Intelligent identification method and system for aging degree of porcelain glaze

Abstract

The invention discloses a porcelain glaze aging degree intelligent identification method and system, and the method comprises the steps: firstly, collecting the reflection characteristics of a target porcelain in visible light and near-infrared bands through a multispectral imaging device, and processing an original image to generate a standardized to-be-identified porcelain image, including an overall or local image; and inputting the image into a pre-trained target model to obtain a target value of aging degree grading of each glaze area, and further determining a target porcelain aging degree grading result. And finally, fusing the result with trace element data detected by an X-ray fluorescence spectrum to generate an identification report containing age inference and glaze preservation state evaluation, thereby realizing intelligent and accurate identification of the porcelain glaze aging degree.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an intelligent identification method and system for the aging degree of porcelain glaze. Background Art

[0002] Traditionally, the identification of the degree of aging of porcelain glaze mainly relies on expert experience and limited instrument testing, which has the problems of strong subjectivity, low efficiency, and accuracy greatly affected by human factors. Expert identification may produce different conclusions due to differences in personal knowledge and experience, and it is difficult to make accurate judgments for some complex situations. Instrument testing can often only provide data in a single dimension and cannot comprehensively and comprehensively evaluate the degree of aging of porcelain glaze. Therefore, there is an urgent need for an objective, accurate, and efficient intelligent identification method to meet the growing demand in the field of porcelain identification. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent identification method and system for the aging degree of porcelain glaze.

[0004] In a first aspect, an embodiment of the present invention provides an intelligent identification method for the aging degree of porcelain glaze, comprising:

[0005] The reflectance characteristics of the target porcelain in the visible light and near-infrared bands are collected by a multispectral imaging device, and the glaze reflection suppression process is performed on the collected original image to generate a standardized porcelain image to be identified; the porcelain image to be identified is a porcelain image of a preset image type of the target porcelain, and the preset image type is an overall image or a partial image;

[0006] Loading the porcelain image to be identified into a pre-trained target glaze aging degree identification model, and obtaining a glaze aging degree target value of the aging degree classification result of each glaze area in the porcelain image output by the target glaze aging degree identification model;

[0007] Determining a target aging degree grading result of the target porcelain according to a target aging degree value of the glaze surface of the aging degree grading result of each glaze surface area in the porcelain image;

[0008] The target aging degree classification results are temporally and spatially aligned and fused with the trace element data detected by X-ray fluorescence spectroscopy to generate a porcelain glaze aging degree identification report that includes the age inference results and glaze preservation status assessment.

[0009] In a second aspect, an embodiment of the present invention provides a server system, including a server, wherein the server is used to execute the method described in the first aspect.

[0010] Compared with the prior art, the beneficial effects provided by the present invention include: using the intelligent identification method and system for the aging degree of porcelain glaze disclosed in the present invention, by using multi-spectral imaging equipment to collect the reflection characteristics of the target porcelain in the visible light and near-infrared bands, processing the original image to generate a standardized image of the porcelain to be identified, including an overall or partial image. The image is input into a pre-trained target model to obtain the target value of the aging degree classification of each glaze area, and then determine the aging degree classification result of the target porcelain. Finally, the result is fused with the trace element data detected by X-ray fluorescence spectroscopy to generate an identification report containing age inference and glaze preservation status assessment, so as to realize intelligent and accurate identification of the aging degree of porcelain glaze. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can also be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic flow chart of the steps of the intelligent identification method for the aging degree of porcelain glaze provided by an embodiment of the present invention;

[0013] Figure 2 A schematic block diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0015] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.

[0016] In order to solve the technical problems in the aforementioned background technology, Figure 1 This is a flow chart of the intelligent identification method for the degree of aging of porcelain glaze provided in an embodiment of the present disclosure. The intelligent identification method for the degree of aging of porcelain glaze is introduced in detail below.

[0017] Step S201, collecting the reflection characteristics of the target porcelain in the visible light and near-infrared bands by a multispectral imaging device, performing glaze reflection suppression processing on the collected original image, and generating a standardized porcelain image to be identified; the porcelain image to be identified is a porcelain image of a preset image type of the target porcelain, and the preset image type is an overall image or a partial image;

[0018] Step S202, loading the porcelain image to be identified into a pre-trained target glaze aging degree identification model, and obtaining a glaze aging degree target value of the aging degree grading result of each glaze area in the porcelain image output by the target glaze aging degree identification model;

[0019] Step S203, determining a target aging degree grading result of the target porcelain according to the aging degree target value of the glaze surface area in the porcelain image;

[0020] Step S204, performing spatiotemporal alignment fusion based on the target aging degree classification result and the trace element data detected by X-ray fluorescence spectroscopy, to generate a porcelain glaze aging degree identification report including the age inference result and the glaze preservation status assessment.

[0021] In an embodiment of the present invention, illustratively, the server controls the multispectral imaging device to collect images of the target porcelain. Assuming that the target porcelain is a blue-and-white porcelain bottle from the Ming Dynasty, the multispectral imaging device will scan it in the visible light and near-infrared bands to obtain the reflection characteristics of the target porcelain in these bands. During the acquisition process, the device will shoot the blue-and-white porcelain bottle from different angles and distances to ensure that its surface features are fully captured. For example, first shoot from directly above to obtain an image of the top of the bottle body, then shoot the main part of the bottle body from different angles from the side, and also take close-up photos of details such as the bottleneck and the bottom of the bottle. The collected original image will have the problem of glaze reflection, which will affect the accuracy of subsequent identification. After receiving the original image, the server will immediately perform glaze reflection suppression processing on it. Taking an area with severe reflection on the bottle body as an example, the server analyzes the pixel value distribution of the area through a specific algorithm and identifies the high-brightness pixels of the reflective part. Then, according to the pixel characteristics of the surrounding normal area, the pixel value of the reflective area is adjusted to reduce its brightness so that it is more consistent with the brightness of the surrounding area. After processing, the server will generate a standardized image of the porcelain to be identified from the image. The preset image types here are overall images or partial images. For this blue and white porcelain bottle, the server will generate an overall image to fully display the shape, patterns and other information of the bottle body; at the same time, it will also generate partial images of some key parts, such as areas with typical patterns on the bottle body and areas with obvious changes in the glaze, so as to conduct a more accurate aging analysis later. The server loads the generated porcelain image to be identified into the pre-trained target glaze aging degree identification model. Assuming that the image of the target porcelain includes an overall image and multiple partial images, the server will input these images into the model in turn. After receiving the image, the model will analyze and process the image. For the overall image, the model will first identify the general outline of the porcelain and the overall glaze characteristics, and judge the overall color and gloss of the glaze. Taking the overall image of the blue and white porcelain bottle as an example, the model can detect the overall color distribution of the glaze of the bottle body, whether there is uneven color, etc. For partial images, the model will analyze the characteristics of each glaze area in more detail. For example, in a local image with cracks, the model will detect the density and shape of the cracks and analyze the color changes at the cracks; at the same time, it will also observe the distribution characteristics of microscopic bubbles, including the size, number, and distribution pattern of the bubbles. After analysis and processing by the model, the target value of glaze aging for the aging degree grading result of each glaze area in the porcelain image is finally output. For example, for a glazed area on the main body of a blue and white porcelain vase, the target aging degree output by the model may indicate that the area is in a moderately aged state; while for a small area at the bottleneck, the output target value shows that its aging degree is relatively light. After the server receives the target glaze aging degree value of the aging degree grading result of each glazed area output by the model, it will conduct a comprehensive analysis of these values ​​to determine the target aging degree grading result of the target porcelain.For this blue and white porcelain vase, the server will count the area proportion of the glazed area of ​​different aging degree grading results. Assuming that the aging degree target value of most areas of the bottle body is shown as moderate aging, and only a few areas are lightly aged and severely aged, the server will calculate a comprehensive aging degree score based on the area weight of each area. Then, according to the pre-set aging degree grading standard, this comprehensive score is mapped to the corresponding grading result. For example, if the comprehensive score falls within the moderate aging range, the server will determine that the target aging degree grading result of this blue and white porcelain vase is moderate aging. After determining the target aging degree grading result of the target porcelain, the server will obtain the trace element data detected by X-ray fluorescence spectroscopy. Assuming that this blue and white porcelain vase has been tested by X-ray fluorescence spectroscopy, the test data will show the types and contents of various trace elements in the porcelain. The server will align and fuse the target aging degree grading results with the trace element data in time and space. Taking the blue and white porcelain vase as an example, the server will analyze the changes in the trace element content corresponding to different aging degree areas. For example, it is found that the content of certain trace elements in the moderately aged area is different from that in the lightly aged area. The server will further study the relationship between this change and the age and preservation status of the porcelain. Through a comprehensive analysis of the target aging degree classification results and trace element data, the server will generate a porcelain glaze aging degree identification report that includes the age inference results and the glaze preservation status assessment. For this Ming Dynasty blue and white porcelain vase, the report will infer the approximate age range of the porcelain and assess the preservation status of its glaze, such as the integrity of the glaze, whether there are obvious signs of damage or repair, etc. At the same time, the report will also give some suggestions on the protection and collection of the porcelain to help collectors better preserve and manage the porcelain.

