Single artwork traceability identification method
By combining macro and micro information extraction technology, using high-resolution cameras and microscopes, combined with color histograms and image analysis, the problem of unstable accuracy of art identification in the existing technology is solved, and higher identification accuracy and reliability.
Patent Information
- Application Number
- CN202411262048.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the method of tracing and identification of single-piece artworks depends on expert experience, resulting in unstable identification accuracy and affecting the effectiveness and reliability of the method.
Using a combination of macro information extraction and micro information extraction, artworks are shot from different angles through high-resolution cameras and microscopes, their macro and micro information data are extracted, and their color histograms and image analysis techniques are compared to judge art. authenticity of the product.
It improves the accuracy and reliability of artwork appraisal, reduces misjudgment, provides an objective basis for appraisal, and enhances identification efficiency.
Smart Images

Figure CN119049048A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of artwork traceability and identification, and in particular to a method for tracing and identifying a single artwork. Background Art
[0002] In the current art market environment, the traceability and identification of individual artworks is of great significance. As art trading activities become more frequent and counterfeiting technology continues to improve, it has become extremely urgent to accurately judge the authenticity, source and history of artworks. At the same time, modern counterfeiting methods are becoming more and more sophisticated. It is difficult to accurately judge the authenticity of an artwork based on its appearance or a single feature. Traditional identification methods mostly rely on expert experience to subjectively determine the authenticity of an artwork.
[0003] However, with current technology, different experts have different levels of experience, cognitive perspectives, and aesthetic standards. A single identification of appearance based solely on subjective experience will lead to unstable identification accuracy, affect the effectiveness and reliability of the identification method, and thus make it impossible to effectively trace and identify. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention provides a method for tracing the origin of a single artwork, which solves the problem that the single identification method leads to unstable identification accuracy and affects the effectiveness and reliability of the identification method.
[0005] To achieve the above purpose, the present invention is implemented through the following technical scheme: a method for tracing the origin of a single artwork comprises the following steps: S1: Macro information extraction: Place the original artwork on the inspection table and use a high-resolution camera to take pictures of the total field of view and the surrounding environment from different angles to extract the complete macro information data of the original artwork; S2: Extraction of microscopic information: Select the texture features of the microscopic image at the characteristic position of the original artwork through a microscope, and adjust the focal length, magnification and lighting conditions of the microscope so that the total field of view is clearly displayed in the field of view of the microscope, and then record the complete microscopic information data of the original artwork; S3: Macroscopic identification of artworks: Place the artwork to be identified on the same inspection table as the original artwork and keep the same lighting environment as the original artwork, take macroscopic photos of it, compare the macroscopic color histograms of the original artwork and the artwork to be identified after preprocessing, and analyze and process the images through preliminary description and digital image processing; S4: Microscopic identification of artworks: extracting the texture features of the microscopic images of the artwork to be tested and the original artwork at the characteristic positions, and comparing them one by one through image analysis technology; S5: Artwork determination: if the artwork to be determined is placed in a location consistent with the original artwork in terms of macro information, a microscopic image analysis is performed; otherwise, it is a fake; if the microscopic inspection is consistent with the microscopic texture information data of the original artwork, it is authentic; otherwise, it is a fake.
[0006] Preferably, the color of the detection platform in S1 is selected as a neutral color, and the neutral color includes gray and black, and a medium format camera and uniform lighting equipment are selected to accurately shoot it, and the angles include front angle, side angle, and overhead angle.
[0007] Preferably, the characteristic positions of the original artwork in S2 are positions of significant artistic features of the artwork, including signatures and inscriptions, decorations and patterns, and areas with rich details and textures. The microscope is selected at 60x to record the characteristic positions.
[0008] Preferably, the image preprocessing in S3 is to first capture a macro image of the artwork to be determined, and at the same time convert its color space into a format that is unified with the macro information data collected from the original artwork, and then sharpen the image with a sharpening radius of 0.3-3 pixels, a number of 50%-200%, and a threshold of 0-20.
[0009] Preferably, the color features in S3 are respectively calculated by using a color histogram analysis tool to calculate the color histograms of the two images, observe the similarities and differences in their color distribution, and compare them by calculating the Bhattacharyya distance according to the color histogram. The calculation formula of the Bhattacharyya distance is: ; in represents the Bhattacharyya distance, P and Q represent the probability distributions corresponding to the two color histograms, i represents each interval or color channel of the color histogram, and P(i) and Q(i) represent the probability values in the intervals, respectively.
