Method and system for constructing digital material evidence chain of cultural relics based on micro-physical feature extraction
By collecting macroscopic images and surface microscopic physical fingerprint features of cultural relics, a high-fidelity digital twin archive is generated and a deep feature vector is extracted using an AI engine. This solves the problem of difficulty in defining responsibility for physical replacement and damage in cultural relic management, realizes a real-time, tamper-proof digital evidence chain, and provides judicial-level verification capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ARTRON ART GRP CO LTD
- Filing Date
- 2026-04-10
- Publication Date
- 2026-07-10
Smart Images
Figure CN122364321A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of cultural relic management technology, and in particular to a method and system for constructing a digital evidence chain for cultural relics based on the extraction of microscopic physical features. Background Technology
[0002] Currently, the management and authentication of cultural relics largely rely on visual inspection, paper seals, or ordinary optical photographs. However, existing methods of cultural relic management have the following drawbacks: First, paper seals are easily forged, and visual inspection is insufficient to detect partial replacements or substitutions down to the millimeter level. Second, ordinary photographs lack quantifiable technical data strongly correlated with the physical characteristics of the cultural relic, making it impossible to form a legally valid chain of evidence in cases of damage or disputes over transfer. Third, existing management methods lack cross-institutional, real-time consistency verification mechanisms based on the physical characteristics of the cultural relic, leading to prominent technical problems such as physical replacement, difficulty in determining liability for damage, and serious regulatory lags during the transfer of cultural relics, including their release and return from storage.
[0003] Therefore, it is necessary to provide a method and system for constructing digital evidence chains of cultural relics based on the extraction of microscopic physical features in order to overcome the above-mentioned defects. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for constructing a digital evidence chain for cultural relics based on the extraction of microscopic physical features. This aims to solve the problems of the risk of physical replacement and the difficulty in accurately defining liability for damage during the circulation of cultural relics, forming a closed-loop system that strongly combines "physical and digital" elements, and providing regulatory authorities with penetrating verification capabilities.
[0005] To achieve the above objectives, this invention provides a method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction, comprising: Step S10: Collect macroscopic images and surface microscopic physical fingerprint features of cultural relics based on predetermined anchor points to generate a high-fidelity digital twin archive containing multimodal features; Step S20: Based on the surface microphysical fingerprint features, a depth feature vector for uniquely identifying the artifact is extracted and generated using a preset AI engine; Step S30: In the cultural relic circulation process, the predetermined anchor points on the cultural relic are re-sampled in situ, and the AI engine is used to compare the re-sampled features with the depth feature vector in milliseconds, and output the consistency verification result. Step S40: Hash the high-fidelity digital twin archive, the deep feature vector, and the consistency verification result to generate a digital digest and store it in the cloud dynamic evidence database; Step S50: Bind the digital digest with the timestamp and upload it to the blockchain network for storage, forming an immutable digital evidence chain.
[0006] In a preferred embodiment, step S10, collecting the microscopic physical fingerprint features of the artifact surface based on predetermined anchor points, includes: collecting corresponding physical features for artifacts of different materials: For paper or ancient book artifacts, collect the random physical topology of paper fibers; for porcelain artifacts, collect the geometric distribution of glaze bubble structure and surface cracks; for bronze or gold and silver artifacts, collect the microscopic crystallization points formed by metal corrosion. Among them, several core verification anchor points were selected on the surface of the cultural relics as the predetermined locations for in-situ re-sampling.
[0007] In a preferred embodiment, in step S10, when collecting surface microscopic physical fingerprint features, a microscopic imaging device is used for 100-200x magnification to ensure that the single pixel resolution is ≤5 micrometers, the positioning error of repeated collection is ≤0.3mm, and the light source deviation is ≤2 degrees; wherein, an X-ray fluorescence spectrometer is used to collect material element composition data at at least 4 typical points on the cultural relic as auxiliary features.
