File electronic information data recovery method

Abstract

The invention discloses an archive electronic information data recovery method, which relates to the technical field of archive management and data recovery, and comprises the following steps: obtaining original archive data, extracting first data and generating second data according to format characteristics, generating a version sequence of the data object based on the structured metadata, matching the third data and the encrypted data block which is not extracted in the first data as fourth data according to a preset strategy, performing decryption, generating fifth data according to a format feature, if the first data has the unextracted non-encrypted damaged data block, generating supplementary repair data, and if the first data has the unextracted non-encrypted damaged data block, generating the supplementary repair data. The first data, the third data, the fifth data and the repair data are integrated according to the hierarchical identifier, the association between the catalogue and the metadata is reconstructed, a complete file is generated, and a three-in-one system is formed by constructing a structured metadata version sequence, repairing heterogeneous data in a graded manner and fusing block chain dual verification and is suitable for file recovery.

Description

Technical Field

[0001] The present invention relates to the technical field of archive management and data recovery, and in particular to a method for recovering archive electronic information data. Background Art

[0002] In recent years, with the deepening of digital transformation, electronic archives, as the core carrier of information records, have been widely used in scenarios such as judicial file management, government document archiving, and corporate contract storage. The core value of electronic archives lies in their integrity, availability, and security. However, factors such as storage media aging, virus attacks, and human errors have led to frequent problems such as loss of text data byte streams, pixel damage in image scans, and broken metadata associations. Traditional recovery technology relies on single-version storage and extensive repair strategies. It is difficult to effectively manage multi-version history in complex branch editing scenarios, and the repair effect on data with high damage rates is poor. It is difficult to meet the stringent requirements of "original uniqueness" for judicial evidence collection, "long-term readability" for government archives, and "audit traceability" for corporate compliance. Against this background, building an efficient data recovery system that takes into account version management, intelligent repair, and security auditing has become a key technical path to solving the pain points of electronic archive management.

[0003] Existing archival data recovery technologies have many problems that need to be solved urgently. In terms of data integrity and version management, traditional methods rely on single version storage and lack systematic management of historical versions. For example, the early version control system of SVN only supports linear backtracking. It cannot record branch modifications and merge conflicts when multiple users edit concurrently, resulting in low version matching accuracy in complex branch scenarios; heterogeneous data repair strategies are rough, text repair relies only on CRC checksum redundancy, lacks semantic association processing, and increases the format disorder rate; image repair is prone to pixel distortion when the damage rate is high and is not combined with OCR text layer repair; the original metadata is not uniform in format, missing fields, and has no standardized processing process, so the association relationship is often broken after repair; there are obvious defects in permission control and security auditing, key management relies on single-factor authentication of static passwords, and the integration of biometrics and dynamic tokens is insufficient. The decryption logs are stored in the centralized database and are easy to be tampered with; the efficiency of file integration and directory reconstruction is low, and directory reconstruction relies only on "annual-institution- The static hierarchical mapping of "classification level" does not combine version information to quickly locate the latest valid version. Metadata association is limited to physical path mapping and cannot generate dynamic hyperlinks of file association relationships, resulting in long average file retrieval time. Summary of the Invention