[0022] In the embodiment of the present invention, the target glaze aging degree identification model is trained in the following manner and can be implemented through the following examples.

[0023] Acquire an initial glaze aging degree identification model, wherein the initial glaze aging degree identification model includes a first glaze recognition model and at least one second glaze recognition model, wherein each first image instance of the first glaze recognition model includes a porcelain image of an overall image of a target porcelain, and each second image instance of the second glaze recognition model includes a porcelain image of a partial image of the target porcelain;

[0024] Acquire a first aging feature mapping parameter of the first glaze surface recognition model and a second aging feature mapping parameter of each of the second glaze surface recognition models when training is completed in the current training phase;

[0025] For each second glaze recognition model, a first orientation parameter group for fusion in the first aging feature mapping parameter and a second orientation parameter group for fusion in the second aging feature mapping parameter are determined according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed; wherein the first parameter update direction is the model parameter update direction of the first aging feature mapping parameter of the first feature mapping component, and the second parameter update direction is the model parameter update direction of the second aging feature mapping parameter of the second feature mapping component;

[0026] For each second glaze recognition model, the first orientation parameter group when the training is completed in the current training stage is merged with the second orientation parameter group of the second glaze recognition model to obtain the second aging feature mapping parameters corresponding to the training in the next training stage, and based on the obtained second aging feature mapping parameters and each second image instance of the second glaze recognition model, the second glaze recognition model is trained once to obtain the second aging feature mapping parameters when the training is completed in the next training stage;

[0027] The second orientation parameter group of each second glaze recognition model when training is completed in the next training stage and the first orientation parameter group of the first glaze recognition model when training is completed in the current training stage are merged to obtain the first aging feature mapping parameters corresponding to the training in the next training stage, and the first glaze recognition model is trained once based on the obtained first aging feature mapping parameters and each first image instance to obtain the first aging feature mapping parameters when training is completed in the next training stage;

[0028] The first glaze recognition model at the completion of the last training phase is determined as the first glaze aging degree identification model, and the second glaze recognition model at the completion of the last training phase is determined as the second glaze aging degree identification model;

[0029] The glaze aging degree identification model including the feature extraction components corresponding to each image type in the preset image types in the trained first glaze aging degree identification model or the second glaze recognition model is determined as the target glaze aging degree identification model.

[0030] In an embodiment of the present invention, illustratively, the server sets out to train the target glaze aging degree identification model, and first obtains the initial glaze aging degree identification model. The initial model includes a first glaze recognition model and multiple second glaze recognition models. The server extracts each first image instance of the first glaze recognition model from the database, and these instances are all overall images of the target porcelain. For example, there are a batch of porcelains from different eras and different kilns, such as Song Dynasty Ru Kiln porcelain plates, Yuan Dynasty blue and white porcelain bottles, Qing Dynasty famille rose porcelain bowls, etc. The server collects their complete appearance images as first image instances. For Song Dynasty Ru Kiln porcelain plates, the overall image clearly shows the shape, size, overall glaze color and gloss of the porcelain plate. At the same time, the server obtains each second image instance of the second glaze recognition model, which are partial images of the target porcelain. Still taking the Song Dynasty Ru Kiln porcelain plate as an example, the server will select the area with cracks on the edge of the porcelain plate, the area with inscriptions on the bottom of the plate, and the area with tiny bubbles on the glaze surface to shoot partial close-up images as second image instances. These partial images can present the microscopic features of the porcelain glaze in more detail. After the training is completed in the current training phase, the server obtains the first aging feature mapping parameters of the first glaze recognition model and the second aging feature mapping parameters of each second glaze recognition model. The first aging feature mapping parameter is the parameter learned by the first feature mapping component in the process of processing the first image instance, which reflects the mapping rule of the first glaze recognition model to the aging features of the porcelain in the overall image. Taking the overall image of the blue and white porcelain bottle of the Yuan Dynasty as an example, the first aging feature mapping parameter can reflect the mapping relationship of aging features such as the overall color change of the bottle body and the degree of blurring of the decoration. The second aging feature mapping parameters of each second glaze recognition model are learned by the second feature mapping component for the second image instance. For example, for the local image of the famille rose porcelain bowl of the Qing Dynasty, the second aging feature mapping parameter can reflect the mapping of local aging features such as the famille rose peeling area and the degree of color fading on the bowl wall. For each second glaze recognition model, the server determines the first orientation parameter group and the second orientation parameter group for fusion according to the directional consistency between the first parameter update direction and the second parameter update direction when each training phase in the latest target training phase is completed. The first parameter update direction is the model parameter update direction of the first aging feature mapping parameter of the first feature mapping component, and the second parameter update direction is the model parameter update direction of the second aging feature mapping parameter of the second feature mapping component. The server determines the degree of consistency by comparing the consistency of the two directions, such as calculating the cosine value of the angle between them. Assume that the server has three second glaze recognition models, which process local images of different parts of Song Dynasty Ru kiln porcelain plates respectively. When comparing the first parameter update direction and the second parameter update direction, the server finds that when a second glaze recognition model processes a local image of the cracked area of ​​the porcelain plate, its second parameter update direction is less consistent with the first parameter update direction.The server further analyzes the texture analysis components of the second feature mapping component of the second glaze recognition model, determines those texture analysis components with low consistency as the second texture analysis components for fusion, and uses the parameters of these components as the second orientation parameter group. Then, the server finds the components other than the corresponding texture analysis components in the first feature mapping component based on these second texture analysis components, and uses the parameters of these components as the first orientation parameter group. For each second glaze recognition model, the server fuses the first orientation parameter group when the training is completed in the current training phase with the second orientation parameter group of the second glaze recognition model. The server first obtains the first feature contribution corresponding to the first glaze recognition model and the second feature contribution corresponding to each second glaze recognition model, which are pre-set or calculated based on the performance and feature extraction capabilities of the model in the previous training phase. For the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier in the second orientation parameter group of each second feature mapping component, the server obtains the model parameter update direction when the current training phase is completed. For example, when processing the second feature mapping component of a local image of a Qing Dynasty famille rose porcelain bowl, a texture analysis component thereof corresponds to a specific feature level identifier, and the server obtains the model parameter update direction of the component when the current training phase is completed. According to these update directions, the server calculates the third feature contribution of each second feature mapping component, and the contribution is negatively correlated with the parameter update direction of the corresponding second aging feature mapping parameter. That is to say, the component with a large change in parameter update direction has a relatively low third feature contribution. Then, the server performs linear superposition processing on the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier with the third feature contribution of the corresponding second feature mapping component, and obtains the first integration result corresponding to the texture analysis component with the same feature level identifier in the next stage. The server determines the mean of the first integration result of the texture analysis component of each feature level identifier as the third orientation parameter group, and then based on the first feature contribution and the second feature contribution, performs weighted fusion on the first orientation parameter group and the third orientation parameter group when the first glaze recognition model completes the training in the current training stage, and obtains the second aging feature mapping parameters corresponding to the training in the next training stage. The server performs a training iteration on the second glaze recognition model based on the obtained second aging feature mapping parameters and each second image instance of the second glaze recognition model. For example, for a second glaze recognition model that processes a local image of a Song Dynasty Ru kiln porcelain plate, the server uses the new second aging feature mapping parameters to train it, inputs the corresponding local image instance, adjusts the model parameters, and finally obtains the second aging feature mapping parameters when the next training phase completes the training. The server fuses the second orientation parameter group of each second glaze recognition model when the next training phase completes the training and the first orientation parameter group of the first glaze recognition model when the current training phase completes the training.The specific fusion process is similar to the above fusion to obtain the second aging feature mapping parameters corresponding to the next training stage training, and the feature contribution and other factors should also be considered. After the server obtains the first aging feature mapping parameters corresponding to the next training stage training, it performs a training iteration on the first glaze recognition model based on the parameters and each first image instance. For example, the server uses the new first aging feature mapping parameters to process the overall image of the Yuan Dynasty blue and white porcelain bottle, and continuously adjusts the model parameters so that the model can better learn the aging features in the overall image, and finally obtains the first aging feature mapping parameters when the training is completed in the next training stage. After multiple training stages of iterative training, the server determines the first glaze recognition model when the last training stage is completed as the first glaze aging degree identification model, and determines the second glaze recognition model when the last training stage is completed as the second glaze aging degree identification model. The server will select the glaze aging degree identification model including the feature extraction components corresponding to each image type in the preset image type from the first glaze aging degree identification model or the second glaze recognition model that has been trained as the target glaze aging degree identification model according to the preset image type, that is, the overall image or the partial image. If the porcelain image to be identified is an overall image, the server will select the first glaze aging degree identification model; if it is a partial image, the server will select the corresponding second glaze aging degree identification model. In this way, the trained target glaze aging degree identification model can be used for subsequent intelligent identification of porcelain glaze aging degree.

[0031] In the embodiment of the present invention, the component architecture of the feature mapping components of each glaze recognition model in the initial glaze aging degree identification model is consistent, and the feature mapping components of each glaze recognition model include a first number of texture analysis components;

[0032] For each second glaze recognition model, the method of determining a first directional parameter group for fusion in the first aging feature mapping parameters and a second directional parameter group for fusion in the second aging feature mapping parameters of the second glaze recognition model according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed can be implemented through the following example.