[0008] In a preferred embodiment, step S20 specifically includes: The original microscopic image containing the surface microscopic physical fingerprint features is subjected to denoising, resampling, and multi-scale spatial enhancement preprocessing. For the preprocessed microscopic images, a deep convolutional network is used to extract global texture features, and a scale-invariant feature transform operator is used to extract local key point features. The global texture features and the local key point features are heterogeneously fused to generate a high-dimensional deep feature vector with a dimension ≥ 512. A local topological constraint mechanism is introduced to analyze the relative geometric relationships between key feature points, and the fused feature vector is transformed into a compact hash code.
[0009] In a preferred embodiment, step S30, the millisecond-level comparison specifically includes: The hash code is used for coarse screening in the cloud-based dynamic evidence database; For the coarse screening results, the similarity of their deep feature vectors is calculated by calculating Hamming distance or cosine similarity in order to perform fine matching; Image overlap analysis is performed on the fine matching results. When the overall matching confidence is greater than or equal to the preset threshold, the identities are determined to be consistent.
[0010] In a preferred embodiment, during the artifact removal verification process, the baseline archive is retrieved and the original location is re-collected; during the artifact return verification process, minute cracks or damage are detected, a new depth feature vector is generated, and a status change analysis report is generated based on the depth feature vector from the previous removal.
[0011] In a preferred embodiment, in step S40, the objects of hashing processing include at least: the high-fidelity digital twin archive, the deep feature vector, the consistency verification result, and the state change analysis report; In step S50, the blockchain network connects to the national-level evidence storage node and performs real-time disaster recovery backup of core data.
[0012] This invention also provides a system for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction, used to execute the method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in any of the above embodiments, including: The high-fidelity acquisition module is used to acquire macroscopic images and surface microscopic physical fingerprint features of cultural relics; The feature extraction and comparison module is used to generate deep feature vectors and perform millisecond-level consistency comparisons. The business closed-loop control module is used to manage the outbound verification and return review process and generate status analysis reports; The blockchain evidence storage module is used to hash the verification data and upload it to the blockchain network.
[0013] In a preferred embodiment, the high-fidelity acquisition module includes: Large format scanners or medium format cameras are used for gridded area acquisition, with a resolution of ≥600ppi. Color reproduction error ΔE≤2; Microscopic imaging equipment used for 100-200x magnification acquisition, with a single pixel resolution of ≤5 micrometers; X-ray fluorescence spectrometer is used to collect material elemental composition data at typical locations of cultural relics.
[0014] The present invention provides a method and system for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction. First, by collecting macroscopic images and microscopic physical fingerprint features of the cultural relic, a high-fidelity digital twin archive is created. Second, an artificial intelligence engine extracts a unique, high-dimensional deep feature vector from the microscopic fingerprint, serving as the cultural relic's "digital DNA." Then, each time the cultural relic is taken out of or returned to storage, its microscopic features are re-collected and compared with the "digital DNA" in the archive at millisecond levels to verify identity consistency. Finally, the key data throughout the process (archive, feature vector, comparison results) are hashed and stored in the blockchain along with a timestamp, forming an immutable digital evidence chain. Therefore, by strongly binding physical features with digital identity, the risk of physical substitution is fundamentally eliminated; by using millisecond-level dynamic comparison and blockchain storage, an evidence chain with legal force is formed; and by cross-institutional digital verification, transparent and real-time monitoring of the cultural relic's status is achieved. Attached Figure Description
[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 A flowchart of the method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction provided by the present invention; Figure 2 The framework diagram of the digital evidence chain construction system for cultural relics based on microscopic physical feature extraction provided by the present invention. Detailed Implementation
[0017] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described in this specification are merely for explaining the invention and are not intended to limit the invention.
[0018] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.
[0019] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0020] Example 1 In an embodiment of the present invention, a method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction is provided. By taking the unreplicable microscopic physical fingerprint of cultural relics as the core of their digital identity, and combining artificial intelligence comparison and blockchain evidence storage technology, a closed-loop system with strong physical-digital binding is formed to completely eliminate the risk of physical replacement in the circulation of cultural relics, accurately define the responsibility for damage, and provide regulatory authorities with penetrating verification capabilities.