[0010] Preferably, in S4, the contrast between the color microscopic images at multiple characteristic positions of the artwork to be tested and the original artwork is increased, and then converted into a grayscale image to extract texture features.
[0011] Preferably, the image analysis technology in S4 is to extract the entropy of texture features through gray level co-occurrence matrix technology and calculate its similarity value, which is between 0-1.
[0012] Preferably, in S5, if the Bhattacharyya distance in the color feature of the original artwork in S3 is consistent with that to be tested, then further analysis of the microscopic image is performed, otherwise it is a fake, and if the similarity value in the microscopic detection is 0.9-1, it is a genuine product, otherwise it is a fake.
[0013] The present invention provides a method for tracing the origin of a single artwork. It has the following beneficial effects: 1. The present invention can comprehensively and accurately identify artworks by combining macroscopic color histogram identification with microscopic image analysis identification. Macroscopic identification can grasp the artwork as a whole to judge its authenticity and age. Microscopic identification deeply analyzes the texture and color details to find microscopic feature differences that are difficult to forge. The combination of the two improves the accuracy of identification and reduces the effect of misjudgment.
[0014] 2. The present invention unifies the color space of two images at the same position in the macro image preprocessing and sharpens the images one by one to enhance the details and contrast, which is convenient for accurate comparison. It also quantifies the color distribution difference through color histogram and Bhattacharyya distance comparison, provides an objective basis for identification, and improves the accuracy, reliability and efficiency of identification. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 The present invention is a flowchart of the method for tracing the origin of a single artwork. DETAILED DESCRIPTION
[0016] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0017] Please refer to the attached Figure 1 The embodiment of the present invention provides a method for tracing the origin of a single artwork, comprising the following steps: S1: Macro information extraction: Place the original artwork on the inspection table and use a high-resolution camera to take pictures of the total field of view and the surrounding environment from different angles to extract the complete macro information data of the original artwork; S2: Extraction of microscopic information: Select the texture features of the microscopic image at the characteristic position of the original artwork through a microscope, and adjust the focal length, magnification and lighting conditions of the microscope so that the total field of view is clearly displayed in the field of view of the microscope, and then record the complete microscopic information data of the original artwork; S3: Macroscopic identification of artworks: Place the artwork to be identified on the same inspection table as the original artwork and keep the same lighting environment as the original artwork, take macroscopic photos of it, compare the macroscopic color histograms of the original artwork and the artwork to be identified after preprocessing, and analyze and process the images through preliminary description and digital image processing; S4: Microscopic identification of artworks: extracting the texture features of the microscopic images of the artwork to be tested and the original artwork at the characteristic positions, and comparing them one by one through image analysis technology; S5: Artwork determination: if the artwork to be determined is placed in a location consistent with the original artwork in terms of macro information, a microscopic image analysis is performed; otherwise, it is a fake; if the microscopic inspection is consistent with the microscopic texture information data of the original artwork, it is authentic; otherwise, it is a fake.
[0018] In the S1, the color of the detection platform is selected as a neutral color, which includes gray and black, and a medium format camera and uniform lighting equipment are selected to accurately shoot it, and the angles include front angle, side angle, and overhead angle.
[0019] Specifically, choosing a neutral color test stand will not reflect strong colors when shooting, thus avoiding the interference of the color of the test stand itself on the color of the artwork, thereby ensuring that the color of the artwork can be accurately captured and analyzed without deviation due to the color reflection of the test stand. Medium format cameras usually have higher pixels and resolutions, which can capture the subtle details of the artwork. At the same time, they have a wider dynamic range and can capture high-quality images under different lighting conditions, thereby improving the accuracy of the artwork when shooting and recording, ensuring the embodiment of the details of the artwork, and through multi-angle shooting, its appearance and characteristics can be fully displayed. The front angle can show the main picture and composition of the artwork, the side angle can show its thickness and three-dimensional sense, and the overhead angle can show the details and overall shape of its top, thereby improving the completeness of the macro data of the original artwork and ensuring the accuracy of subsequent identification.
[0020] The characteristic positions of the original artwork in S2 are the positions of significant artistic features of the artwork, including signatures and inscriptions, decorations and patterns, and areas with rich details and textures. The microscope is selected at 60x to record the characteristic positions.
