Method and system for classifying types of ancient porcelain bowls based on artificial intelligence big data algorithm

Abstract

The invention discloses a method and system for classifying ancient porcelain bowl shapes based on an artificial intelligence big data algorithm, and belongs to the technical field of science and technology archaeological. The method comprises the following steps: firstly, obtaining three-dimensional structure data of an ancient pottery bowl through a three-dimensional scanning technology, and carrying out modeling correction on an incomplete utensil; secondly, performing standardization processing on the three-dimensional data, and generating a virtual sample to expand a training set; thirdly, utilizing a hierarchical clustering algorithm and principal component analysis to extract shape features of kiln mouths in different times, and training a supervised learning classifier; and finally, inputting the three-dimensional data of the to-be-identified ancient porcelain bowl into the classifier, and automatically outputting the identification results of the age, the kiln mouth and the style of the to-be-identified ancient porcelain bowl. According to the method, the artificial intelligence technology is used for replacing traditional manual identification, the classification efficiency and accuracy are remarkably improved, subjective errors are reduced, and the method is particularly suitable for large-scale archaeological exploration and automatic arrangement of museum collections. In addition, the system supports continuous algorithm optimization and database updating, the classification precision is further improved, and efficient and reliable technical support is provided for ancient ceramic research.

Description

Technical Field

[0001] This invention relates to the field of scientific archaeology, specifically to a method and system for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms. Background Technology

[0002] Bowls from different eras in China exhibit diverse shapes, reflecting the craftsmanship and aesthetic tastes of their time. Neolithic bowls were mostly handmade, with simple and rugged shapes, primarily flat or round bottoms, such as the painted pottery bowls of the Yangshao culture. Bowls from the Shang and Zhou dynasties gradually became more regular in shape, with thicker walls, and often featured high-footed bowls (like the dou), decorated primarily with geometric patterns. During the Qin and Han dynasties, bowl shapes diversified, with glazed pottery bowls appearing; these had thinner walls, flared rims, and increased practicality. During the Wei, Jin, and Northern and Southern dynasties, celadon bowls became mainstream, with simple and elegant shapes, vibrant celadon glazes, and common lotus petal decorations. Bowls from the Sui and Tang dynasties were full and imposing, with white porcelain bowls having a fine body and celadon bowls having a lustrous glaze; the rims were often petal-shaped or sunflower-shaped. Bowls from the Song and Yuan dynasties were renowned for their elegance and refinement, such as the lotus-shaped bowls of Ru ware, the carved bowls of Ding ware, and the blue-and-white bowls of the Yuan dynasty, all with regular shapes and pure glazes. During the Ming and Qing dynasties, bowls became more diverse in shape. Blue and white bowls, wucai (five-color) bowls, and famille rose bowls were decorated with intricate designs, and the rims and feet were finely finished, showcasing an extremely high level of craftsmanship.

[0003] As archaeological research deepens, a large number of ceramic samples from different periods have been unearthed, accumulating more and more information, and the relationships between various pieces of information have become increasingly complex. Archaeologists and artifact appraisal experts mainly rely on their experience to observe unearthed ceramic samples with the naked eye. Moreover, the identification of ancient ceramic shapes is a comprehensive task, requiring appraisers to possess profound historical knowledge, rich practical experience, and keen observation skills. Therefore, the accuracy of current unearthed ceramic identification largely depends on the knowledge and experience of the appraisers, which can easily lead to errors in identification. Summary of the Invention

[0004] To address the problems existing in current technologies, this invention provides a method and system for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms. This significantly improves identification efficiency and reduces subjective errors, making it particularly suitable for large-scale archaeological excavations or the organization of museum collections. Furthermore, this method can be further improved in classification accuracy through continuous database updates and algorithm optimization, providing strong technical support for ancient ceramics research.