[0021] like Figure 1 As shown, the method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction includes steps S10-S50. The step numbers are for simplicity and are not intended to limit the execution order. A specific embodiment (e.g., a blue-and-white porcelain plate from the Ming and Qing dynasties) will be used to describe the technical solution of this invention in detail.
[0022] Step S10: Based on the macroscopic images and surface microscopic physical fingerprint features of the cultural relics collected from the predetermined anchor points, a high-fidelity digital twin archive containing multimodal features is generated.
[0023] It should be noted that corresponding physical characteristics are collected for cultural relics made of different materials: for paper or ancient books, the random physical topological structure of paper fibers is collected; for porcelain, the geometric distribution of glaze bubble structure and surface cracks is collected; for bronze or gold and silver artifacts, the microscopic crystallization points generated by metal corrosion are collected; and multiple core verification anchor points are selected on the surface of the cultural relics as predetermined locations for in-situ resampling.
[0024] In this exemplary embodiment, for the blue and white porcelain plate, the operator first uses a 100-megapixel medium format camera to take pictures according to a grid-based partitioning method. The camera resolution is set to 600ppi, and the color reproduction error ΔE is controlled within 2 to ensure high color and detail reproduction of the macroscopic image.
[0025] Next, based on the material of the artifact, corresponding microscopic physical fingerprint features are collected. Since this example involves porcelain, the system guides the operator to focus on collecting the geometric distribution of glaze bubble structures and surface cracks. The operator selects four core verification anchor points on the porcelain plate surface; for example, a densely populated bubble area in the main pattern area at the center of the plate, the edge of a medallion pattern on the rim, and two locations with prominent glaze features within the foot ring. These anchor points will serve as predetermined locations for all future in-situ resampling.
[0026] During data acquisition, operators used a microscopic imaging device with motorized focusing and positioning capabilities to magnify and acquire data at each anchor point 120 times. The system ensured a single-pixel resolution of 3 micrometers, used an image matching algorithm to control the positioning error of repeated acquisitions to within 0.2 millimeters, and stabilized the illumination deviation angle to within 1 degree using a standard light source. Simultaneously, a portable X-ray fluorescence spectrometer was used to collect elemental composition data at four typical points on the porcelain plate surface (e.g., areas with deep blue-and-white color, near glaze cracks, and exposed clay) as auxiliary identification features. All the macroscopic images, microscopic structural images, and elemental composition data acquired together constituted a high-fidelity digital twin archive of the blue-and-white porcelain plate.
[0027] Step S20: Based on the collected surface micro-physical fingerprint features, a depth feature vector for uniquely identifying the artifact is extracted and generated using a preset AI engine.
[0028] The AI engine first preprocesses the acquired microscopic images of anchor points (e.g., bubble images): it uses a nonlocal mean filtering algorithm for noise reduction, uses bicubic interpolation for resampling to unify image size, and constructs a Gaussian pyramid for multi-scale spatial enhancement, effectively eliminating interference caused by ambient light and device differences.
[0029] After preprocessing, the system enters a composite feature extraction network. On one hand, a pre-trained deep convolutional network (e.g., ResNet-50) extracts global texture features (e.g., 2048 dimensions) from the image, such as the overall bubble density distribution and tonal patterns of the glaze. On the other hand, an improved Scale Invariant Feature Transform (SIFT) operator is used to extract local keypoint features. Its simplified steps include: first, constructing a Gaussian difference scale space; second, detecting extrema in each scale space as candidate keypoints (e.g., the center point of a single bubble); third, precisely determining the keypoint location and scale; and finally, calculating the gradient orientation histogram within the keypoint neighborhood to form a 128-dimensional local feature descriptor.
