient corresponding to each candidate recommended path; and determining the recommended path for the intangible cultural heritage project materials in each candidate recommended path based on the comprehensive recommendation coefficient.

[0088] Among them, topographic data can be digital elevation model data; land use types include: industrial land, residential land, mountainous and hilly areas, coastal land, agricultural land, etc.; recommended resistance can be migration resistance.

[0089] Specifically, the terrain slope of each candidate recommended path is calculated using digital elevation model data, and the land use type is determined. Then, the slope resistance is determined based on the terrain slope, and the land resistance is determined based on the land use type. For example, the steeper the slope, the greater the resistance; clay has greater resistance than sand. The relative importance ratio of terrain slope to land use type to the recommended resistance is determined using multiple expert ratings or historical data (e.g., slope:land = 3:1). For instance, 5-10 domain experts conduct independent ratings. Experts judge the degree of influence of terrain slope and land use type on the recommended resistance based on experience. If experts believe that the influence of terrain slope on resistance is significantly greater than that of land use type (scale 5), then the ratio of land use type to terrain slope is 1 / 5. Based on the expert evaluation results, a judgment matrix is ​​constructed, as shown below:

[0090]

[0091] Among them, a 12 a represents the ratio of the relative importance of terrain slope and land use type to the recommended resistance. 21 =1 / a 12 Furthermore, the largest eigenvalue of the judgment matrix and its corresponding eigenvector are calculated, and the eigenvector is normalized to obtain the weight vector, which represents the resistance weight coefficients for terrain slope and land use type. Then, based on these resistance weight coefficients, slope resistance and land resistance are weighted and summed to obtain the comprehensive recommendation coefficient for each candidate recommended path. Finally, the path with the smallest comprehensive recommendation coefficient is selected as the recommended path. This embodiment of the invention determines recommended paths by considering path resistance, which can save time and energy consumption in migrating intangible cultural heritage materials through optimized path planning.

[0092] Furthermore, the raw materials used in the production of different intangible cultural heritage items may have overlapping suitable growing areas. Therefore, protection strategies need to be established for materials from at least two intangible cultural heritage items with overlapping suitable growing areas to achieve balanced protection. Based on this, the method includes: determining the total number of pixels in the suitable growing areas and the number of overlapping pixels in the suitable growing areas of any two intangible cultural heritage items, based on their suitable growing areas; determining the ratio of the number of overlapping pixels to the total number of pixels in the suitable growing areas as the similarity of the suitable growing areas between the two intangible cultural heritage items; and generating protection strategy information for the two intangible cultural heritage items based on the similarity of the suitable growing areas.

[0093] Specifically, if the similarity of suitable habitats exceeds a preset threshold (which is set based on actual needs), then the materials of any two intangible cultural heritage items will undergo balanced protection treatment in the overlapping suitable habitats. Otherwise, the materials of one type of intangible cultural heritage item from any two intangible cultural heritage items will undergo protection treatment in the overlapping suitable habitats. This achieves balanced protection for different intangible cultural heritage items and helps to build a sustainable intangible cultural heritage protection system.

[0094] According to another method for predicting suitable habitats for intangible cultural heritage (ICH) materials provided by this invention, compared with the current method of determining the suitable habitat for the corresponding materials based on the area where ICH is inherited, this invention identifies the raw materials for the target ICH project, selects candidate suitable habitats, and then, based on the environmental information of the candidate suitable habitats, determines the suitability probability of the raw materials in each grid image of the candidate area. Based on the suitability probability, suitable grid images are selected from each grid image, and at least one suitable habitat corresponding to the ICH material is determined from each suitable grid image. In other words, this invention automatically determines the suitable habitats for ICH materials by considering environmental information, which improves the accuracy of determining suitable habitats for ICH materials. Through suitable habitats, the effective propagation of ICH materials can be achieved, thereby improving the effective protection of ICH projects.

[0095] Furthermore, as Figure 1 In specific implementation, embodiments of the present invention provide a device for predicting the suitable habitat of materials for intangible cultural heritage projects, such as... Figure 3 As shown, the device includes: an acquisition unit 31, a first determination unit 32, and a second determination unit 33.