[0005] This invention is achieved through the following technical solution: A method for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms includes the following steps: Step 1: Obtain the three-dimensional structural data of the ancient pottery bowl; Step 2: Standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Generate virtual sample data based on the standardized data. Construct a training sample set based on the standardized data and virtual sample data. Step 3: Cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape characteristics of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape characteristics. Step 4: Obtain the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape characteristics of the ancient porcelain bowl based on the three-dimensional structural data, input the shape characteristics into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

[0006] Preferably, obtaining the three-dimensional structural data of the ancient pottery bowl includes: Three-dimensional scanning was performed on ancient porcelain bowls with unclear origins and dating. A three-dimensional model was constructed based on the three-dimensional scanning data. The three-dimensional model was then sliced ​​to obtain the three-dimensional structural data of the ancient pottery bowl.

[0007] Preferably, when the ancient porcelain bowl is incomplete due to a lack of provenance and dating, the pottery bowl is modeled and corrected, and three-dimensional structural data is obtained based on the corrected ancient pottery bowl model.

[0008] Preferably, the standardization process for the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl includes: The three-dimensional structural data of the ancient pottery bowl were subjected to noise reduction, outlier removal, and normalization to obtain standardized data of the ancient pottery bowl.

[0009] Preferably, constructing the training sample set based on standardized data and virtual sample data includes: The standardized data is rotated, scaled, translated, or / and distorted to generate virtual sample data. The standardized data and virtual sample data are then integrated to obtain the training sample set.

[0010] Preferably, the training sample set is clustered to obtain training samples from different eras and kilns, and the vessel shape characteristics of the training samples from different eras and kilns are obtained, including: Hierarchical clustering algorithm is used to classify the training sample set to obtain training samples from different eras and kilns. Principal component analysis is then used as a dimensionality reduction technique to reduce the dimensionality of the clustered training samples and obtain the shape features of the dimensionality-reduced training samples.

[0011] Preferably, the step of training the ancient porcelain bowl classifier based on vessel shape features includes: A supervised learning algorithm was used as the classifier. The shape features and corresponding classification labels were input into the classifier for training. The performance of the classifier was evaluated using cross-validation. The structure and parameters of the classifier were adjusted to optimize the performance. The ancient porcelain bowl classifier was obtained through iterative training.

[0012] Preferably, the identification results of the ancient porcelain bowl include the era, kiln, and style of the ancient porcelain bowl.

[0013] A system for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms, including The acquisition module is used to obtain the three-dimensional structural data of the ancient pottery bowl; The preprocessing module is used to standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Virtual sample data is generated based on the standardized data, and a training sample set is constructed based on the standardized data and the virtual sample data. The training module is used to cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape features of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape features. The identification module is used to acquire the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape features of the ancient porcelain bowl based on the three-dimensional structural data, input the shape features into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

[0014] An electronic device, characterized in that it comprises: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms.

[0015] Compared with the prior art, the present invention has the following beneficial technical effects: This application provides a method for classifying ancient porcelain bowls based on artificial intelligence and big data algorithms. First, 3D scanning technology is used to acquire the 3D structural data of the ancient pottery bowls, ensuring the comprehensiveness and accuracy of the data. In particular, the modeling and correction function for incomplete artifacts solves the problem of inaccurate analysis due to incomplete artifacts in traditional identification methods. Second, standardized processing steps (including noise reduction, outlier removal, and normalization) effectively eliminate noise and bias during data acquisition, laying a high-quality data foundation for subsequent analysis. Furthermore, generating virtual sample data (such as through rotation, scaling, and translation operations) not only expands the diversity of training samples but also enhances the model's generalization ability, enabling it to adapt to ancient porcelain bowls in different preservation states and forms. In the feature extraction and classifier training stages, the combination of hierarchical clustering algorithms and principal component analysis (PCA) achieves dimensionality reduction of high-dimensional data and extraction of key features, thereby simplifying model complexity and improving classification efficiency. The introduction of supervised learning algorithms (such as neural networks) further optimizes the classifier's performance. Through cross-validation and parameter tuning, high accuracy in classifying by era, kiln, and style is ensured. Ultimately, this method can quickly and automatically output identification results, including confidence scores, providing objective and quantifiable scientific evidence for archaeological research and artifact authentication. Compared with traditional identification methods that rely on expert experience, this invention not only significantly reduces labor costs and subjective errors, but also continuously improves classification accuracy through continuous database updates and algorithm optimization. It is particularly suitable for large-scale archaeological excavations and the systematic organization of museum collections, demonstrating significant technological advancement and practical value.

