Porcelain identification system and method based on artificial intelligence

Abstract

The invention relates to the technical field of cultural relic identification, in particular to a porcelain identification system and method based on artificial intelligence, and the system comprises a short-wave infrared camera module which is used for collecting an infrared transmission image of porcelain in a wave band of 760-1mm; the visible light shooting module is used for acquiring a porcelain surface visible light image; the image synchronization control module ensures that the short infrared shooting module and the visible light shooting module synchronously take images at the same angle and position; and the AI image processing module is used for analyzing the glaze layer reflection characteristic, the carcass material reaction characteristic and the aging structure difference of the double-spectrum image based on a deep learning architecture, comparing with the annotation database, and outputting the kiln mouth, the age and the authenticity judgment result of the porcelain. According to the porcelain recognition system and method based on artificial intelligence, the aging characteristics, the kiln mouth and the age of porcelain are presumed through AI image analysis, and the interpretation efficiency and accuracy of the porcelain are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cultural relics identification, and in particular to a porcelain identification system and method based on artificial intelligence. Background Art

[0002] Traditional porcelain dating and authenticity authentication relies heavily on expert experience, judging by the appearance, style, decorative patterns, and inscriptions on the base. While existing image recognition technologies are increasingly being applied to cultural relic analysis, they currently focus on visible light imaging or microscopic features, with limited ability to detect underglaze structures, cosmetic layers, and signs of porcelain aging. This is particularly true when porcelain is repaired, re-fired glazes are used to imitate, or new porcelain with an old base is used as a base. Traditional detection methods are often affected by interference, making identification difficult. Summary of the Invention

[0003] In order to overcome the shortcomings and deficiencies in the prior art, the purpose of the present invention is to provide an artificial intelligence-based porcelain identification system and method, which improves the efficiency and accuracy of porcelain interpretation by inferring the aging characteristics, kilns and age of porcelain through AI image analysis.

[0004] The present invention is achieved through the following technical solutions:

[0005] In a first aspect, the present invention discloses a porcelain identification system based on artificial intelligence, which comprises:

[0006] Short-wave infrared camera module, used to collect infrared transmission images of porcelain in the 760nm to 1mm band;

[0007] Visible light photography module, used to obtain visible light images of the porcelain surface;

[0008] Image synchronization control module ensures that the short infrared camera module and the visible light camera module capture images synchronously at the same angle and position;

[0009] The AI ​​image processing module, based on a deep learning architecture, analyzes the glaze reflection characteristics, body material reaction characteristics, and aging structure differences of the dual-spectral image, and compares them with the annotation database to output the kiln, age, and authenticity determination results of the porcelain.

[0010] In combination with the first aspect, further, the short infrared camera module is provided with an InGaAs sensor, and the wavelength coverage range of the InGaAs sensor includes a 900-1700 nanometer band;

[0011] The visible light photography module is provided with a CCD sensor, and the CCD sensor is used for multi-angle image acquisition in the visible light band of 400-700 nanometers.

[0012] In combination with the first aspect, further, the training data set of the AI ​​image processing module includes the following annotation parameters:

[0013] distribution pattern of underglaze cracks;

[0014] Differences in infrared absorption spectra between the cosmetic soil layer and the carcass;

[0015] The decorative style characteristics of historical kiln samples;

[0016] Structured recognition template for official inscription text.

[0017] In combination with the first aspect, further comprising:

[0018] The multi-spectral UV detection unit integrates a 365nm long-wave UV lamp and a 254nm short-wave UV lamp to identify abnormal reactions of fluorescent agents and use the detection results as exclusion conditions for analysis by the AI ​​image processing module.

[0019] In combination with the first aspect, further, the short-wave infrared camera module and the visible light photography module are integrated into an adjustable slide rail bracket, and the bracket is equipped with a laser positioner and an angle encoder;

[0020] The slide rail bracket is provided with a standard light source box, which is used to provide a uniform lighting environment with a color temperature of 2000±50K;

[0021] The device has a built-in image alignment algorithm that automatically corrects the pixel-level spatial matching of bispectral images.