[0030] Subsequently, the system heterogeneously fuses the 2048-dimensional global feature vector extracted by the convolutional network with the 128-dimensional local feature vectors extracted by multiple SIFT operators. The fusion method uses feature concatenation followed by a fully connected layer for dimensionality reduction, ultimately generating a 512-dimensional high-dimensional depth feature vector, which is the "digital DNA" of the blue and white porcelain plate.
[0031] To further improve retrieval efficiency and anti-counterfeiting capabilities, the system introduces a local topological constraint mechanism. The algorithm analyzes the relative geometric relationships between key feature points (such as the center points of adjacent bubbles) in the anchor image, including distance and angle, to form a topological constraint code. Finally, this feature vector, which integrates global, local, and topological information, is transformed into a compact hash code using the Locality Sensitive Hash (LSH) algorithm, facilitating rapid subsequent indexing. This hash code, along with the original 512-dimensional feature vector, is stored in a cloud-based dynamic evidence database and linked to the digital twin archive of the artifact.
[0032] When the blue and white porcelain plate needs to be taken out of the warehouse for exhibition, step S30 is executed. In the process of cultural relic circulation, the system re-samples the predetermined anchor points on the cultural relic in situ, and uses an AI engine to compare the re-sampled features with the deep feature vector in milliseconds, and outputs the consistency verification result.
[0033] Specifically, the operator retrieves the baseline file for the porcelain plate with a single click using a handheld terminal. The system automatically guides the microscope to the four core verification anchor points recorded in the previous step, re-acquires data at the original locations, and obtains new microscopic images.
[0034] The AI engine then begins a millisecond-level comparison. First, it uses the hash code of the new image to perform a coarse screening in a dynamic evidence database in the cloud, quickly locating the archive containing the porcelain plate. Then, it performs a fine-grained match by calculating the cosine similarity between the newly extracted deep feature vector and the 512-dimensional feature vector stored in the archive. The formula for calculating cosine similarity is: Similarity (A,B) = (A·B) / (||A||×||B||), where A and B are two feature vectors. The closer the calculated similarity value is to 1, the more similar the two are.
[0035] Finally, the system performs SSIM (Structural Similarity Index) analysis on the original and resampled images. SSIM compares the similarity between two images across three dimensions: brightness, contrast, and structure, outputting a score between 0 and 1. The system combines the cosine similarity and the SSIM score to obtain a comprehensive matching confidence level between 0 and 100. When the confidence level is greater than or equal to 98, the system determines that the identities match. The entire process, from resampled image to output result, takes only 2.5 minutes per artifact, far below the industry standard. Simultaneously, while ensuring rapid retrieval efficiency for massive amounts of data, it significantly improves the accuracy and robustness of the final matching results, ensuring that the decision threshold is scientifically reliable.
[0036] Step S40: Hash the high-fidelity digital twin archive, deep feature vector, and consistency verification results to generate a digital digest and store it in the cloud dynamic evidence database.
[0037] Step S50: Bind the digital digest with the timestamp and upload it to the blockchain network for evidence storage, forming an immutable digital evidence chain.
[0038] During the artifact return verification process, the system not only verifies identity but also pays close attention to minute signs of damage. Step S30, involving re-sampling and comparison, is executed again. At this point, the AI engine's sensitivity is increased, enabling it to detect minute cracks or glaze wear as small as 0.2 millimeters. The system generates a new depth feature vector and, based on the depth feature vector stored during the previous retrieval, produces a "State Change Analysis Report." The report highlights the locations of newly appearing cracks and estimates their size.
[0039] Subsequently, steps S40 and S50 are executed. The system hashes the high-fidelity digital twin archive, the original deep feature vector, the consistency verification result of this database restore, and the newly generated "State Change Analysis Report" using the SHA-256 algorithm to generate a fixed-length digital digest. This digital digest is bound to a timestamp accurate to milliseconds and uploaded to the network for notarization via a blockchain node.