[0016] This application also proposes a system for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms, an electronic device, and a computer storage medium, which possess all the advantages of the aforementioned method for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a model diagram of the sample after calibration according to the present invention; Figure 2 This is a sample slice and information extraction diagram after calibration according to the present invention; Figure 3 This is a diagram illustrating the dimensionality reduction analysis of the extracted information in this invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] See Figure 1-3 A method for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms includes the following steps: Step 1: Obtain the three-dimensional structural data of the ancient pottery bowl.

[0022] Using 3D scanning technology or high-precision image acquisition equipment, comprehensive and multi-dimensional data collection and analysis are conducted on ancient porcelain bowls whose origins and dating have been lost. The surface morphology of the ancient porcelain bowls is accurately captured, generating high-resolution digital models. Comprehensive data on the shape of the unearthed ancient porcelain bowls is extracted, including key parameters such as mouth diameter, bottom diameter, height, curvature, and thickness.

[0023] For ancient pottery bowls with fragments unearthed, a model was created and corrected. The corrected shape was then sliced, and the radius of each segment was precisely measured and extracted.

[0024] S1.1 Select a high-precision 3D scanner. Adjust the scanner's resolution and scanning range according to the size and complexity of the ancient porcelain bowl. Before scanning, calibrate the scanner to ensure the accuracy of the scanned data.

[0025] S1.2 Place the antique porcelain bowl on the scanner's worktable and start the scanner for a comprehensive, multi-dimensional scan. During the scanning process, adjust the position or angle of the antique porcelain bowl to ensure that all its surfaces are scanned.

[0026] For large or complex antique porcelain bowls, multiple scans may be necessary to obtain complete data. After scanning, professional 3D data processing software is used to stitch and merge the data from multiple scans to generate a complete high-resolution 3D model.

[0027] Choose a high-resolution digital camera or scanner as your image acquisition device.

[0028] Based on the characteristics of the antique porcelain bowl, adjust camera parameters such as focal length, aperture, and exposure time to obtain clear images. Shoot the bowl from multiple angles, including the front, side, top, and bottom, to ensure all surface information is captured. For complex shapes, more detailed images may be required.

[0029] S1.3. Use image processing software to preprocess the captured images, such as denoising and contrast enhancement. Then, use a 3D reconstruction algorithm to convert the processed images into a 3D model.

[0030] Whether through 3D scanning or high-precision image acquisition, the final result will be point cloud data of the ancient porcelain bowl. The point cloud data contains a large amount of 3D coordinate information of the surface of the ancient porcelain bowl.

[0031] Professional 3D modeling software is used to reconstruct the surface of point cloud data, generating a 3D surface model of the ancient porcelain bowl. During the reconstruction process, optimization operations such as surface smoothing and repairing damaged areas may be necessary to improve the model's accuracy and aesthetics. The optimized 3D surface model is then a high-resolution digital model. This model can be exported to common 3D file formats (such as STL and OBJ) for subsequent data analysis and processing.

[0032] S1.4. Use the measurement tools in the 3D modeling software to measure the key parameters in the digital model, including the diameter, bottom diameter, height, curvature, and thickness.

[0033] Record the measured data, and then organize and analyze it. Data reports or charts can be generated as needed to visually display the shape and characteristics of the ancient porcelain bowl.

[0034] S1.5 For ancient pottery bowls unearthed in fragments, the first step is to piece together and model the fragments. Professional 3D modeling software can be used, or the position and angle of the fragments can be manually adjusted to achieve accurate assembly. After assembly, the model is calibrated to ensure its accuracy and integrity. Then, the calibrated model is sliced ​​to obtain detailed information such as the radius of each segment. This information is equally important for subsequent classification and identification.

[0035] Step 2: Standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Generate virtual sample data based on the standardized data. Construct a training sample set based on the standardized data and the virtual sample data.

[0036] Remove noise and outliers (such as scanning errors or missing data) from the collected data. Unify data from different sources to the same units and ranges to facilitate subsequent analysis. Generate more training samples through operations such as rotation and scaling to improve the model's generalization ability.