[0022] In a second aspect, the present invention discloses a porcelain identification method based on artificial intelligence, which comprises the following steps:

[0023] Through the image synchronization control module, the short-wave infrared image and visible light image of the porcelain are obtained in the same spatial coordinate system.

[0024] Extract glaze penetration features from shortwave infrared images, including body joint marks, internal repair areas, and material aging gradients;

[0025] Perform style matching analysis between the glaze color and patterns of visible light images and the historical kiln feature database;

[0026] By integrating dual-spectral feature data through the AI ​​image processing module, an identification report is generated that includes the probability of age, kiln ownership and authenticity level.

[0027] Beneficial effects of the present invention:

[0028] This invention utilizes dual-spectral integration of shortwave infrared and visible light, combined with AI image processing and analysis, to effectively reveal detailed aging information previously difficult to discern with the naked eye, such as underglaze cracks, carcass inhomogeneities, and traces of joints or repairs, thereby complementing the shortcomings of traditional authentication methods. Its establishment of a multispectral database and AI interpretation model enhances the objectivity and reproducibility of authentication, while also increasing the speed and accuracy of identification. Even for porcelain with minimal plain decoration, this system can reliably interpret it through microstructural comparison, providing practical and forward-looking technical support for porcelain authenticity and age verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0030] Figure 1 This is a module block diagram of the porcelain identification system of the present invention.

[0031] Figure 2 It is a structural schematic diagram of the short-wave infrared camera module and the visible light photography module of the present invention.

[0032] Figure 3 It is a flow chart of the porcelain identification method of the present invention.

[0033] Reference numerals

[0034] Shortwave infrared camera module--100, visible light photography module--200, image synchronization control module--300, AI image processing module--400. DETAILED DESCRIPTION

[0035] To make the above-mentioned objects, features, and advantages of the present invention more readily apparent, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. The following description sets forth numerous specific details to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art may make similar modifications without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0036] Example 1

[0037] like Figure 1 and Figure 2 As shown, this embodiment discloses an artificial intelligence-based porcelain recognition system, which includes:

[0038] Short-wave infrared camera module, used to collect infrared transmission images of porcelain in the 760nm to 1mm band;

[0039] Visible light photography module, used to obtain visible light images of the porcelain surface;

[0040] Image synchronization control module ensures that the short infrared camera module and the visible light camera module capture images synchronously at the same angle and position;

[0041] The AI ​​image processing module, based on a deep learning architecture, analyzes the glaze reflection characteristics, body material reaction characteristics, and aging structure differences of the dual-spectral image, and compares them with the annotation database to output the kiln, age, and authenticity determination results of the porcelain.

[0042] To ensure data quality, the system's operating manual specifies that shortwave infrared image capture must fully capture the interface between the porcelain surface and the base to avoid missing subtle aging features. The operating manual requires that image capture include the contact surface between the porcelain surface and the base to prevent forgeries where new porcelain meets old base. For plain porcelain, detailed decorative details such as engravings, prints, or decals should be preserved during imaging to enhance AI comparison accuracy. All images must be captured repeatedly at a fixed angle and distance to serve as the basis for comparison and archiving.

[0043] Furthermore, the short infrared camera module is provided with an InGaAs sensor, and the wavelength coverage range of the InGaAs sensor includes the 900-1700 nanometer band;

[0044] The visible light photography module is provided with a CCD sensor, and the CCD sensor is used for multi-angle image acquisition in the visible light band of 400-700 nanometers.

[0045] Furthermore, the training data set of the AI ​​image processing module includes the following annotation parameters:

[0046] distribution pattern of underglaze cracks;

[0047] Differences in infrared absorption spectra between the cosmetic soil layer and the carcass;

[0048] The decorative style characteristics of historical kiln samples;

[0049] Structured recognition template for official inscription text.

[0050] Furthermore, the porcelain identification system of this embodiment also includes:

[0051] The multi-spectral UV detection unit integrates a 365nm long-wave UV lamp and a 254nm short-wave UV lamp to identify abnormal reactions of fluorescent agents and use the detection results as exclusion conditions for analysis by the AI ​​image processing module.