[0033] Determine, according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed, a second texture analysis component for fusion among the texture analysis components of the second feature mapping component of the second glaze recognition model, and use the parameters of each second texture analysis component as a second directional parameter group for fusion in the second aging feature mapping parameters of the second glaze recognition model;

[0034] According to each of the second texture analysis components, a first texture analysis component for fusion among each of the texture analysis components of the first feature mapping component is determined, and the parameters of each of the first texture analysis components are used as the first orientation parameter group for fusion in the first aging feature mapping parameters, wherein the first texture analysis component is a texture analysis component other than the texture analysis components corresponding to each of the second texture analysis components among the texture analysis components of the first feature mapping component.

[0035] In an embodiment of the present invention, exemplarily, in the initial glaze aging degree identification model obtained by the server, the feature mapping component architecture of each glaze recognition model is the same, and all contain the first number of texture analysis components. During the training process, the server determines the first orientation parameter group and the second orientation parameter group for fusion for each second glaze recognition model. When each training stage of the latest target training stage is completed, the server compares the consistency of the first parameter update direction and the second parameter update direction. Taking a Tang Dynasty Yue Kiln celadon pot as an example, the first glaze recognition model processes its overall image, and the second glaze recognition model processes the image of the local crack area of ​​the pot body. The server calculates the parameter update direction of each texture analysis component in the first feature mapping component and the second feature mapping component at the end of each training stage. For the second glaze recognition model that processes the image of the local crack area of ​​the pot body, the server selects the second texture analysis component for fusion according to the direction consistency. For example, the server finds that in multiple training stages, the parameter update direction of a certain texture analysis component has a large angle with the parameter update direction of the corresponding texture analysis component in the first feature mapping component, and the consistency is low. The server determines this texture analysis component as the second texture analysis component for fusion. The server will make such a judgment on all the texture analysis components of the second glaze recognition model, select the components that meet the conditions, and use the parameters of these second texture analysis components as the second directional parameter group for fusion in the second aging feature mapping parameters. Then, the server finds the first texture analysis component for fusion in the first feature mapping component based on the determined second texture analysis component. Taking the Tang Dynasty Yue Kiln celadon pot as an example, the server will check the texture analysis components of the first feature mapping component, exclude those components corresponding to the second texture analysis component, and the remaining components are the first texture analysis components. For example, the second texture analysis component corresponds to the texture feature of the crack of the pot body, so the texture analysis component that analyzes other features such as overall color and overall shape in the first feature mapping component may be determined as the first texture analysis component. The server uses the parameters of these first texture analysis components as the first directional parameter group for fusion in the first aging feature mapping parameters. In this way, the server accurately determines the first directional parameter group and the second directional parameter group of each second glaze recognition model, which provides a basis for subsequent parameter fusion and model training iteration, and helps to improve the model's ability to recognize the degree of aging of porcelain glaze, thereby completing the identification of the degree of aging of porcelain glaze more accurately.

[0036] In an embodiment of the present invention, the second orientation parameter group of each second glaze recognition model when the training is completed in the next training stage and the first orientation parameter group of the first glaze recognition model when the training is completed in the current training stage are fused to obtain the first aging feature mapping parameters corresponding to the training in the next training stage, which can be implemented through the following examples.

[0037] Acquire a first feature contribution corresponding to the first glaze surface recognition model, and a second feature contribution corresponding to each of the second glaze surface recognition models;

[0038] Integrating the second orientation parameter groups of each of the second glaze recognition models when training is completed in the next training phase to obtain a third orientation parameter group;

[0039] Based on the first feature contribution and the second feature contribution, the first orientation parameter group and the third orientation parameter group of the first glaze recognition model when the training is completed in the current training stage are weighted fused to obtain the first aging feature mapping parameters corresponding to the next training stage.

[0040] In an embodiment of the present invention, exemplarily, when the server trains each second glaze recognition model and the first glaze recognition model, it is necessary to fuse the second directional parameter group of each second glaze recognition model when the training is completed in the next training stage with the first directional parameter group of the first glaze recognition model when the training is completed in the current training stage, so as to obtain the first aging feature mapping parameters corresponding to the training in the next training stage. First, the server obtains the first feature contribution corresponding to the first glaze recognition model, and the second feature contribution corresponding to each second glaze recognition model. Taking the identification of a batch of ancient porcelain as an example, the first glaze recognition model processes the overall image of the porcelain. It can grasp the overall aging characteristics of the porcelain from a macro perspective, such as the overall color, shape changes, etc. The server determines a relatively high first feature contribution, such as 0.6, based on its performance in previous training and its ability to extract overall aging characteristics. Each second glaze recognition model processes a local image of the porcelain, and can deeply analyze the local aging characteristics, such as cracking details, bubble distribution, etc. The server determines different second feature contributions for each second glaze recognition model based on factors such as the accuracy of their extraction of different local features. For example, the second feature contributions of the three second glaze recognition models are 0.1, 0.15, and 0.15, respectively. Then, the server integrates the second orientation parameter groups of each second glaze recognition model when it completes training in the next training stage to obtain a third orientation parameter group. Assuming that the three second glaze recognition models process the local images of the bottle mouth, bottle body, and bottle bottom of the porcelain respectively, after they complete training in the next training stage, the server integrates the corresponding parameters in the second orientation parameter group of each model. For example, for the parameters of the texture analysis component, the server calculates the average or weighted average of the parameters of the component of each model, and finally obtains the third orientation parameter group. Finally, based on the first feature contribution and the second feature contribution, the server weightedly fuses the first orientation parameter group and the third orientation parameter group of the first glaze recognition model when it completes training in the current training stage. The server multiplies the first directional parameter group by the first feature contribution 0.6, multiplies the third directional parameter group by the sum of the three second feature contributions 0.4, and then adds the two results to obtain the first aging feature mapping parameters corresponding to the next training stage. Through this fusion method, the server combines the first glaze recognition model's grasp of the overall features and the analysis of the local features by each second glaze recognition model, so that the first aging feature mapping parameters in the next stage can more comprehensively and accurately reflect the aging characteristics of porcelain, thereby improving the accuracy of the model's identification of the degree of aging of porcelain glaze.

[0041] In the embodiment of the present invention, the second orientation parameter groups of each of the second glaze recognition models when training is completed in the next training stage are integrated to obtain a third orientation parameter group, which can be implemented through the following example.

[0042] For the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier in the second orientation parameter group of each second feature mapping component, the target second orientation parameter group of each second glaze recognition model when the training is completed in the next training stage is integrated to obtain the first integration result corresponding to the texture analysis component with the same feature level identifier in the next training stage;

[0043] The first integration results of the texture analysis components of each feature level identifier are integrated to obtain the third orientation parameter group.

[0044] In an embodiment of the present invention, exemplarily, the server will operate according to specific steps in the process of integrating the second orientation parameter groups of each second glaze recognition model when the training is completed in the next training stage to obtain the third orientation parameter group. The server faces multiple second glaze recognition models, and the second feature mapping component of each model has its own second orientation parameter group. Taking the identification of a Song Dynasty official kiln porcelain bowl as an example, it is assumed that there are three second glaze recognition models that process the local images of the rim, bowl body and bowl bottom of the porcelain bowl respectively. The server pays attention to the target second orientation parameter group of the texture analysis component corresponding to the same feature level identification in the second orientation parameter group of each second feature mapping component. For example, for the feature level identification of analyzing the glaze crackle texture, the three second glaze recognition models have corresponding texture analysis components, and their parameters constitute the target second orientation parameter group. The server integrates the target second orientation parameter groups of these three models when the training is completed in the next training stage. For the target second orientation parameter group for the crackle texture of the rim, bowl body and bowl bottom of the porcelain bowl, the server calculates the average value. If the corresponding parameters of the rim model are 0.2, the bowl body model is 0.3, and the bowl bottom model is 0.25, then the average value is (0.2+0.3+0.25)÷3=0.25, and this 0.25 is the first integrated result corresponding to the texture analysis component of the same feature level identification (analyzing the glaze crackle texture) in the next training stage. The server performs this operation on the texture analysis component of each feature level identification. In addition to the glaze crackle texture, there are also texture analysis components for different feature level identifications such as bubble distribution and color changes. The server will calculate the respective first integrated results according to the above method. The server integrates the first integrated results of the texture analysis components of each feature level identification to obtain the third directional parameter group. For the first integrated results of analyzing different features, the server integrates them again. For example, the first integrated results of analyzing crackle texture, bubble distribution, color change and other features are processed according to certain rules (such as averaging or weighted averaging again) to finally form the third directional parameter group. Through such an integration process, the server effectively integrates the information of each second glaze recognition model for different local images and different feature levels, so that the third directional parameter group can comprehensively reflect the training results of each second glaze recognition model, laying the foundation for the subsequent fusion with the first directional parameter group to obtain more accurate first aging feature mapping parameters corresponding to the next training stage, thereby improving the performance of the porcelain glaze aging degree identification model.

[0045] In an embodiment of the present invention, the target second orientation parameter groups of each second glaze recognition model when completing training in the next training stage are integrated to obtain the first integration result corresponding to the texture analysis component with the same feature level identification in the next training stage, which can be implemented through the following example.