[0004] The technical problem solved by the present invention is that the existing archival data recovery technology has many problems that need to be solved urgently. In terms of data integrity and version management, the traditional method relies on single version storage and lacks systematic management of historical versions. For example, the early version control system of SVN only supports linear backtracking. It cannot record branch modifications and merge conflicts when multiple users edit concurrently, resulting in low version matching accuracy in complex branch scenarios; the heterogeneous data repair strategy is extensive, text repair only relies on CRC checksum redundancy, lacks semantic association processing, and increases the format disorder rate; image repair is prone to pixel distortion when the damage rate is high and is not combined with OCR text layer repair; the original metadata is not uniform in format, the field is missing, and there is no standardized processing flow, so the association relationship is often broken after repair; there are obvious defects in permission control and security auditing, key management relies on single-factor authentication of static passwords, the integration of biometrics and dynamic tokens is insufficient, and the decryption log is stored in the centralized database and is easy to be tampered with; the efficiency of file integration and directory reconstruction is low, and directory reconstruction only relies on "annual-institution- The static hierarchical mapping of "classification level" does not combine version information to quickly locate the latest valid version. Metadata association is limited to physical path mapping and cannot generate dynamic hyperlinks of file association relationships, resulting in long average file retrieval time.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: A method for recovering electronic information data of archives comprises the following steps: Step S1, obtaining original archive data, extracting first data and generating second data according to format features of file extension and byte encoding rules; The second data includes a hierarchical identifier and structured metadata; Step S2, generating a version sequence of the data object based on the version number and timestamp of the structured data, and matching the third data according to a preset strategy; Step S3: The encrypted data blocks not extracted from the first data are used as fourth data, and are decrypted, and then fifth data is generated according to the format characteristics of the file extension and byte encoding rules; If there is an unextracted non-encrypted damaged data block in the first data, generating supplementary repair data; Step S4: Integrate the first data, the third data, the fifth data and the repair data according to the level identifier of the second data, and rebuild the directory and metadata association to generate a complete volume.

[0006] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: the original archival data consists of two or more data objects; The first data includes text file data, image scan data and original file metadata; The structured metadata is obtained by unifying the format of the original archive metadata and completing missing fields; The structured metadata includes version number, modification timestamp and file association relationship; The version sequence is generated based on the version number and the modification timestamp; The third data is a valid data object version that has passed verification; The third data includes the complete content of the data object and the corresponding structured metadata snapshot; The complete content of the data object includes binary data of the text file and binary data of the image file; The structured metadata snapshot corresponding to the complete content of the data object includes a version number, a timestamp, an associated file ID, and a content hash value.

[0007] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: generating a version sequence of a data object based on structured metadata and matching the third data according to a preset strategy specifically includes: Based on the version number and modification timestamp in the structured metadata, a linear version timeline and a tree-like version branch structure are constructed; The linear version timeline is arranged in ascending order of modification timestamps to form a complete modification history axis of the data object; The tree-like version branch structure takes the initial version as the root node to form the main branch. When a branch modification occurs in a data object, a version node is generated according to the version number to record the branch source and merge relationship. The branch modification includes simultaneous editing by different users; Each version node includes a structured metadata snapshot and a content hash value, which is used to quickly verify the complete content consistency of the data object; The verification specifically includes: When generating a version node, a hash value is calculated for the complete content of the data object of the current version corresponding to the node and stored in the structured metadata snapshot; During verification, the hash value of the complete content of the data object actually stored in the current system is recalculated and compared with the content hash value recorded in the version node; If the hash values ​​are consistent, it is determined that the complete content of the data object has not been tampered with or damaged; If the hash values ​​are inconsistent, it is determined that the complete content of the data object has been changed or damaged, and a repair process is triggered.

[0008] As a preferred solution of the archival electronic information data recovery method described in the present invention, the application scenarios of the verification specifically include: After repairing an encrypted data block or a damaged non-encrypted data block, the correctness of the repair result is verified by comparing the hash value; When matching the third data according to the preset strategy, filtering out the undamaged version by the hash value; When parallel branches in the tree-like version branch structure are merged, content conflicts are identified by comparing hash values.

[0009] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: the preset strategy matches the third data, specifically including: A priority matching strategy based on version integrity, including selecting the version of the data object that has passed content hash verification and has the latest modification timestamp; Matching strategies based on version relevance include filtering data object versions based on file relevance and version number continuity in structured metadata; Conflict resolution strategies based on branch structure, including trunk branch priority and multi-user editing merge strategies; Special matching strategies based on data types, including text format feature verification and image metadata verification; Abnormal situation handling strategies, including version rollback and hash value conflict verification; The multi-user editing merge strategy includes: Sort by user permission level, giving priority to versions modified by users with higher permissions; If the permissions are the same, merge them by modification timestamp and keep the latest modification content; The abnormal situation handling strategy includes: When all candidate versions are damaged, automatically roll back to the last available version of the data object; When the hash value is the same but the content is different, a second verification is performed; The secondary check verifies the data object by comparing the file size and modification timestamp.