[0096] The acquisition unit 31 can be used to respond to the suitable habitat prediction instruction of intangible cultural heritage project materials, acquire the raw materials for making the material carrier required to form the target intangible cultural heritage project, and determine the candidate area image of the candidate suitable habitat corresponding to the raw materials, as well as the environmental information under the candidate suitable habitat that affects the raw materials.

[0097] The first determining unit 32 can be used to determine the suitability probability of the raw materials for production under each grid image in the candidate region image based on the environmental information.

[0098] The second determining unit 33 can be used to select a suitable raster image in each of the raster images based on the suitability probability, and determine at least one suitable area corresponding to the intangible cultural heritage material from each of the suitable raster images.

[0099] In specific application scenarios, the raw materials used in the production process are at least one type; in order to determine at least one suitable habitat for the materials of the intangible cultural heritage project, such as... Figure 4 As shown, the second determining unit 33 includes a clustering module 331 and a determining module 332.

[0100] The clustering module 331 can be used to take any one of the production raw materials as a target production raw material, and based on the survival probability of the target production raw material under each grid image in the candidate region image, determine the survival grid images with a survival probability greater than a preset threshold in each grid image, and perform spatial clustering on each of the survival grid images. Based on the clustering results, at least one material survival zone is determined for the target production raw material.

[0101] The determining module 332 can be used to determine the overlapping material suitable areas that are common to each of the raw materials based on the material suitable areas corresponding to each of the raw materials, and to determine the overlapping material suitable areas as the suitable areas for the materials of the intangible cultural heritage project.

[0102] In specific application scenarios, to determine the suitable growing area of ​​overlapping materials as the suitable growing area of ​​intangible cultural heritage materials, the determining module 332 can be specifically used to obtain a preset cross-shaped structural element template and determine the eroded binary image corresponding to the image of the suitable growing area of ​​overlapping materials. The preset cross-shaped structural element template is constructed with a central pixel as the origin, extending one pixel horizontally and one pixel vertically from the origin. The pixel values ​​of the central pixel and the extended pixels are all set to 1. Any pixel in the eroded binary image is taken as a pixel to be eroded, and the central element of the preset cross-shaped structural element template is aligned with the pixel to be eroded. It is determined whether the pixel values ​​of the pixels in the eroded binary image covered by the position with a value of 1 in the aligned preset cross-shaped structural element template are all 1. If so, the pixel to be eroded is determined as the first foreground image pixel; otherwise, the pixel to be eroded is determined as the first... The background image pixels are composed of each first foreground image pixel and each first background image pixel, forming the overlapping material suitable area after erosion processing. Any pixel in the dilated binarized image corresponding to the overlapping material suitable area after erosion processing is taken as a pixel to be dilated. The center element of the preset cross structure element template is aligned with the pixel to be dilated. It is determined whether there is at least one 1 in the pixel value of the pixel in the dilated binarized image covered by the position with a value of 1 in the aligned preset cross structure element template. If so, the pixel to be dilated is determined as a second foreground image pixel; otherwise, the pixel to be dilated is determined as a second background image pixel. The overlapping material suitable area after dilation processing is composed of each second foreground image pixel and each second background image pixel. The overlapping material suitable area after dilation processing is determined as the suitable area for intangible cultural heritage materials.

[0103] In specific application scenarios, in order to determine the fitness probability under each raster image, the first determining unit 32 can be specifically used to determine the raster environment information f corresponding to each raster image z in the environment information. i (z), and determine the environment information f for each of the grids respectively. i The weighting coefficient λ corresponding to (z) i Based on each of the grid environment information f i (z) and its corresponding weighting coefficient λ i Determine the fitness probability p(z) corresponding to each of the raster images z. Where n is the number of raster environment information, i is the identifier of each raster environment information, and a is a normalization constant.

[0104] In specific application scenarios, in order to recommend suitable habitats, the device further includes a recommendation unit 34.