[0046] Obtaining a model parameter update direction for each second aging feature map parameter when the current training phase is completed;

[0047] Obtaining a third feature contribution of each second feature mapping component according to a model parameter update direction of each second aging feature mapping parameter when the current training phase is completed, wherein the third feature contribution is negatively correlated with a parameter update direction of the corresponding second aging feature mapping parameter;

[0048] For the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier in the second orientation parameter group of each second feature mapping component, a linear superposition process is performed with the third feature contribution of the corresponding second feature mapping component to obtain the first integration result corresponding to the texture analysis component with the same feature level identifier in the next stage;

[0049] The integrating the first integrated results of the texture analysis components identified by each feature level to obtain the third orientation parameter group may be implemented through the following example.

[0050] The mean value of the first integrated result of the texture analysis component identified by each feature level is determined as the third orientation parameter group.

[0051] In an embodiment of the present invention, exemplarily, when the server integrates the target second directional parameter groups of each second glaze recognition model to obtain the first integration result and then obtain the third directional parameter group, it will operate according to a specific process. The server obtains the model parameter update direction of each second aging feature mapping parameter when the current training stage is completed. Taking the identification of Ming Dynasty blue and white porcelain bottles as an example, there are three second glaze recognition models that process the local images of the bottle mouth, bottle body, and bottle bottom respectively. After the current training stage is completed, the server analyzes the changes in the second aging feature mapping parameters of each model to determine its parameter update direction. For example, for the model that processes the local bottle mouth, its second aging feature mapping parameter changes from value A to value B in the current stage, and the server determines the update direction of this parameter of the model by calculating the difference between the two. According to the model parameter update direction of each second aging feature mapping parameter when the current training stage is completed, the server obtains the third feature contribution of each second feature mapping component. Since the third feature contribution is negatively correlated with the parameter update direction of the corresponding second aging feature mapping parameter, if the update direction of a second aging feature mapping parameter changes greatly, it means that the learning fluctuation of the model at this stage is large and the stability is poor, and its third feature contribution is low. For example, if the update direction of the model parameter for the local part of the bottle body changes dramatically, the server assigns it a lower third feature contribution, such as 0.2; while the update of the model parameter for the local part of the bottle bottom is relatively stable, the server assigns it a higher third feature contribution, such as 0.8. For the second orientation parameter group of each second feature mapping component, the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier, the server performs linear superposition processing with the third feature contribution of the corresponding second feature mapping component. For example, for the texture analysis component with the feature level identifier of analyzing the local glaze color change of the blue and white porcelain bottle, the target second orientation parameter groups of the bottle mouth, bottle body, and bottle bottom models are 0.3, 0.4, and 0.5 respectively, and the corresponding third feature contributions are 0.3, 0.2, and 0.8. The server performs linear superposition: 0.3×0.3+0.4×0.2+0.5×0.8=0.53, and this 0.53 is the first integration result corresponding to the texture analysis component with the same feature level identifier in the next stage. The server integrates the first integrated results of the texture analysis components of each feature level identifier to obtain the third directional parameter group. The server will analyze the first integrated results of the texture analysis components of different feature level identifiers (such as glaze color changes, crack density, etc.) and calculate the average. Assuming that the first integrated results of the analysis of three different feature level identifiers are 0.53, 0.6, and 0.55, respectively, the server calculates (0.53+0.6+0.55)÷3=0.56, and determines 0.56 as the third directional parameter group. Through such operations, the server can integrate the information of each second glaze recognition model to provide more accurate data for subsequent fusion with the first directional parameter group, thereby improving the performance of the porcelain glaze aging degree identification model.

[0052] In an embodiment of the present invention, for each second glaze recognition model, a first directional parameter group for fusion in the first aging feature mapping parameter and a second directional parameter group for fusion in the second aging feature mapping parameter are determined according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed. The implementation can be performed through the following examples.

[0053] If the directional consistency between the first parameter update direction and the second parameter update direction at the completion of each training stage in the latest target training stage does not exceed the preset consistency threshold, the feature selection matrix of the second glaze recognition model at the completion of this training stage is updated to obtain the feature selection matrix corresponding to the training of the next training stage; wherein the feature selection matrix is ​​used to indicate the second directional parameter group used for fusion in the second aging feature mapping parameters of the second glaze recognition model, and the feature selection matrix corresponding to the training of the first training stage is a preset feature selection matrix;

[0054] According to the feature selection matrix corresponding to the training in the next training stage, a first orientation parameter group for fusion in the first aging feature mapping parameters and a second orientation parameter group for fusion in the second aging feature mapping parameters of the second glaze recognition model are determined, wherein the first orientation parameter group is an array of the first aging feature mapping parameters excluding the parameters corresponding to the second orientation parameter group.

[0055] In an embodiment of the present invention, exemplarily, when processing each second glaze recognition model, the server will determine the directional parameter group for fusion based on the directional consistency between the first parameter update direction and the second parameter update direction. The server takes the identification of a Qing Dynasty famille rose porcelain bottle as an example. The porcelain bottle has multiple second glaze recognition models that process local images of different parts such as the bottleneck, bottle body, and bottle bottom. After each training stage of the latest target training stage is completed, the server will calculate the directional consistency between the first parameter update direction and the second parameter update direction of each second glaze recognition model. Assume that the preset consistency threshold set by the server is 0.8. For the second glaze recognition model that processes the bottleneck local image, the server checks the directional consistency data when each training stage of the latest target training stage is completed, and finds that the directional consistency of all stages is lower than 0.8. At this time, the server will update the feature selection matrix of the second glaze recognition model when this training stage is completed. The feature selection matrix is ​​like a signboard, which can clearly indicate which parameters in the second aging feature mapping parameters of the second glaze recognition model are the second directional parameter group for fusion. The feature selection matrix corresponding to the training in the first training phase is pre-set, for example, the parameters corresponding to certain elements in the matrix are initially set as parameters for fusion. The server adjusts the elements in the matrix according to the fact that the directional consistency does not exceed the threshold, for example, the element values ​​that originally indicate that certain texture analysis component parameters are used for fusion are changed, so as to obtain the feature selection matrix corresponding to the training in the next training phase. The server determines the directional parameter group for fusion according to the updated feature selection matrix corresponding to the training in the next training phase. For the second glaze recognition model that processes the bottleneck local image, the server finds the corresponding parameters in the second aging feature mapping parameters as the second directional parameter group based on the new feature selection matrix. Then, the server excludes the parameters corresponding to the second directional parameter group from the first aging feature mapping parameters, and the remaining parameter array is the first directional parameter group. The same operation is also applied to other second glaze recognition models that process local images such as the bottle body and the bottom of the bottle. In this way, the server can dynamically adjust the feature selection matrix according to the directional consistency, and then accurately determine the first directional parameter group and the second directional parameter group for fusion, so that the subsequent parameter fusion and model training are more accurate, improve the performance of the porcelain glaze aging degree identification model, and more accurately identify the glaze aging degree of Qing Dynasty famille rose porcelain bottles.

[0056] In the embodiment of the present invention, the directional consistency between the first parameter update direction and the second parameter update direction includes a consistency coefficient between the first sub-parameter update direction and the second sub-parameter update direction of each corresponding texture analysis component in the first glaze recognition model and the second glaze recognition model;

[0057] If the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed does not exceed the preset consistency threshold, the feature selection matrix of the second glaze recognition model when this training stage is completed is updated to obtain the feature selection matrix corresponding to the next training stage training, which can be implemented through the following example.

[0058] For each texture analysis component in each second glaze recognition model, if the consistency coefficient of the texture analysis component at the completion of each training stage in the latest target training stage does not exceed the preset consistency threshold, the texture analysis component is determined as a target texture analysis component;

[0059] In the feature selection matrix of the second glaze recognition model at the completion of the current training phase, the parameters corresponding to the target texture analysis component are unassigned for fusion, and the feature selection matrix corresponding to the next training phase is obtained.

[0060] In an embodiment of the present invention, exemplarily, when the server processes the feature selection matrix update of the second glaze recognition model, it will operate according to the consistency coefficient between the first and second parameter update directions. Taking the identification of a Song Dynasty Ru kiln porcelain plate as an example, there are multiple second glaze recognition models that process different local images such as the edge, center, and pattern of the porcelain plate, and the first glaze recognition model processes the overall image of the porcelain plate. After the completion of each training stage of the latest target training stage, the server will calculate the consistency coefficient between the first sub-parameter update direction and the second sub-parameter update direction of each corresponding texture analysis component in the first glaze recognition model and the second glaze recognition model. For example, for the second glaze recognition model that processes the edge image of the porcelain plate, one of the texture analysis components is responsible for analyzing the cracking characteristics of the edge glaze. The server will compare its second sub-parameter update direction with the first sub-parameter update direction of the texture analysis component corresponding to the analysis of the overall cracking characteristics in the first glaze recognition model to obtain the consistency coefficient. Assume that the preset consistency threshold is set to 0.7. For each texture analysis component in each second glaze recognition model, the server will check its consistency coefficient when each training stage of the latest target training stage is completed. If in the second glaze recognition model that processes the central image of the porcelain plate, there is a texture analysis component responsible for analyzing the bubble distribution characteristics of the central glaze, and its consistency coefficient is lower than 0.7 at the completion of each training stage, the server will determine this texture analysis component as the target texture analysis component. The server will operate in the feature selection matrix of the second glaze recognition model at the completion of this training stage. The feature selection matrix originally indicates which parameters in the second aging feature mapping parameters of the second glaze recognition model are used for fusion. The server releases the parameters corresponding to the target texture analysis component from the designation for fusion. For example, a row of elements in the original feature selection matrix corresponds to the texture analysis component parameters that process the bubble distribution characteristics in the center of the porcelain plate, and the value of the element indicates that this parameter is used for fusion. The server modifies the value of this element so that it no longer indicates that the parameter is used for fusion, thereby obtaining the feature selection matrix corresponding to the training in the next training stage. Similarly, the server will perform the same operation on the second glaze recognition model that processes other local images of the porcelain plate. In this way, the server dynamically adjusts the feature selection matrix according to the consistency coefficient and accurately determines the parameters used for fusion, providing a more accurate basis for subsequent parameter fusion and model training, improving the performance of the porcelain glaze aging degree identification model, and more accurately identifying the glaze aging of Song Dynasty Ru kiln porcelain plates.