[0010] As a preferred solution of the method for recovering electronic information data of archives described in the present invention, wherein: natural language processing technology is used to identify the encoding format, paragraph structure and font characteristics of the text archive data; Extract resolution, color mode, file format and OCR recognition text layer data from image scan data; The format features include the file extension of the original archive data, byte encoding rules and original archive metadata field definition specifications.

[0011] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: verifying user authority through a key management module, obtaining a corresponding decryption key according to the hierarchical identifier, and decrypting the fourth data specifically includes: The key management module adopts a hierarchical key system, hierarchical identification and decryption keys, which are associated through access control lists; Each of the hierarchical identifiers corresponds to a key authority; The key permissions include browsing permissions, editing permissions, and management permissions; The verification of user rights adopts a double verification mechanism, and the key acquisition process generates an operation log and stores it in encrypted form on the blockchain node; The dual verification mechanism includes biometric authentication and dynamic token verification; The biometric authentication verifies the consistency between the user's identity and the biometric information recorded in the authority database through the biometric collection device; The biometric feature includes fingerprint recognition, facial recognition or iris scanning; The dynamic token verification, the user inputs a dynamic token based on time synchronization or event triggering, and the system verifies whether the token matches the real-time token generated by the server; If the biometric authentication and dynamic token verification are passed at the same time, the decryption key acquisition process of the corresponding hierarchical identification is triggered; During the key acquisition process, an operation log is generated and encrypted and stored through the blockchain node.

[0012] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: if there are unextracted non-encrypted damaged data blocks in the first data, then generating supplementary repair data; The repair data is repaired by a repair algorithm, specifically including: Use byte stream completion algorithm to repair text archive data; The interpolation repair algorithm is used to repair the image scan data; The repair strategy for non-encrypted damaged data blocks is graded based on the degree of damage, including: Directly repair when the damage rate is less than the first threshold; When the damage rate reaches the second threshold and the third threshold, the historical version difference data is called to complete the task; When the damage rate is greater than a fourth threshold, a version sequence matching is triggered for the data object to which the current damaged data block belongs to obtain an alternative version.

[0013] As a preferred solution of the archival electronic information data recovery method of the present invention, wherein: in the integration process of step S4, the hierarchical identifier corresponds to the tree structure of the file directory, and the directory index is rebuilt by a depth-first traversal algorithm; The hierarchical identifier corresponds to the tree structure of the file directory, the root node is the file number, and the version nodes are the year label, institution label and confidentiality level label; The structured metadata association mechanism includes converting the file association relationship in the structured metadata into a directory hyperlink, and the version number and modification timestamp generate the unique storage path of the data object.

[0014] As a preferred solution of the archival electronic information data recovery method of the present invention, after the complete file is generated, the integrity verification process is automatically executed: Recalculate the content hash values ​​of all data objects in the file and compare them with the version node record values. If the verification pass rate falls below the fifth threshold, a secondary repair is triggered. At the same time, a digital fingerprint of the file is generated and matched with the creation fingerprint in the original archive metadata for verification. After verification, a timestamp certificate is added and the file is archived in a read-only storage area.