[0105] The recommendation unit 34 can be used to acquire the predicted environmental information of each suitable habitat area after a preset time, and based on the predicted environmental information, determine the predicted suitable habitat probability corresponding to each grid image in the suitable habitat area image of each suitable habitat area; based on the predicted suitable habitat probability, determine the predicted suitable habitat area of ​​the intangible cultural heritage material after the preset time in each suitable habitat area; based on each suitable habitat area and its corresponding predicted suitable habitat area, determine the degree of area reduction of each suitable habitat area after the preset time; based on the degree of area reduction of the current suitable habitat area where the intangible cultural heritage material is located, determine whether it is necessary to recommend a suitable habitat area for the intangible cultural heritage material; if so, determine a recommended suitable habitat area in each suitable habitat area, and determine a recommendation path from the current suitable habitat area to the recommended suitable habitat area for the intangible cultural heritage material; based on the recommendation path, recommend the intangible cultural heritage material from the current suitable habitat area to the recommended suitable habitat area; otherwise, prohibit the recommendation of a suitable habitat area for the intangible cultural heritage material.

[0106] In specific application scenarios, to determine recommended paths from the current suitable habitat area to the recommended suitable habitat area, the recommendation unit 34 can specifically be used to determine multiple candidate recommended paths from the current suitable habitat area to the recommended suitable habitat area, and based on the terrain data of the area where each candidate recommended path is located, determine the terrain slope and land use type of each candidate recommended path, as well as the slope resistance corresponding to the terrain slope and the land resistance corresponding to the land use type; determine the relative importance ratio of the terrain slope and the land use type to the recommended resistance, and based on the relative importance ratio, construct a judgment matrix of the terrain slope and the land use type to the recommended resistance, and based on the judgment matrix, determine the resistance weight coefficients of the terrain slope and the land use type respectively; based on the resistance weight coefficients, perform a weighted summation of the slope resistance and the land resistance to obtain a comprehensive recommendation coefficient corresponding to each candidate recommended path, and based on the comprehensive recommendation coefficient, determine the recommended path for the intangible cultural heritage project materials in each candidate recommended path.

[0107] In specific application scenarios, in order to obtain the raw materials for the material carrier required to form the target intangible cultural heritage project, the acquisition unit 31 includes a carrier prediction module 311 and a material prediction module 312.

[0108] The carrier prediction module 311 can be used to obtain the descriptive text information of the target intangible cultural heritage project, search for supplementary corpus information in a preset corpus that has a similarity greater than a preset similarity threshold with the descriptive text information, and input the descriptive text information and the supplementary corpus information into a large model with added sequence label heads for carrier prediction, so as to obtain the material carrier required to form the target intangible cultural heritage project.

[0109] The material prediction module 312 can be used to obtain material prediction prompt information, input the material prediction prompt information and the material carrier into a preset material prediction model to perform material prediction, and obtain the raw materials for making the material carrier. The preset material prediction model is pre-constructed based on a sample material carrier dataset with manufacturing material labels.

[0110] It should be noted that other corresponding descriptions of the functional modules involved in the device for predicting suitable growing areas of intangible cultural heritage materials provided in this embodiment of the invention can be found in the following references. Figure 1 The corresponding description of the method shown will not be repeated here.

[0111] Based on the above, Figure 1Accordingly, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps: in response to a suitable habitat prediction instruction for intangible cultural heritage materials, acquiring the raw materials for forming the material carrier required for the target intangible cultural heritage project, and determining candidate region images of candidate suitable habitats corresponding to the raw materials, as well as environmental information affecting the raw materials under the candidate suitable habitats; based on the environmental information, determining the suitability probability of the raw materials under each grid image in the candidate region images; based on the suitability probability, selecting suitable grid images in each grid image, and determining at least one suitable habitat corresponding to the intangible cultural heritage material from each suitable grid image.

[0112] Based on the above, Figure 1 The method shown and as Figure 3 The embodiment of the device shown in the invention also provides a physical structure diagram of a computer device, such as... Figure 5 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor. Both the memory 42 and the processor 41 are mounted on a bus 43. When the processor 41 executes the program, it performs the following steps: in response to a prediction instruction for the suitable growing area of ​​intangible cultural heritage materials, it acquires the raw materials for forming the material carrier required for the target intangible cultural heritage project, and determines candidate region images of the candidate suitable growing areas corresponding to the raw materials, as well as environmental information affecting the raw materials under the candidate suitable growing areas; based on the environmental information, it determines the probability of the raw materials' suitability under each grid image in the candidate region images; based on the suitability probability, it selects a suitable grid image in each grid image, and determines at least one suitable growing area corresponding to the intangible cultural heritage material from each suitable grid image.