[0061] In the embodiment of the present invention, the model parameter update direction of the second aging feature mapping parameter when the current training stage is completed can be implemented through the following examples.

[0062] Fusing a first orientation parameter group of the first aging feature mapping parameter when the previous training stage is completed with a second orientation parameter group of the second feature mapping component when the current training stage is completed, to obtain a second fusion result corresponding to the current training stage;

[0063] Obtaining a model parameter update direction of the second aging feature mapping parameter when the current training stage is completed according to a deviation between the second aging feature mapping parameter and the second fusion result when the current training stage is completed;

[0064] The model parameter update direction of the first aging feature mapping parameter when the current training phase is completed includes:

[0065] According to the deviation between the completion time of the current training phase and the completion time of the previous training phase when the first aging characteristic mapping parameter is completed, the model parameter update direction of the first aging characteristic mapping parameter when the current training phase is completed is obtained.

[0066] In an embodiment of the present invention, exemplarily, the server has multiple second glaze recognition models that process local images of the blue and white porcelain jar, such as the jar mouth, jar body, and jar bottom. For the second glaze recognition model that processes the local image of the jar body, the server first obtains the first orientation parameter group of the first aging feature mapping parameter when the previous training stage is completed, and the second orientation parameter group of the second feature mapping component when the current training stage is completed. For example, the first orientation parameter group contains parameters for analyzing the overall color change, and the second orientation parameter group contains parameters for analyzing the local cracking characteristics of the jar body. The server fuses these two groups of parameters according to certain rules, for example, by weighted average, to obtain the second fusion result corresponding to the current training stage. The server calculates the deviation between the second aging feature mapping parameter and the second fusion result when the current training stage is completed. If the current second aging feature mapping parameter indicates that the density of the local cracking of the jar body is a numerical value, and the cracking density corresponding to the second fusion result is another numerical value, the server obtains the deviation value by subtracting the two. Based on this deviation, the server determines the model parameter update direction of the second aging feature mapping parameter when the current training stage is completed. If the deviation is positive, it means that the current parameter value is large, and the update direction may be to reduce the parameter value; if the deviation is negative, the update direction may be to increase the parameter value. For the overall image of the blue-and-white porcelain jar processed by the first glaze recognition model, the server obtains the value of the first aging feature mapping parameter when the current training stage is completed and the previous training stage is completed. For example, the parameter for analyzing the overall color uniformity in the previous stage is one value, and the parameter becomes another value in the current stage. The server calculates the deviation between the two values ​​and obtains the deviation amount by simple numerical subtraction. Based on this deviation, the server determines the update direction of the model parameter of the first aging feature mapping parameter when the current training stage is completed. If the current value is larger than the value of the previous stage, the update direction may be to reduce the parameter; otherwise, increase the parameter. The server also performs the same operation on the second glaze recognition model that processes other local images of the blue-and-white porcelain jar, so as to accurately grasp the update direction of each model parameter and provide a basis for subsequent model training iterations, thereby improving the accuracy and reliability of the porcelain glaze aging degree identification model, and more accurately identifying the glaze aging degree of the Yuan Dynasty blue-and-white porcelain jar.

[0067] In the embodiment of the present invention, each image instance further includes a glaze aging degree target value of an aging degree grading result of each glaze area in the corresponding porcelain image; the aging degree grading result is determined based on at least two indicators of glaze crack density, color change rate, and microscopic bubble distribution characteristics;

[0068] For each second glaze recognition model, each training stage also includes the following implementation methods.

[0069] For the second feature extraction component of the second glaze recognition model, the first feature extraction component parameters of the first feature extraction component of the same image type when the previous training phase is completed are determined as the second feature extraction component parameters of the second feature extraction component after being updated in the current training phase;

[0070] The second glaze recognition model after updating each second feature extraction component parameter and the second aging feature mapping parameter is determined as the second glaze recognition model in the present training phase, so as to train the second glaze recognition model in the present training phase based on the image instance according to the second glaze recognition model until a second training termination state is satisfied, and obtain the second glaze recognition model when the present training phase is completed;

[0071] For the first glaze recognition model, each training stage is also implemented as follows.

[0072] For each first feature extraction component, the second feature extraction component parameters of all second feature extraction components of the same image type at the completion of this training phase are merged to obtain updated first feature extraction component parameters of the same image type of the first feature extraction component;

[0073] The first glaze recognition model after updating each first feature extraction component parameter and the first aging feature mapping parameter is determined as the first glaze recognition model in this training stage, and the first glaze recognition model in this training stage is trained according to the image instance of the first glaze recognition model until the first training termination state is met, so as to obtain the first glaze recognition model when this training stage is completed.

[0074] In an embodiment of the present invention, exemplarily, when training the porcelain glaze aging degree identification model, the server will iteratively train the first glaze recognition model and the second glaze recognition model according to the glaze aging degree target value in the image instance. Taking the identification of a Ming Dynasty blue and white porcelain bottle as an example, the image instance covers the overall image of the vase and the local images such as the bottle mouth, bottle body, and bottle bottom, and each image instance contains the glaze aging degree target value of the aging degree classification result of each glaze area, and these target values ​​are determined according to at least two indicators of the glaze crack density, color change rate, and microscopic bubble distribution characteristics. For the second glaze recognition model that processes the local image of the bottle mouth, when its second feature extraction component is updated in this training phase, the server directly determines the first feature extraction component parameters of the first feature extraction component of the same image type (local image) when the previous training phase is completed as the second feature extraction component parameters after the second feature extraction component is updated. For example, the parameters of the first feature extraction component when analyzing the overall bottle mouth shape characteristics will be applied to the second feature extraction component that processes the local image of the bottle mouth. After the server updates each second feature extraction component parameter and the second aging feature mapping parameter of the second glaze recognition model, it determines it as the second glaze recognition model in the process of this training stage. Then, the server trains the model of this stage according to the image instance of the model, that is, the local image of the bottle mouth, until the second training termination state is met, such as the prediction accuracy of the model reaches a certain threshold or the number of training times reaches a preset value, thereby obtaining the second glaze recognition model when this training stage is completed. The same process will also be applied to other second glaze recognition models that process local images such as the bottle body and bottle bottom. For the overall image of the blue and white porcelain bottle processed by the first glaze recognition model, the server will update each first feature extraction component in each training stage. The server merges the second feature extraction component parameters of all second feature extraction components of the same image type (overall image) when this training stage is completed. For example, the parameters of the second feature extraction component that processes the local images of the bottle mouth, bottle body, and bottle bottom at the completion of this stage on the color feature extraction are weighted averaged, etc., to obtain the updated first feature extraction component parameters of the first feature extraction component on the color feature extraction. After the server updates each first feature extraction component parameter and the first aging feature mapping parameter of the first glaze recognition model, it determines it as the first glaze recognition model in this training phase. Then, the server trains the model of this phase based on the image instance of the model, that is, the overall image of the blue and white porcelain vase, until the first training termination state is met, such as reaching the preset loss function value or the upper limit of the number of training rounds, and finally obtains the first glaze recognition model when this training phase is completed. Through such an interactive training method, the server continuously optimizes the first glaze recognition model and the second glaze recognition model to improve the accuracy of the model in identifying the degree of aging of the porcelain glaze.

[0075] In the embodiment of the present invention, the first feature mapping component and each second feature mapping component have the same feature mapping component structure, both including at least one aging texture analysis unit, and each second feature extraction component of the second glaze recognition model includes a feature extraction unit corresponding to the aging texture analysis unit of the second feature mapping component in the second glaze recognition model;

[0076] The training of the second glaze recognition model in the present training phase based on the image instance according to the second glaze recognition model until a second training termination state is satisfied may be implemented through the following examples.