[0021] Specifically, signatures and inscriptions are the artist's identification of the work, which often has a unique style and characteristics. Extracting microscopic information from signatures and inscriptions can help determine the author's identity and the authenticity of the work. At the same time, decorations and patterns are important components of artworks. Artworks of different periods and styles often have unique characteristics in decorations and patterns. Extracting microscopic information from decorations and patterns can help determine the age, style and production process of the work. Details and rich textures can often reflect the artist's skills and the uniqueness of the work. Extracting microscopic information from these locations can help analyze the production process, material characteristics and preservation status of the work. Through the above, the recognition of the position during identification can be better guaranteed, the accuracy of preliminary identification, and the credibility of subsequent identification and evaluation can be improved. At the same time, the use of a 60x magnification can provide a higher resolution, making the texture features of the microscopic image more clearly visible. For the extraction of microscopic information of artworks, high magnification can help discover some details that are difficult to detect with the naked eye and improve the accuracy of identification.
[0022] The image preprocessing in S3 is to first take a macro image of the artwork to be determined, and at the same time convert its color space into a format that is unified with the macro information data collected from the original artwork, and then sharpen the image with a sharpening radius of 0.3-3 pixels, an amount of 50%-200%, and a threshold of 0-20.
[0023] Specifically, converting the color space of the artwork to be determined into a format that is unified with the macro information data collected from the original artwork can ensure that in subsequent analysis and comparison, the two artworks are in the same color system. Different color spaces have different ways of representing colors. If they are not unified, inaccurate color contrast may result. This facilitates subsequent color feature extraction and comparison, and improves the accuracy and comparability of color feature extraction. By adjusting the sharpening radius, amount, and threshold, the detailed features of the texture, brushstrokes, patterns, etc. in the image can be highlighted, making the color features more distinct, thereby optimizing the feature extraction effect.
[0024] The color features in S3 are calculated by using a color histogram analysis tool to calculate the color histograms of the two images respectively, observe the similarities and differences in their color distribution, and compare them by calculating the Bhattacharyya distance according to the color histogram. The calculation formula of the Bhattacharyya distance is: ; in represents the Bhattacharyya distance, P and Q represent the probability distributions corresponding to the two color histograms, i represents each interval or color channel of the color histogram, and P(i) and Q(i) represent the probability values in the intervals, respectively.
[0025] Specifically, the color histogram shows the distribution of different colors in the image in an intuitive way. By observing the color histogram, the appraiser can quickly understand the main color components of the artwork, the brightness and saturation distribution of the color, etc., and then by calculating the color histograms of the original artwork and the artwork to be determined, the similarities and differences in color distribution between the two can be intuitively compared. The Bhattacharyya distance measures the similarity between the two color histograms. A smaller Bhattacharyya distance indicates that the color distribution of the two artworks is more similar, and vice versa.
[0026] In S4, the contrast between the color microscopic images at multiple characteristic positions of the artwork to be tested and the original artwork is improved, and then converted into a grayscale image to extract texture features.
[0027] Specifically, improving the contrast can make the differences between different colors in the color microscopic image more significant, which can help discover the similarities and differences in color usage between the artwork to be tested and the original artwork, and provide richer information for texture feature extraction. Subsequently, converting the color image to a grayscale image can remove color interference and make the texture features more prominent. By converting both into grayscale images, texture feature extraction and comparison can be performed on the same basis to ensure the reliability of the identification results.
[0028] The image analysis technology in S4 is to extract the entropy of texture features through gray level co-occurrence matrix technology and calculate its similarity value, which is between 0 and 1.
[0029] Specifically, the gray-level co-occurrence matrix is a method for analyzing image texture features. For a gray-level image, the gray-level co-occurrence matrix describes the frequency of occurrence of pixel pairs with different gray-level values under a specific spatial relationship. A variety of texture features can be extracted from the gray-level co-occurrence matrix, among which entropy is an important feature. Entropy as a texture feature can quantify the complexity of the image texture. The similarity value is between 0 and 1. The closer to 1, the more similar the texture features of the two artworks are, and the closer to 0, the greater the difference.
[0030] In S5, if the Bhattacharyya distance in the color feature of the original artwork in S3 is consistent with that to be tested, then further analysis of the microscopic image is performed, otherwise it is a fake. If the similarity value in the microscopic detection is 0.9-1, it is a genuine product, otherwise it is a fake.