[0052] In combination with the first aspect, further, the short-wave infrared camera module and the visible light photography module are integrated into an adjustable slide rail bracket, and the bracket is equipped with a laser positioner and an angle encoder;

[0053] The slide rail bracket is provided with a standard light source box, which is used to provide a uniform lighting environment with a color temperature of 2000±50K;

[0054] The device has a built-in image alignment algorithm that automatically corrects the pixel-level spatial matching of bispectral images.

[0055] Example 2

[0056] like Figure 3 As shown, this embodiment discloses a porcelain identification method based on artificial intelligence, which includes the following steps:

[0057] Through the image synchronization control module, the short-wave infrared image and visible light image of the porcelain are obtained in the same spatial coordinate system.

[0058] Extract glaze penetration features from shortwave infrared images, including body joint marks, internal repair areas, and material aging gradients;

[0059] Perform style matching analysis between the glaze color and patterns of visible light images and the historical kiln feature database;

[0060] By integrating dual-spectral feature data through the AI ​​image processing module, an identification report is generated that includes the probability of age, kiln ownership and authenticity level.

[0061] In summary, the present invention, through dual-spectral integration of shortwave infrared and visible light, combined with AI image processing and analysis, can effectively reveal detailed aging information previously difficult to discern with the naked eye, such as underglaze cracks, carcass inhomogeneities, and traces of joints or repairs, thereby complementing the shortcomings of traditional identification methods. Its establishment of a multispectral database and AI interpretation model can enhance the objectivity and reproducibility of identification, and improve the speed and accuracy of identification. Even for porcelain with minimal plain decoration, this system can reliably interpret it through microstructure comparison, providing practical and forward-looking technical support for the authentication of porcelain authenticity and age.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A porcelain identification system based on artificial intelligence, characterized in that: include: Short-wave infrared camera module, used to collect infrared transmission images of porcelain in the 760nm to 1mm band; Visible light photography module, used to obtain visible light images of the porcelain surface; Image synchronization control module ensures that the short infrared camera module and the visible light camera module capture images synchronously at the same angle and position; The AI ​​image processing module, based on a deep learning architecture, analyzes the glaze reflection characteristics, body material reaction characteristics, and aging structure differences of the dual-spectral image, and compares them with the annotation database to output the kiln, age, and authenticity determination results of the porcelain.

2. The artificial intelligence-based porcelain identification system according to claim 1, characterized in that: The short infrared camera module is provided with an InGaAs sensor, and the wavelength coverage range of the InGaAs sensor includes the 900-1700 nanometer band; The visible light photography module is provided with a CCD sensor, and the CCD sensor is used for multi-angle image acquisition in the visible light band of 400-700 nanometers.

3. The artificial intelligence-based porcelain identification system according to claim 1, characterized in that: The training data set of the AI ​​image processing module includes the following annotation parameters: distribution pattern of underglaze cracks; Differences in infrared absorption spectra between the cosmetic soil layer and the carcass; The decorative style characteristics of historical kiln samples; Structured recognition template for official inscription text.

4. The artificial intelligence-based porcelain identification system according to claim 1, characterized in that: Also includes: The multi-spectral UV detection unit integrates a 365nm long-wave UV lamp and a 254nm short-wave UV lamp to identify abnormal reactions of fluorescent agents and use the detection results as exclusion conditions for analysis by the AI ​​image processing module.

5. The porcelain identification system based on artificial intelligence according to claim 1, characterized in that: The short-wave infrared camera module and the visible light camera module are integrated into an adjustable slide rail bracket, and the bracket is equipped with a laser positioner and an angle encoder; The slide rail bracket is provided with a standard light source box, which is used to provide a uniform lighting environment with a color temperature of 2000±50K; The device has a built-in image alignment algorithm that automatically corrects the pixel-level spatial matching of bispectral images.

6. A porcelain identification method based on artificial intelligence, characterized in that: The following steps are involved: Through the image synchronization control module, the short-wave infrared image and visible light image of the porcelain are obtained in the same spatial coordinate system. Extract glaze penetration features from shortwave infrared images, including body joint marks, internal repair areas, and material aging gradients; Perform style matching analysis between the glaze color and patterns of visible light images and the historical kiln feature database; By integrating dual-spectral feature data through the AI ​​image processing module, an identification report is generated that includes the probability of age, kiln ownership and authenticity level.

Images