[0037] S2.1. Inspect the acquired 3D structural data of the ancient pottery bowl, identifying and removing noise and outliers. Noise may originate from errors during the scanning process, while outliers may be due to incomplete data acquisition or equipment malfunction. These defective data can be identified and removed through statistical methods, filtering algorithms, or manual inspection.

[0038] Data must be standardized to the same scale and range: Since data from different sources may have different scales and ranges, they need to be standardized to the same scale and range for ease of subsequent analysis. This can be achieved through methods such as linear transformation, normalization, or standardization, ensuring that all data are compared and analyzed on the same scale.

[0039] S2.2, Generate virtual sample data: Rotation is used to increase the diversity of training samples and improve the generalization ability of the model. It can be performed on standardized data. By rotating the three-dimensional model of the ancient pottery bowl, data samples at different angles can be generated, thereby simulating various situations that may be encountered in the actual identification process.

[0040] By scaling the data and changing the size of the 3D model of the ancient pottery bowl, data samples of different sizes can be generated. This helps the model better adapt to ancient pottery bowls of various sizes that may be encountered in actual identification.

[0041] Optionally, other transformation operations, such as translation and distortion, can be performed to further increase the diversity of training samples.

[0042] S2.3 Construct a training sample set.

[0043] The standardized data and the generated virtual sample data are integrated to form a complete training sample set. This sample set should contain a sufficient number of samples to cover ancient pottery bowls from various eras, kilns, and styles.

[0044] Data annotation is necessary for training the classifier, requiring each sample in the training set to be labeled. The annotation information should include key details such as the era, kiln, and style of the ancient pottery bowl. This information can be obtained through expert authentication, historical documents, or other reliable sources.

[0045] To evaluate the model's performance, the training sample set can be divided into a training set, a validation set, and a test set. The training set is used to train the classifier, the validation set is used to tune the classifier's parameters and evaluate its performance, and the test set is used to finally evaluate the classifier's accuracy and generalization ability.

[0046] Step 3: Cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape characteristics of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape characteristics.

[0047] The extracted ancient porcelain bowl shapes are stored in a database. Hierarchical clustering is used to group the shape data, initially classifying them into categories based on different eras and kilns. Dimensionality reduction techniques (such as PCA) are used to extract key features, generating shape features for different eras and kilns. A classifier is trained using supervised learning algorithms (such as neural networks) to optimize the accuracy of the feature model. The data is then categorized and labeled to clarify its era, kiln, and style information.

[0048] S3.1 Select a database system for storing ancient porcelain bowl data, such as a relational database (MySQL, PostgreSQL) or a non-relational database (MongoDB). Based on the data volume and query requirements, set up the database environment and create the corresponding table structures or collections to store the data.

[0049] The extracted data on the shapes of ancient porcelain bowls (including key parameters such as rim diameter, base diameter, height, curvature, and thickness) were imported into the database. The integrity and accuracy of the data were ensured, and preprocessing was performed, such as data cleaning and format conversion, for subsequent analysis.

[0050] S3.2 Select a hierarchical clustering algorithm, such as AGNES (bottom-up) or DIANA (top-down). Based on the data characteristics and requirements, determine the distance metric (such as Euclidean distance, Manhattan distance, etc.) and clustering criteria (such as maximum distance, minimum distance, etc.).

[0051] The vessel shape data is input into a hierarchical clustering algorithm for cluster analysis. Based on the clustering results, categories are initially divided according to different eras and kilns. Clustering parameters, such as the number of clusters, can be adjusted to optimize the clustering effect. Evaluation metrics (such as silhouette coefficient, Calinski-Harabasz index, etc.) are used to evaluate the clustering results to ensure the effectiveness and accuracy of the clustering.

[0052] S3.3 Principal Component Analysis (PCA) as a dimensionality reduction technique. PCA is a commonly used unsupervised dimensionality reduction method that can project high-dimensional data into a low-dimensional space through linear transformation while preserving the main features of the data.