[0113] Through the technical solution of this invention, the present invention determines the raw materials for the production of target intangible cultural heritage projects and selects candidate suitable growing areas. Then, based on the environmental information of the candidate suitable growing areas, it determines the probability of the raw materials' suitability for each grid image in the candidate area image. According to the suitability probability, suitable grid images are selected from each grid image, and at least one suitable growing area corresponding to the intangible cultural heritage project material is determined from each suitable grid image. In other words, this invention automatically determines the suitable growing areas of intangible cultural heritage project materials by considering environmental information, which can improve the accuracy of determining the suitable growing areas of intangible cultural heritage project materials. Through suitable growing areas, the effective propagation of intangible cultural heritage project materials can be achieved, thereby improving the effective protection of intangible cultural heritage projects.

[0114] It is obvious to those skilled in the art that the modules or steps of the present invention 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. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting a suitable area for a non-heritage project material, characterized by, The method comprises the following steps: In response to the non-heritage project material suitable area prediction instruction, obtain the production raw materials of the material carrier required for forming the target non-heritage project, determine the candidate area image of the candidate suitable area corresponding to the production raw materials, and the environmental information of the candidate suitable area affecting the production raw materials; Based on the environmental information, determine the suitability probability of the production raw materials under each grid image in the candidate area image; Based on the suitability probability, select suitable grid images in each grid image, and determine at least one suitable area of non-heritage project material corresponding to each suitable grid image; Wherein, the production raw materials are at least one; based on the suitability probability, select suitable grid images in each grid image, and determine at least one suitable area of non-heritage project material corresponding to each suitable grid image, comprising: Take any one of the production raw materials as a target production raw material, and determine the suitable grid image with a suitability probability greater than a preset threshold in each grid image based on the suitability probability of the target production raw material in each grid image in the candidate area image. Spatial clustering is performed on each suitable grid image, and at least one material suitable area is determined for the target production raw material according to the clustering result. Based on each material suitable area corresponding to each production raw material, determine an overlapping material suitable area commonly corresponding to each production raw material, and determine the overlapping material suitable area as the suitable area of non-heritage project material; The method comprises the following steps: determining each of the grid images in the environment information corresponding grid environment information , and determining each of the grid environment information respectively corresponding weight coefficient ; determining each of the grid images based on each of the grid environment information and the corresponding weight coefficient thereof corresponding fitness probability , wherein, is the number of grid environment information, is the identification of each of the grid environment information, is a normalization constant.​ 2. The method of claim 1, wherein, Determine the overlapping material suitable area as the suitable area of non-heritage project material, comprising: Obtain a preset cross-shaped structural element template, and determine an eroded binary image corresponding to the overlapping material suitable area image, wherein the preset cross-shaped structural element template is composed of a center pixel as the origin, and each pixel value of the center pixel and the extended pixel is set to 1; Align the center element of the preset cross-shaped structural element template with the to-be-eroded pixel point; Determine whether the pixel values of the pixel points in the eroded binary image covered by the positions with a value of 1 in the aligned preset cross-shaped structural element template are all 1, if yes, determine the to-be-eroded pixel point as a first foreground image pixel point, otherwise, determine the to-be-eroded pixel point as a first background image pixel point, and the eroded overlapping material suitable area is composed of each first foreground image pixel point and each first background image pixel point; Align the center element of the preset cross-shaped structural element template with the to-be-inflated pixel point; determining whether there is at least one 1 in the pixel value of the pixel point in the dilated binary image covered by the position of the value 1 in the preset cross structure element template after alignment, if yes, determining the pixel point to be dilated as a second foreground image pixel point, otherwise, determining the pixel point to be dilated as a second background image pixel point, and each of the second foreground image pixel points and each of the second background image pixel points constitute the suitable area of the overlapping material after dilated processing; determining the suitable area of the overlapping material after dilated processing as the suitable area of the non-heritage project material.