[0040] In this embodiment, the blockchain network connects to evidence storage nodes recognized by the State Administration of Cultural Heritage and performs real-time disaster recovery backups of all core data to ensure long-term data security and immutability. This ultimately forms a complete, legally credible digital evidence chain, ensuring the compliance, security, and immutability of long-term data preservation, meeting national-level archiving requirements, and providing a solid technical foundation for forensic identification. Furthermore, it achieves closed-loop management of the entire lifecycle of cultural relics, efficiently preventing swapping during the outbound process and accurately defining liability for damage during the return process, providing objective evidence for liability determination.
[0041] Example 2 This invention also provides a system 100 for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction, used to perform all the above methods. For example... Figure 2 As shown, the system comprises four core modules: High-fidelity acquisition module 10: Responsible for acquiring macroscopic images and surface microscopic physical fingerprint features of cultural relics. For example, this module specifically includes a 100-megapixel medium format camera (for gridded regional acquisition, resolution 600ppi, color reproduction error ΔE≤2), a 100-200x magnified microscopic imaging device (single pixel resolution ≤5 micrometers), and an X-ray fluorescence spectrometer (for acquiring elemental composition data).
[0042] Feature extraction and comparison module 20: Internally running the aforementioned composite AI algorithm architecture, it is responsible for generating deep feature vectors and performing millisecond-level consistency comparisons. The encoding generation response time of this module is less than 2 seconds, the success rate of repeated comparisons of the same cultural relic reaches over 99.5%, and the false recognition rate of different cultural relics is less than 0.1%.
[0043] Business closed-loop control module 30: Responsible for managing the verification process for cultural relics leaving the warehouse and the review process for returning them to the warehouse. It can automatically guide operators to complete the original location re-sampling, call the comparison module for identity verification, detect status changes at the 0.2 mm level, and finally automatically generate the "Status Change Analysis Report" and the "Return Consistency Report".
[0044] Blockchain Evidence Storage Module 40: It is responsible for hashing the key data (such as reports, feature vectors, comparison results) generated by each verification operation, binding them with timestamps, and uploading them to the blockchain network connected to the national-level evidence storage node to form an immutable digital evidence chain.
[0045] In summary, this invention uses macroscopic images and microscopic physical fingerprints (such as fibers and bubbles) of cultural relics as uncopyable "physical DNA" and utilizes AI to generate unique high-dimensional feature vectors, thereby achieving a strong binding between physical entities and digital identities. This fundamentally eliminates the risk of physical replacement of cultural relics during the circulation process, because no replacement can match the original archive at the microscopic physical fingerprint level.
[0046] Furthermore, a closed-loop process of "in-situ re-sampling - millisecond-level comparison" is adopted in the circulation process, and the verification process and results are stored in real time through blockchain, forming a traceable and tamper-proof judicial-grade physical evidence chain. Any damage or change in status can be accurately recorded and the responsible link can be locked, which greatly improves the legal effect of the evidence.
[0047] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0048] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0049] Those skilled in the art will recognize that the units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0050] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0051] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0052] The present invention is not limited to the description in the specification and embodiments, and thus other advantages and modifications can be readily realized by those skilled in the art. Therefore, the present invention is not limited to the specific details, representative devices and illustrated examples shown and described herein without departing from the spirit and scope of the general concept as defined by the claims and their equivalents.
Claims
1. A method for constructing a digital evidence chain for cultural relics based on the extraction of microscopic physical features, characterized in that, include: Step S10: Collect macroscopic images and surface microscopic physical fingerprint features of cultural relics based on predetermined anchor points to generate a high-fidelity digital twin archive containing multimodal features; Step S20: Based on the surface microphysical fingerprint features, a depth feature vector for uniquely identifying the artifact is extracted and generated using a preset AI engine; Step S30: In the cultural relic circulation process, the predetermined anchor points on the cultural relic are re-sampled in situ, and the AI engine is used to compare the re-sampled features with the depth feature vector in milliseconds, and output the consistency verification result. Step S40: Hash the high-fidelity digital twin archive, the deep feature vector, and the consistency verification result to generate a digital digest and store it in the cloud dynamic evidence database; Step S50: Bind the digital digest with the timestamp and upload it to the blockchain network for storage, forming an immutable digital evidence chain.
2. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 1, characterized in that, In step S10, collecting the microscopic physical fingerprint features of the artifact surface based on predetermined anchor points includes: collecting corresponding physical features for artifacts of different materials: For paper or ancient book artifacts, collect the random physical topology of paper fibers; for porcelain artifacts, collect the geometric distribution of glaze bubble structure and surface cracks; for bronze or gold and silver artifacts, collect the microscopic crystallization points formed by metal corrosion. Among them, several core verification anchor points were selected on the surface of the cultural relics as the predetermined locations for in-situ re-sampling.
3. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 1, characterized in that, In step S10, when collecting surface microscopic physical fingerprint features, a microscopic imaging device is used for 100-200 times magnification to ensure that the single pixel resolution is ≤5 micrometers, the positioning error of repeated collection is ≤0.3mm, and the light source deviation is ≤2 degrees; among them, an X-ray fluorescence spectrometer is used to collect material element composition data at at least 4 typical points on the cultural relic as auxiliary features.
4. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 1, characterized in that, Step S20 specifically includes: The original microscopic image containing the surface microscopic physical fingerprint features is subjected to denoising, resampling, and multi-scale spatial enhancement preprocessing. For the preprocessed microscopic images, a deep convolutional network is used to extract global texture features, and a scale-invariant feature transform operator is used to extract local key point features. The global texture features and the local key point features are heterogeneously fused to generate a high-dimensional deep feature vector with a dimension ≥ 512. A local topological constraint mechanism is introduced to analyze the relative geometric relationships between key feature points, and the fused feature vector is transformed into a compact hash code.
5. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 4, characterized in that, In step S30, the millisecond-level comparison specifically includes: The hash code is used for coarse screening in the cloud-based dynamic evidence database; For the coarse screening results, the similarity of their deep feature vectors is calculated by calculating Hamming distance or cosine similarity in order to perform fine matching; Image overlap analysis is performed on the fine matching results. When the overall matching confidence is greater than or equal to the preset threshold, the identities are determined to be consistent.
6. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 1, characterized in that, During the verification process of cultural relics leaving the warehouse, the baseline archive is retrieved and the original location is re-collected; during the verification process of cultural relics returning to the warehouse, minute cracks or damage are detected, new depth feature vectors are generated, and a status change analysis report is generated based on the depth feature vectors from the previous time the relics were left out of the warehouse.
7. The method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 6, characterized in that, In step S40, the objects of hashing processing include at least: the high-fidelity digital twin archive, the deep feature vector, the consistency verification result, and the state change analysis report; In step S50, the blockchain network connects to the national-level evidence storage node and performs real-time disaster recovery backup of core data.
8. A system for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction, used to execute the method for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in any one of claims 1-7, characterized in that, include: The high-fidelity acquisition module is used to acquire macroscopic images and surface microscopic physical fingerprint features of cultural relics; The feature extraction and comparison module is used to generate deep feature vectors and perform millisecond-level consistency comparisons. The business closed-loop control module is used to manage the outbound verification and return review process and generate status analysis reports; The blockchain evidence storage module is used to hash the verification data and upload it to the blockchain network.
9. The system for constructing a digital evidence chain for cultural relics based on microscopic physical feature extraction as described in claim 8, characterized in that, The high-fidelity acquisition module includes: Large format scanners or medium format cameras are used for gridded area acquisition, with a resolution of ≥600ppi. Color reproduction error ΔE≤2; Microscopic imaging equipment used for 100-200x magnification acquisition, with a single pixel resolution of ≤5 micrometers; X-ray fluorescence spectrometer is used to collect material elemental composition data at typical locations of cultural relics.