[0031] Specifically, by comparing the Bhattacharyya distance in the color features of the original artwork and the artwork to be tested in the macro information, a preliminary screening can be performed before conducting micro image analysis. The Bhattacharyya distance measures the similarity between two color histograms and can quickly determine the difference in color distribution between the two artworks, indicating that the two artworks are relatively similar in color features and that it is necessary to further conduct micro image analysis. This can avoid unnecessary micro analysis of counterfeits with large color differences, thereby improving identification efficiency. At the same time, micro texture information is usually difficult to forge. By comparing micro textures, subtle differences between counterfeits and authentic works can be found. The judgment of similarity values can quantify such differences, making the identification results more objective and reliable, thereby ensuring the comprehensiveness and accuracy of the identification results.
[0032] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for tracing the origin of a single artwork, characterized in that: The following steps are included: S1: Macro information extraction: Place the original artwork on the inspection table and use a high-resolution camera to take pictures of the total field of view and the surrounding environment from different angles to extract the complete macro information data of the original artwork; S2: Extraction of microscopic information: Select the texture features of the microscopic image at the characteristic position of the original artwork through a microscope, and adjust the focal length, magnification and lighting conditions of the microscope so that the total field of view is clearly displayed in the field of view of the microscope, and then record the complete microscopic information data of the original artwork; S3: Macroscopic identification of artworks: Place the artwork to be identified on the same inspection table as the original artwork and keep the same lighting environment as the original artwork, take macroscopic photos of it, compare the macroscopic color histograms of the original artwork and the artwork to be identified after preprocessing, and analyze and process the images through preliminary description and digital image processing; S4: Microscopic identification of artworks: extracting the texture features of the microscopic images of the artwork to be tested and the original artwork at the characteristic positions, and comparing them one by one through image analysis technology; S5: Artwork determination: if the artwork to be determined is placed in a location consistent with the original artwork in terms of macro information, a microscopic image analysis is performed; otherwise, it is a fake; if the microscopic inspection is consistent with the microscopic texture information data of the original artwork, it is authentic; otherwise, it is a fake.
2. The method for tracing the origin of a single artwork according to claim 1, characterized in that: In the S1, the color of the detection platform is selected as a neutral color, which includes gray and black, and a medium format camera and uniform lighting equipment are selected to accurately shoot it, and the angles include front angle, side angle, and overhead angle.
3. The method for tracing the origin of a single artwork according to claim 1, characterized in that: The characteristic positions of the original artwork in S2 are the positions of significant artistic features of the artwork, including signatures and inscriptions, decorations and patterns, and areas with rich details and textures. The microscope is selected at 60x to record the characteristic positions.
4. The method for tracing the origin of a single artwork according to claim 1, characterized in that: The image preprocessing in S3 is to first take a macro image of the artwork to be determined, and at the same time convert its color space into a format that is unified with the macro information data collected from the original artwork, and then sharpen the image with a sharpening radius of 0.3-3 pixels, an amount of 50%-200%, and a threshold of 0-20.
5. The method for tracing the origin of a single artwork according to claim 1, characterized in that: The color features in S3 are calculated by using a color histogram analysis tool to calculate the color histograms of the two images respectively, observe the similarities and differences in their color distribution, and compare them by calculating the Bhattacharyya distance according to the color histogram. The calculation formula of the Bhattacharyya distance is: ; in represents the Bhattacharyya distance, P and Q represent the probability distributions corresponding to the two color histograms, i represents each interval or color channel of the color histogram, and P(i) and Q(i) represent the probability values in the intervals, respectively.
6. The method for tracing the origin of a single artwork according to claim 1, characterized in that: In S4, the contrast between the color microscopic images at multiple characteristic positions of the artwork to be tested and the original artwork is improved, and then converted into a grayscale image to extract texture features.
7. The method for tracing the origin of a single artwork according to claim 1, characterized in that: The image analysis technology in S4 is to extract the entropy of texture features through gray level co-occurrence matrix technology and calculate its similarity value, which is between 0 and 1.
8. The method for tracing the origin of a single artwork according to claim 1, characterized in that: In S5, if the Bhattacharyya distance in the color feature of the original artwork in S3 is consistent with that to be tested, then further analysis of the microscopic image is performed, otherwise it is a fake. If the similarity value in the microscopic detection is 0.9-1, it is a genuine product, otherwise it is a fake.
Citation Information
Cited By
Artwork traceability management method and system
CN119539839A
Method for implementing a learning system integrated into a mobile application for the recognition and generation of artistic works
MA69151A1
Method for implementing a learning system integrated into a mobile application for the recognition and generation of artistic works
MA69151B1