[0077] For each aging texture analysis unit of the second glaze recognition model in the current training phase, integrating feature information output by a feature extraction unit corresponding to each second feature extraction component of the second glaze recognition model, to obtain a first integrated feature corresponding to the aging texture analysis unit;

[0078] For each aging texture analysis unit of the second glaze recognition model in the current training phase, according to each reference feature identifier corresponding to the aging texture analysis unit when the previous training phase is completed, gated fusion is used to optimize the first integrated feature corresponding to the aging texture analysis unit, and the first integrated features before and after the optimization of the aging texture analysis unit are integrated to obtain the second integrated feature corresponding to the aging texture analysis unit;

[0079] Determine the second integrated feature corresponding to the first aging texture analysis unit as the mapping result of the first aging texture analysis unit, and for non-first aging texture analysis units, perform feature mapping according to the second integrated feature corresponding to the non-first aging texture analysis unit and the mapping result of the previous aging texture analysis unit to obtain the mapping result of the non-first aging texture analysis unit;

[0080] According to the mapping result of the last aging texture analysis unit, the predicted aging degree classification result of each glaze area in the corresponding porcelain image of the image instance is obtained, and according to the deviation between the predicted aging degree classification result of each glaze area and the glaze aging degree target value, each second feature extraction component parameter and second aging feature mapping parameter of the second glaze recognition model in the current training stage is optimized;

[0081] Each reference feature identifier corresponding to each aging texture analysis unit is obtained by feature grouping the mapping features output by the corresponding aging texture analysis unit for the glazed area of ​​the same aging degree grading result in each first image instance.

[0082] In an embodiment of the present invention, exemplarily, when the server trains the second glaze recognition model, it is necessary to train the model in the present training phase according to the image example until the second training termination state is satisfied. Taking the identification of a secret color porcelain bowl of the Tang Dynasty as an example, the first feature mapping component and the second feature mapping component have the same structure, both have aging texture analysis units, and the second feature extraction component has a corresponding feature extraction unit. The server starts to integrate feature information for each aging texture analysis unit of the second glaze recognition model in the current training phase. For example, the second glaze recognition model that processes the local image of the bowl mouth has three aging texture analysis units, which analyze the crack texture, color texture and bubble distribution texture respectively. There are corresponding feature extraction units in its second feature extraction component, which extract the feature information of the bowl mouth crack, color and bubble distribution respectively. The server integrates the feature information output by each feature extraction unit corresponding to the same aging texture analysis unit. For the aging texture analysis unit that analyzes the crack texture, the server integrates the information output by the bowl mouth crack feature extraction unit shot at different angles to obtain the first integrated feature corresponding to the aging texture analysis unit. The server optimizes the first integrated feature of each aging texture analysis unit. Each aging texture analysis unit has a corresponding reference feature identifier when the previous training phase is completed. These identifiers are obtained by feature grouping the mapping features output by the glazed area of ​​the corresponding aging texture analysis unit in the first image instance with the same aging degree grading result. For example, for the aging texture analysis unit that analyzes color texture, the server uses gated fusion to optimize the first integrated feature of the unit according to the reference feature identifier obtained in the previous phase. Gated fusion is like an intelligent switch that selectively enhances or weakens certain parts of the first integrated feature according to the reference feature identifier. After optimization, the server integrates the first integrated features before and after optimization again to obtain the second integrated feature corresponding to the aging texture analysis unit. The server determines the mapping result of the first aging texture analysis unit. Taking the aging texture analysis unit that analyzes crackle texture as an example, its corresponding second integrated feature is directly determined as the mapping result of the unit. For non-first aging texture analysis units, such as units that analyze color texture, the server performs feature mapping based on its corresponding second integrated feature and the mapping result of the previous aging texture analysis unit (cracking texture analysis unit). Through a specific mapping function, the features of the two are associated and converted to obtain the mapping result of the non-first aging texture analysis unit. In this way, the feature mapping of all aging texture analysis units is completed. Based on the mapping result of the last aging texture analysis unit, the server obtains the predicted aging degree classification results of each glaze area in the image instance (partial image of the bowl mouth). The server compares these prediction results with the preset glaze aging degree target value in the image instance and calculates the deviation.For example, if a certain area of ​​the bowl mouth is predicted to be slightly aged, while the target value is shown as moderately aged, the server optimizes each second feature extraction component parameter and second aging feature mapping parameter of the second glaze recognition model in the current training phase based on this deviation. By adjusting the parameters, the model can more accurately predict the degree of aging in subsequent training. The server repeats the above process to train the second glaze recognition model that processes other local images such as the bowl body and bottom of the bowl until the second training termination state is met, such as the deviation between the prediction result and the target value is within an acceptable range, or the preset number of training rounds is reached, thereby improving the second glaze recognition model's accuracy in identifying the degree of aging of the local glaze of the Tang Dynasty secret color porcelain bowl.

[0083] In the embodiment of the present invention, according to the reference feature identification when the previous training stage is completed, the first integrated feature corresponding to the aging texture analysis unit is optimized by gated fusion, which can be implemented through the following examples.

[0084] According to the reference feature identifier at the completion of the previous training stage, the first integrated feature corresponding to the aging texture analysis unit, and the number of texture analysis components in the aging texture analysis unit, gated fusion is used to perform a standardization operation to obtain a fourth feature contribution;

[0085] The first integrated feature corresponding to the aging texture analysis unit is weighted by the fourth feature contribution to obtain an optimized first integrated feature.

[0086] In an embodiment of the present invention, exemplarily, when training the second glaze recognition model, the server will optimize the first integrated feature corresponding to the aging texture analysis unit. Taking the identification of a Song Dynasty official kiln porcelain plate as an example, the server optimizes the first integrated feature of each aging texture analysis unit in the second glaze recognition model that processes the local image of the porcelain plate. The server starts to optimize a certain aging texture analysis unit of the second glaze recognition model in the current training phase, such as an aging texture analysis unit that analyzes the local crack texture of the porcelain plate. The unit has reference feature identifiers when the previous training phase is completed. These identifiers are obtained by feature grouping the mapping features output by the corresponding aging texture analysis unit in the glaze area of ​​the same aging degree grading result in the first image instance. At the same time, the server has obtained the first integrated feature corresponding to the aging texture analysis unit and knows the number of texture analysis components in the unit. The server uses gated fusion to perform standardization operations to obtain the fourth feature contribution. Gated fusion is like an intelligent filter that calculates based on the reference feature identifier, the first integrated feature, and the number of texture analysis components. Assume that the aging texture analysis unit for analyzing crack texture has three texture analysis components, the reference feature identification of the previous stage indicates the correlation between a certain crack pattern and the degree of aging, and the first integrated feature contains various feature information of the current local crack of the porcelain plate. The server integrates and standardizes this information through a specific algorithm to obtain the fourth feature contribution. This contribution reflects the importance of each feature in the aging texture analysis unit in the current optimization process. The server uses the obtained fourth feature contribution to weight the first integrated feature corresponding to the aging texture analysis unit. If the fourth feature contribution shows that a certain crack feature is more important in identifying the degree of aging, the server will increase the weight of the feature in the first integrated feature; otherwise, it will reduce the weight. For example, if the fourth feature contribution shows that the density feature of the crack has a greater impact on the identification of the degree of aging, the server will increase the proportion of the feature in the first integrated feature. Through such a weighted operation, the server obtains the optimized first integrated feature. The server also performs the same operation on the aging texture analysis unit that processes other texture features of the porcelain plate (such as color texture and bubble distribution texture). In this way, the server continuously optimizes the features of each aging texture analysis unit in the second glaze recognition model, so that the model can more accurately identify the degree of glaze aging based on the local image of the porcelain plate, thereby improving the performance and accuracy of the entire identification model.

[0087] In the embodiment of the present invention, the reference feature identifiers of each aging texture analysis unit are updated in the following manner:

[0088] For each first image instance, according to the glaze aging degree target value of the first image instance, determine the aging degree classification result corresponding to the mapping feature of each glaze area output by the aging texture analysis unit when the previous training stage is completed;

[0089] For each first image instance, integrating the mapping features of each glazed area of ​​the same aging degree grading result output by the aging texture analysis unit, to obtain the recognition processing features of the corresponding aging degree grading result of the first image instance by the aging texture analysis unit;

[0090] For the same aging degree grading result, the recognition processing features of the aging degree grading result of each first image instance by the aging texture analysis unit are grouped according to a preset number of categories, and the features of the cluster center of each feature group in the current training stage are determined;

[0091] For each feature group, the features of the cluster centers of the feature group in the current training stage and the previous training stage are integrated to obtain a reference feature identifier corresponding to the feature group when the aging texture analysis unit completes the current training stage.