3. The method of claim 1, wherein, After determining at least one suitable area corresponding to the non-heritage project material from each of the suitable grid images, the method further comprises: obtaining prediction environment information of each of the suitable areas after a preset time, and respectively determining a prediction suitable probability corresponding to each grid image in a suitable area image of each of the suitable areas based on the prediction environment information; respectively determining a prediction suitable area of the non-heritage project material after the preset time in each of the suitable areas based on the prediction suitable probability; based on each of the suitable areas and the corresponding prediction suitable area, determining the area reduction degree of each of the suitable areas after the preset time; based on the area reduction degree of the current suitable area where the non-heritage project material is located, determining whether the non-heritage project material needs to be recommended to a suitable area, if yes, determining a recommended suitable area in each of the suitable areas, and determining a recommended path for the non-heritage project material from the current suitable area to the recommended suitable area, recommending the non-heritage project material from the current suitable area to the recommended suitable area based on the recommended path, otherwise, prohibiting the non-heritage project material from being recommended to a suitable area.

4. The method of claim 3, wherein, The method for determining a recommended path for the non-heritage project material from the current suitable area to the recommended suitable area comprises: determining a plurality of candidate recommended paths between the current suitable area and the recommended suitable area, and respectively determining a terrain slope and a land use type of each of the candidate recommended paths based on the terrain data of the area where each of the candidate recommended paths is located, as well as a slope resistance corresponding to the terrain slope and a land resistance corresponding to the land use type; determining the relative importance ratio of the terrain slope and the land use type to the recommended resistance, and based on the relative importance ratio, constructing a judgment matrix of the terrain slope and the land use type to the recommended resistance, and respectively determining the resistance weight coefficients of the terrain slope and the land use type based on the judgment matrix; based on the resistance weight coefficients, weighting and summing the slope resistance and the land resistance to obtain a comprehensive recommended coefficient corresponding to each of the candidate recommended paths, and determining the recommended path of the non-heritage project material in each of the candidate recommended paths based on the comprehensive recommended coefficient.

5. The method of claim 1, wherein, The method for obtaining the production raw materials of the material carrier required to form the target non-heritage project comprises: Obtaining description text information of the target non-heritage project, searching for supplementary corpus information with a similarity greater than a preset similarity threshold in a preset corpus, and inputting the description text information and the supplementary corpus information into a large model with an added sequence labeling head to perform carrier prediction to obtain a material carrier required to form the target non-heritage project; Obtaining material prediction prompt information, inputting the material prediction prompt information and the material carrier into a preset material prediction model to perform material prediction, and obtaining a raw material of the material carrier, wherein the preset material prediction model is constructed in advance based on a sample material carrier data set with a manufacturing material label.

6. A device for predicting a suitable area for a non-heritage project material, characterized by, Comprise: The acquisition unit is configured to, in response to a non-heritage project material habitat prediction instruction, acquire a raw material of a material carrier required to form a target non-heritage project, and determine a candidate area image of a candidate habitat corresponding to the raw material and environmental information affecting the raw material under the candidate habitat; The first determining unit is configured to determine the survival probability of the production raw material under each grid image in the candidate region image based on the environment information. The determination of the survival probability of the production raw material under each grid image in the candidate region image based on the environment information comprises: determining each grid image in the environment information corresponding grid environment information , and determining each grid environment information respectively corresponding weight coefficient ; determining each grid image corresponding survival probability based on each grid environment information and the corresponding weight coefficient , wherein, the number of grid environment information is n, the identification of each grid environment information is i, and the normalization constant is c. The second determination unit is configured to select a suitable grid image in each grid image based on the habitat probability, and determine at least one suitable habitat corresponding to the non-heritage project material from each suitable grid image, wherein the raw material is at least one; the suitable grid image in each grid image is selected based on the habitat probability, and at least one suitable habitat corresponding to the non-heritage project material is determined from each suitable grid image, comprising: taking any of the raw materials as a target raw material, determining a suitable grid image with a habitat probability greater than a preset threshold in each grid image based on the habitat probability of the target raw material in each grid image of the candidate area image, and performing spatial clustering on each suitable grid image, and determining at least one material habitat for the target raw material according to the clustering result; based on each material habitat corresponding to each raw material, determine an overlapping material habitat corresponding to each raw material, and determine the overlapping material habitat as the habitat of the non-heritage project material.

7. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 5.

8. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 5.

Images