[0053] The clustered vessel shape data is input into the PCA algorithm for dimensionality reduction. Based on the dimensionality reduction results, key features are extracted. These features represent the vessel shape characteristics of different eras and kilns. The dimensionality reduction effect is evaluated by visualizing the dimensionality-reduced data or calculating the variance ratio before and after dimensionality reduction, ensuring that the dimensionality-reduced data still retains the main characteristics of the original data.

[0054] S3.4 Select a suitable supervised learning algorithm as the classifier, such as a neural network (including multilayer perceptrons, convolutional neural networks, etc.). Determine the structure (such as the number of layers, number of neurons, etc.) and parameters (such as the learning rate, number of iterations, etc.) of the neural network based on the characteristics and requirements of the data.

[0055] The dimensionality-reduced vessel shape data and corresponding classification labels (era, kiln, style, etc.) are input into a neural network for classifier training. During training, methods such as cross-validation are used to evaluate the classifier's performance, and the neural network's structure and parameters are adjusted to optimize performance.

[0056] The trained classifier is evaluated using a test set, and metrics such as accuracy, recall, and F1 score are calculated to ensure the accuracy and generalization ability of the classifier.

[0057] Step 4: Obtain the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape characteristics of the ancient porcelain bowl based on the three-dimensional structural data, input the shape characteristics into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

[0058] For newly unearthed or unidentified ancient porcelain bowls, the same method is used to extract their shape features. These extracted features are then input into an ancient porcelain bowl classifier. A machine learning classifier calculates similarity scores, automatically determining the era, kiln, and style, and outputting the identification results, including era, kiln, style, and confidence score.

[0059] S4.1 Conduct comprehensive and multi-dimensional data collection on the antique porcelain bowl to be identified.

[0060] For complete ancient porcelain bowls, 3D scanning or image acquisition is performed directly; for fragments, splicing and modeling correction are required before data acquisition. The acquired 3D structural data undergoes preprocessing, including noise removal, missing data filling, and surface smoothing, to improve data quality.

[0061] S4.2. Using the same feature extraction method as in step 3, extract key vessel features from the three-dimensional structural data, such as caliber, bottom diameter, height, curvature, thickness, and other possible morphological features.

[0062] S4.3 Input the standardized shape features into the classifier for classification calculation.

[0063] The classifier internally determines the degree of match between the input features and the features of each category by calculating the similarity (e.g., using Euclidean distance, cosine similarity, etc.). Based on the similarity calculation results, the classifier automatically selects the most matching category as the classification of the ancient porcelain bowl to be identified by era, kiln, and style.

[0064] S4.4 The classifier outputs the identification results, including the era, kiln, and style classification information of the ancient porcelain bowl to be identified. In addition to the classification information, the classifier should also provide a confidence score, indicating the reliability of the classification results. The confidence score can be calculated using the classifier's internal probability output, similarity score, or other evaluation metrics.

[0065] This method for classifying ancient porcelain bowl shapes based on artificial intelligence and big data algorithms extracts overall information from unearthed bowl samples to establish a database. Through cluster analysis, it extracts the most representative features of bowl shapes from different periods and eras as influencing factors. After extracting the shape information of unearthed ceramic bowls and inputting it into the database, the artificial intelligence and big data algorithms directly classify the ancient porcelain bowl shapes. Replacing traditional human visual identification with artificial intelligence reduces the difficulty of identification, making it faster and more convenient to determine the origin and age of unearthed ceramic bowls. It also addresses the problem of insufficient identification resources due to the scarcity of ancient ceramic identification experts. This invention, through artificial intelligence and big data algorithms, can efficiently and accurately process massive amounts of ancient porcelain bowl shape data, extracting typical features from different periods and kilns, thereby achieving automated classification and dating of ancient porcelain bowls. Compared with traditional manual identification, this invention not only significantly improves identification efficiency but also reduces subjective errors, making it particularly suitable for large-scale archaeological excavations or the organization of museum collections. Furthermore, this method can further improve classification accuracy through continuous database updates and algorithm optimization, providing strong technical support for ancient ceramic research.