[0092] In an embodiment of the present invention, exemplarily, when the server trains the porcelain glaze aging degree identification model, it will update the reference feature identification of each aging texture analysis unit. Taking the identification of a batch of ancient porcelain as an example, the first image instances of these porcelains contain the whole image, and each instance has a corresponding glaze aging degree target value. For each first image instance, the server analyzes the mapping features of each glazed area output by the aging texture analysis unit when the previous training stage is completed according to its glaze aging degree target value. For example, for the first image instance of a Song Dynasty Ru kiln porcelain plate, its glaze aging degree target value shows that some areas are slightly aged and some are moderately aged. The server checks the mapping features output by the aging texture analysis unit of the porcelain plate in the previous stage, determines the aging degree classification result corresponding to the mapping features of each glazed area, and clarifies which mapping features correspond to the slightly aged area and which correspond to the moderately aged area. For each first image instance, the server integrates the mapping features of each glazed area of ​​the same aging degree classification result output by the aging texture analysis unit. Still taking the Song Dynasty Ru kiln porcelain plate as an example, the server integrates the mapping features of all glaze areas corresponding to mild aging to obtain the recognition processing features of the aging texture analysis unit for the mild aging degree classification result of the porcelain plate. Similarly, the same operation is performed on the medium aging area to obtain the corresponding recognition processing features. For the same aging degree classification result, the server performs feature grouping according to the preset number of preset categories for the recognition processing features of the aging degree classification result of each first image instance by the aging texture analysis unit. Assuming that the preset number of categories is 3, for the recognition processing features corresponding to mild aging in all first image instances, the server uses a clustering algorithm to divide them into 3 groups. Then determine the features of the cluster center of each group in the current training stage, and the cluster center features represent the typical situation of the group of features. For each feature group, the server integrates the features of the cluster center of the feature group in the current training stage and the previous training stage. For example, the cluster center feature of a feature group in the current stage is A, and the cluster center feature of the previous stage is B, and the server integrates A and B. Through such an integration operation, the server obtains a reference feature identifier corresponding to the feature group when the aging texture analysis unit completes the current training stage. The server performs this operation on all feature groups to complete the update of the reference feature identifiers of each aging texture analysis unit. By updating the reference feature identifiers, the server enables the aging texture analysis unit to more accurately optimize the first integrated feature based on these identifiers in subsequent training, thereby improving the performance and accuracy of the porcelain glaze aging degree identification model.

[0093] In an embodiment of the present invention, the method is applied in a distributed collaborative training system, the first glaze recognition model is deployed in a global aggregator, and a second glaze recognition model is deployed in a local training node, and the method is executed by the global aggregator, wherein the second aging feature mapping parameters of each second glaze recognition model when the training is completed in the current training stage are obtained, which can be implemented through the following example.

[0094] Receiving model update parameters of each second glaze recognition model sent by each local training node, wherein the model update parameters include second aging feature mapping parameters of the second glaze recognition model when training is completed in the current training stage;

[0095] For each second glaze recognition model, after obtaining the second aging feature mapping parameters corresponding to the next training stage, the method further includes:

[0096] The model update parameters corresponding to the next training stage of the second glaze recognition model are transmitted to the corresponding local training node, so that the local training node performs a training iteration on the second glaze recognition model based on the model update parameters and the corresponding second image instance, and obtains the model update parameters of the second glaze recognition model when the training is completed in the next stage.

[0097] In an embodiment of the present invention, exemplarily, in a distributed collaborative training system, the server, as a global aggregator, executes a training method for a porcelain glaze aging degree identification model. Taking the identification of a batch of ancient porcelain as an example, the first glaze recognition model is deployed in the global aggregator, and multiple second glaze recognition models are respectively deployed in different local training nodes. The global aggregator waits for each local training node to complete the current training phase. Each local training node is responsible for the training of a second glaze recognition model. For example, there are three local training nodes, which process the local images of the bottle mouth, bottle body and bottle bottom of the porcelain respectively. When the training is completed, each local training node will send the model update parameters of its own second glaze recognition model to the global aggregator. These model update parameters include the second aging feature mapping parameters of the second glaze recognition model when the training is completed in the current training phase. For example, the local training node that processes the local image of the bottle mouth will send the mapping parameters for analyzing the aging features such as cracking and color of the bottle mouth glaze. The global aggregator receives these model update parameters from different local training nodes, thereby obtaining the second aging feature mapping parameters of each second glaze recognition model. The global aggregator performs parameter fusion and other operations on each second glaze recognition model to obtain the second aging feature mapping parameters corresponding to the next training stage. For the second glaze recognition model that processes the local image of the bottle mouth, the global aggregator transmits the model update parameters containing the new second aging feature mapping parameters back to the corresponding local training node. After receiving the parameters, the local training node will perform a training iteration on the second glaze recognition model based on this and the corresponding second image instance (local image of the bottle mouth). During the iteration process, the model will relearn and analyze the glaze aging features in the local image of the bottle mouth according to the new parameters and adjust its own parameters. After the training is completed, the local training node obtains the model update parameters of the second glaze recognition model when the training is completed in the next stage. Similarly, the global aggregator will perform the same operation on the second glaze recognition model that processes the local images of the bottle body, bottle bottom, etc., and transmit the updated parameters to the corresponding local training nodes, allowing them to continue training iterations. Through this distributed collaborative training method, the global aggregator and each local training node cooperate with each other to continuously optimize the second glaze recognition model and improve the accuracy of the identification of the aging degree of the porcelain glaze.

[0098] The embodiment of the present invention provides a computer device 100, which includes a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device 100 executes the aforementioned intelligent identification method for the aging degree of porcelain glaze. Figure 2 As shown, Figure 2The block diagram of the computer device 100 provided in the embodiment of the present invention. The computer device 100 includes a memory 111, a processor 112 and a communication unit 113. To achieve data transmission or interaction, the memory 111, the processor 112 and the communication unit 113 are directly or indirectly electrically connected to each other. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0099] For illustrative purposes, the foregoing description is made with reference to specific embodiments. However, the above illustrative discussion is not intended to be exhaustive or to limit the present disclosure to the precise form disclosed. Numerous modifications and variations are possible in accordance with the above teachings. These embodiments are selected and described in order to best illustrate the principles of the present disclosure and its practical application, so that those skilled in the art can best utilize the present disclosure and utilize various embodiments with different modifications to suit the intended specific application.

Claims

1. An intelligent identification method for the aging degree of porcelain glaze, characterized in that: include: The reflectance characteristics of the target porcelain in the visible light and near-infrared bands are collected by a multispectral imaging device, and the glaze reflection suppression process is performed on the collected original image to generate a standardized porcelain image to be identified; the porcelain image to be identified is a porcelain image of a preset image type of the target porcelain, and the preset image type is an overall image or a partial image; Loading the porcelain image to be identified into a pre-trained target glaze aging degree identification model, and obtaining a glaze aging degree target value of the aging degree classification result of each glaze area in the porcelain image output by the target glaze aging degree identification model; Determining a target aging degree grading result of the target porcelain according to a target aging degree grading result of each glaze area in the porcelain image; The target aging degree classification results are temporally and spatially aligned and fused with the trace element data detected by X-ray fluorescence spectroscopy to generate a porcelain glaze aging degree identification report that includes the age inference results and glaze preservation status assessment.

2. The method according to claim 1, characterized in that The target glaze aging degree identification model is trained by the following methods, including: Acquire an initial glaze aging degree identification model, wherein the initial glaze aging degree identification model includes a first glaze recognition model and at least one second glaze recognition model, wherein each first image instance of the first glaze recognition model includes a porcelain image of an overall image of a target porcelain, and each second image instance of the second glaze recognition model includes a porcelain image of a partial image of the target porcelain; Acquire a first aging feature mapping parameter of the first glaze recognition model and a second aging feature mapping parameter of each of the second glaze recognition models when training is completed in the current training phase; For each second glaze recognition model, a first orientation parameter group for fusion in the first aging feature mapping parameter and a second orientation parameter group for fusion in the second aging feature mapping parameter are determined according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed; wherein the first parameter update direction is the model parameter update direction of the first aging feature mapping parameter of the first feature mapping component, and the second parameter update direction is the model parameter update direction of the second aging feature mapping parameter of the second feature mapping component; For each second glaze recognition model, the first orientation parameter group when the training is completed in the current training stage is merged with the second orientation parameter group of the second glaze recognition model to obtain the second aging feature mapping parameters corresponding to the training in the next training stage, and based on the obtained second aging feature mapping parameters and each second image instance of the second glaze recognition model, the second glaze recognition model is trained once to obtain the second aging feature mapping parameters when the training is completed in the next training stage; The second orientation parameter group of each second glaze recognition model when training is completed in the next training stage and the first orientation parameter group of the first glaze recognition model when training is completed in the current training stage are merged to obtain the first aging feature mapping parameters corresponding to the training in the next training stage, and the first glaze recognition model is trained once based on the obtained first aging feature mapping parameters and each first image instance to obtain the first aging feature mapping parameters when training is completed in the next training stage; The first glaze recognition model at the completion of the last training phase is determined as the first glaze aging degree identification model, and the second glaze recognition model at the completion of the last training phase is determined as the second glaze aging degree identification model; The glaze aging degree identification model including the feature extraction components corresponding to each image type in the preset image types in the trained first glaze aging degree identification model or the second glaze recognition model is determined as the target glaze aging degree identification model.

3. The method according to claim 2, characterized in that The component architecture of the feature mapping components of each glaze recognition model in the initial glaze aging degree identification model is consistent, and the feature mapping components of each glaze recognition model include a first number of texture analysis components; For each second glaze recognition model, the method of determining a first directional parameter group for fusion in the first aging feature mapping parameter and a second directional parameter group for fusion in the second aging feature mapping parameter of the second glaze recognition model according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed, comprises: Determine, according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed, a second texture analysis component for fusion among the texture analysis components of the second feature mapping component of the second glaze recognition model, and use the parameters of each second texture analysis component as a second directional parameter group for fusion in the second aging feature mapping parameters of the second glaze recognition model; According to each of the second texture analysis components, a first texture analysis component for fusion among each of the texture analysis components of the first feature mapping component is determined, and the parameters of each of the first texture analysis components are used as the first orientation parameter group for fusion in the first aging feature mapping parameters, wherein the first texture analysis component is a texture analysis component other than the texture analysis components corresponding to each of the second texture analysis components among the texture analysis components of the first feature mapping component.