[0066] Correspondingly, this application also provides a system for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms, including: The acquisition module is used to obtain the three-dimensional structural data of the ancient pottery bowl; The preprocessing module is used to standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Virtual sample data is generated based on the standardized data, and a training sample set is constructed based on the standardized data and the virtual sample data. The training module is used to cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape features of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape features. The identification module is used to acquire the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape features of the ancient porcelain bowl based on the three-dimensional structural data, input the shape features into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

[0067] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0068] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0069] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms as described in any of the above embodiments.

[0070] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), and wireless connection (including Wi-Fi, Bluetooth, Bluetooth Low Energy, and IEEE 802.11s-based communication technology).

[0071] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms as described in any of the above embodiments.

[0072] For descriptions of relevant parts of the system, electronic device, and computer-readable storage medium for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms provided in this application, please refer to the detailed description of the corresponding parts in the method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms provided in this application, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0073] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms, characterized in that, Includes the following steps: Step 1: Obtain the three-dimensional structural data of the ancient pottery bowl; Step 2: Standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Generate virtual sample data based on the standardized data. Construct a training sample set based on the standardized data and virtual sample data. Step 3: Cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape characteristics of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape characteristics. Step 4: Obtain the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape characteristics of the ancient porcelain bowl based on the three-dimensional structural data, input the shape characteristics into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

2. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, The acquisition of the three-dimensional structural data of the ancient pottery bowl includes: Three-dimensional scanning was performed on ancient porcelain bowls with unclear origins and dating. A three-dimensional model was constructed based on the three-dimensional scanning data. The three-dimensional model was then sliced ​​to obtain the three-dimensional structural data of the ancient pottery bowl.

3. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 2, characterized in that, When an ancient porcelain bowl with an unknown origin and date is incomplete, a model of the pottery bowl is created and corrected, and three-dimensional structural data is obtained based on the corrected model of the ancient pottery bowl.

4. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, The standardization process for the three-dimensional structural data of the ancient pottery bowl yields standardized data including: The three-dimensional structural data of the ancient pottery bowl were subjected to noise reduction, outlier removal, and normalization to obtain standardized data of the ancient pottery bowl.

5. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, The construction of the training sample set based on standardized data and virtual sample data includes: The standardized data is rotated, scaled, translated, or / and distorted to generate virtual sample data. The standardized data and virtual sample data are then integrated to obtain the training sample set.

6. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, Clustering the training sample set yields training samples from different eras and kilns. The vessel shape characteristics of these training samples from different eras and kilns are then obtained, including: Hierarchical clustering algorithm is used to classify the training sample set to obtain training samples from different eras and kilns. Principal component analysis is then used as a dimensionality reduction technique to reduce the dimensionality of the clustered training samples and obtain the shape features of the dimensionality-reduced training samples.

7. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, The training of the ancient porcelain bowl classifier based on vessel shape characteristics includes: A supervised learning algorithm was used as the classifier. The shape features and corresponding classification labels were input into the classifier for training. The performance of the classifier was evaluated using cross-validation. The structure and parameters of the classifier were adjusted to optimize the performance. The ancient porcelain bowl classifier was obtained through iterative training.

8. The method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms according to claim 1, characterized in that, The identification results of the ancient porcelain bowl include its era, kiln, and style.

9. A system for classifying the shapes of ancient porcelain bowls based on artificial intelligence big data algorithms, characterized in that, include: The acquisition module is used to obtain the three-dimensional structural data of the ancient pottery bowl; The preprocessing module is used to standardize the three-dimensional structural data of the ancient pottery bowl to obtain standardized data of the ancient pottery bowl. Virtual sample data is generated based on the standardized data, and a training sample set is constructed based on the standardized data and the virtual sample data. The training module is used to cluster the training sample set to obtain training samples from different eras and kilns, obtain the vessel shape features of the training samples from different eras and kilns, and train the ancient porcelain bowl classifier based on the vessel shape features. The identification module is used to acquire the three-dimensional structural data of the ancient porcelain bowl to be identified, obtain the shape features of the ancient porcelain bowl based on the three-dimensional structural data, input the shape features into the trained ancient porcelain bowl classifier, and obtain the identification result of the ancient porcelain bowl.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for classifying ancient porcelain bowl shapes based on artificial intelligence big data algorithms as described in any one of claims 1-8.

Images