4. The method according to claim 2, characterized in that: The second orientation parameter group of each second glaze recognition model when training is completed in the next training stage and the first orientation parameter group of the first glaze recognition model when training is completed in the current training stage are fused to obtain the first aging feature mapping parameter corresponding to the training in the next training stage, including: Acquire a first feature contribution corresponding to the first glaze surface recognition model, and a second feature contribution corresponding to each of the second glaze surface recognition models; For the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier in the second orientation parameter group of each second feature mapping component, obtain the model parameter update direction of each second aging feature mapping parameter when the current training phase is completed; Obtaining a third feature contribution of each second feature mapping component according to a model parameter update direction of each second aging feature mapping parameter when the current training phase is completed, wherein the third feature contribution is negatively correlated with a parameter update direction of the corresponding second aging feature mapping parameter; For the target second orientation parameter group of the texture analysis component corresponding to the same feature level identifier in the second orientation parameter group of each second feature mapping component, a linear superposition process is performed with the third feature contribution of the corresponding second feature mapping component to obtain the first integration result corresponding to the texture analysis component with the same feature level identifier in the next stage; determining a mean value of the first integrated result of the texture analysis component identified by each feature level as a third orientation parameter group; Based on the first feature contribution and the second feature contribution, the first orientation parameter group and the third orientation parameter group of the first glaze recognition model when the training is completed in the current training stage are weighted fused to obtain the first aging feature mapping parameters corresponding to the next training stage.

5. The method according to claim 2, characterized in that: The directional consistency between the first parameter update direction and the second parameter update direction includes a consistency coefficient between the first sub-parameter update direction and the second sub-parameter update direction of each corresponding texture analysis component in the first glaze recognition model and the second glaze recognition model; For each second glaze recognition model, according to the directional consistency between the first parameter update direction and the second parameter update direction when each training stage in the latest target training stage is completed, a first directional parameter group for fusion in the first aging feature mapping parameter and a second directional parameter group for fusion in the second aging feature mapping parameter are determined, including: For each texture analysis component in each second glaze recognition model, if the consistency coefficient of the texture analysis component at the completion of each training stage in the latest target training stage does not exceed the preset consistency threshold, the texture analysis component is determined as a target texture analysis component; In the feature selection matrix of the second glaze recognition model at the completion of the current training phase, the parameters corresponding to the target texture analysis component are unassigned for fusion, and the feature selection matrix corresponding to the training in the next training phase is obtained; wherein the feature selection matrix is ​​used to indicate the second directional parameter group used for fusion in the second aging feature mapping parameters of the second glaze recognition model, and the feature selection matrix corresponding to the training in the first training phase is a pre-set feature selection matrix; According to the feature selection matrix corresponding to the training in the next training stage, a first orientation parameter group for fusion in the first aging feature mapping parameters and a second orientation parameter group for fusion in the second aging feature mapping parameters of the second glaze recognition model are determined, wherein the first orientation parameter group is an array of the first aging feature mapping parameters excluding the parameters corresponding to the second orientation parameter group.

6. The method according to claim 2, characterized in that The model parameter update direction of the second aging feature mapping parameter when the current training phase is completed includes: Fusing a first orientation parameter group of the first aging feature mapping parameter when the previous training stage is completed with a second orientation parameter group of the second feature mapping component when the current training stage is completed, to obtain a second fusion result corresponding to the current training stage; Obtaining a model parameter update direction of the second aging feature mapping parameter when the current training stage is completed according to a deviation between the second aging feature mapping parameter and the second fusion result when the current training stage is completed; The model parameter update direction of the first aging feature mapping parameter when the current training phase is completed includes: According to the deviation between the completion time of the current training phase and the completion time of the previous training phase when the first aging characteristic mapping parameter is completed, the model parameter update direction of the first aging characteristic mapping parameter when the current training phase is completed is obtained.

7. The method according to claim 2, characterized in that Each image instance also includes a glaze aging degree target value of an aging degree grading result of each glaze area in the corresponding porcelain image; the aging degree grading result is determined based on at least two indicators of glaze crack density, color change rate, and microscopic bubble distribution characteristics; the first feature mapping component and each second feature mapping component have the same feature mapping component structure, both including at least one aging texture analysis unit, and each second feature extraction component of the second glaze recognition model includes a feature extraction unit corresponding to the aging texture analysis unit of the second feature mapping component in the second glaze recognition model; For each second glaze recognition model, each training stage also includes: For the second feature extraction component of the second glaze recognition model, the first feature extraction component parameters of the first feature extraction component of the same image type when the previous training phase is completed are determined as the second feature extraction component parameters of the second feature extraction component after being updated in the current training phase; Determining the second glaze recognition model after updating each second feature extraction component parameter and the second aging feature mapping parameter as the second glaze recognition model in the present training phase; For each aging texture analysis unit of the second glaze recognition model in the current training phase, integrating feature information output by a feature extraction unit corresponding to each second feature extraction component of the second glaze recognition model, to obtain a first integrated feature corresponding to the aging texture analysis unit; For each aging texture analysis unit of the second glaze recognition model in the current training phase, a standardization operation is performed using gated fusion according to the reference feature identifier at the completion of the previous training phase, the first integrated feature corresponding to the aging texture analysis unit, and the number of texture analysis components in the aging texture analysis unit to obtain a fourth feature contribution; The first integrated feature corresponding to the aging texture analysis unit is weighted by the fourth feature contribution to obtain an optimized first integrated feature, and the first integrated feature before and after the optimization of the aging texture analysis unit is integrated to obtain a second integrated feature corresponding to the aging texture analysis unit; Determine the second integrated feature corresponding to the first aging texture analysis unit as the mapping result of the first aging texture analysis unit, and for non-first aging texture analysis units, perform feature mapping according to the second integrated feature corresponding to the non-first aging texture analysis unit and the mapping result of the previous aging texture analysis unit to obtain the mapping result of the non-first aging texture analysis unit; According to the mapping result of the last aging texture analysis unit, the predicted aging degree grading result of each glaze area in the corresponding porcelain image of the image instance is obtained, and according to the deviation between the predicted aging degree grading result of each glaze area and the glaze aging degree target value, each second feature extraction component parameter and the second aging feature mapping parameter of the second glaze recognition model in the current training stage are optimized to obtain the second glaze recognition model when the training stage is completed; wherein each reference feature identifier corresponding to each aging texture analysis unit is obtained by feature grouping the mapping features output by the corresponding aging texture analysis unit for the glaze area of ​​the same aging degree grading result in each first image instance; For the first glaze recognition model, each training stage further includes: For each first feature extraction component, the second feature extraction component parameters of all second feature extraction components of the same image type at the completion of this training phase are merged to obtain updated first feature extraction component parameters of the same image type of the first feature extraction component; The first glaze recognition model after updating each first feature extraction component parameter and the first aging feature mapping parameter is determined as the first glaze recognition model in this training stage, and the first glaze recognition model in this training stage is trained according to the image instance of the first glaze recognition model until the first training termination state is met, so as to obtain the first glaze recognition model when this training stage is completed.

8. The method according to claim 7, characterized in that The reference feature identifiers of each aging texture analysis unit are updated in the following way: For each first image instance, according to the glaze aging degree target value of the first image instance, determine the aging degree classification result corresponding to the mapping feature of each glaze area output by the aging texture analysis unit when the previous training stage is completed; For each first image instance, integrating the mapping features of each glazed area of ​​the same aging degree grading result output by the aging texture analysis unit, to obtain the recognition processing features of the corresponding aging degree grading result of the first image instance by the aging texture analysis unit; For the same aging degree grading result, the recognition processing features of the aging degree grading result of each first image instance by the aging texture analysis unit are grouped according to a preset number of categories, and the features of the cluster center of each feature group in the current training stage are determined; For each feature group, the features of the cluster centers of the feature group in the current training stage and the previous training stage are integrated to obtain a reference feature identifier corresponding to the feature group when the aging texture analysis unit completes the current training stage.

9. The method according to claim 2, characterized in that: The method is applied in a distributed collaborative training system, wherein the first glaze recognition model is deployed in a global aggregator, and one of the second glaze recognition models is deployed in a local training node, and the method is executed by the global aggregator, wherein obtaining the second aging feature mapping parameters of each of the second glaze recognition models when training is completed in the current training phase includes: Receiving model update parameters of each second glaze recognition model sent by each local training node, wherein the model update parameters include second aging feature mapping parameters of the second glaze recognition model when training is completed in the current training stage; For each second glaze recognition model, after obtaining the second aging feature mapping parameters corresponding to the next training stage, the method further includes: The model update parameters corresponding to the next training stage of the second glaze recognition model are transmitted to the corresponding local training node, so that the local training node performs a training iteration on the second glaze recognition model based on the model update parameters and the corresponding second image instance, and obtains the model update parameters of the second glaze recognition model when the training is completed in the next stage.

10. A server system, characterized in that: The method comprises a server, wherein the server is used to execute the method described in any one of claims 1 to 9.

Images