[0015] The present invention has the following beneficial effects: by constructing a structured metadata-driven version sequence, extracting and standardizing original archival metadata, generating structured metadata containing version numbers, timestamps, and file associations, building a linear timeline and tree-like branching structure, recording the history of multi-user editing branches, and achieving systematic management of data object versions, it meets the version traceability requirements of the National Archives Administration. A hierarchical strategy is designed for heterogeneous data repair, using direct repair, version difference completion, and alternative version matching according to the degree of damage. Text semantic repair, image interpolation, and OCR text-layer association technologies are combined to address formatting issues, pixel distortion, and metadata association breaks. The security system integrates blockchain evidence storage and dual verification, using biometrics and dynamic tokens for dual authentication, binding key permissions to file levels, and real-time on-chain encrypted storage and decryption logs to ensure that audit records cannot be tampered with, meeting network security level protection requirements. The above-mentioned technologies are integrated to form a "version management-intelligent repair-security audit" trinity system, breaking through the bottlenecks of traditional solutions in integrity, efficiency, and security, and providing an innovative full-process path from underlying algorithms to upper-level applications for electronic archive recovery in the judicial and government fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the basic flow of a method for recovering electronic information data of an archive provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0018] Example 1, reference Figure 1 , as one embodiment of the present invention, provides a method for recovering electronic information data of archives, comprising the following steps: Step S1, obtaining original archive data, extracting first data and generating second data according to format features of file extension and byte encoding rules; The second data includes a hierarchical identifier and structured metadata; Step S2, generating a version sequence of the data object based on the version number and timestamp of the structured data, and matching the third data according to a preset strategy; Step S3: The encrypted data blocks not extracted from the first data are used as fourth data, which are decrypted and then fifth data is generated according to the format characteristics of the file extension and byte encoding rules; If there is an unextracted non-encrypted damaged data block in the first data, generating supplementary repair data; Step S4: Integrate the first data, the third data, the fifth data and the repair data according to the level identifier of the second data, and rebuild the directory and metadata association to generate a complete file.

[0019] In one embodiment, the original archival data is first acquired through a data acquisition interface, and textual archival data, image scanned copy data, and original archival metadata are extracted as first data. At the same time, the format of the original archival metadata is unified and missing fields are completed according to the format characteristics of the file extension and byte encoding rules, thereby generating second data including hierarchical identifiers (such as file numbers, year labels) and structured metadata (version numbers, timestamps, and file association relationships). Then, based on the version numbers and timestamps of the structured metadata, a linear version timeline and a tree-like version branch structure are constructed to form a complete version sequence of the data object. Furthermore, a preset strategy is combined with content hash value verification to match the data object from the version sequence. A valid version that has passed verification is used as the third data; then the encrypted data block that has not been extracted in the first data is defined as the fourth data, and the decryption key is obtained after verifying the user authority through the key management module for decryption. After decryption, the fifth data is generated according to the format characteristics. If there is an unextracted non-encrypted damaged data block, the supplementary repair data is generated according to the graded processing according to the degree of damage; finally, based on the hierarchical identification of the second data, the depth-first traversal algorithm is used to reconstruct the file directory index, integrate various types of data hierarchically, and convert the file association relationship in the structured metadata into a directory hyperlink. The unique storage path of the data object is generated through the version number and timestamp, and the directory and metadata association reconstruction is completed to generate a complete file.

[0020] The original archival data consists of two or more data objects; The first data includes text file data, image scan data and original file metadata; Structured metadata is obtained by standardizing the format of original archival metadata and completing missing fields; Structured metadata includes version numbers, modification timestamps, and file associations; The version sequence is generated based on the version number and modification timestamp; The third data is a valid data object version that has passed verification; The third data includes the complete content of the data object and the corresponding structured metadata snapshot; The complete content of the data object, including binary data of text files and binary data of image files; A structured metadata snapshot corresponding to the complete content of the data object. The structured metadata snapshot includes the version number, timestamp, associated file ID, and content hash value.

[0021] In one embodiment, during the implementation process, the original archival data includes two or more data objects, from which textual archival data, image scanned copy data and original archival metadata are extracted as the first data. For the original archival metadata, structured metadata is generated through unified format processing (such as standardized field naming, data type conversion) and missing field completion (such as filling in default values ​​based on business rules or obtaining missing information by association). The metadata includes a version number (identifying the version iteration of the data object), a modification timestamp (recording the version update time) and a file association relationship (marking the master-slave relationship or dependency relationship between data objects). Based on the version number and modification timestamp in the structured metadata, a version sequence of the data objects is generated by chronological arrangement, which intuitively presents the complete modification of the data object. History, through preset verification rules, such as content hash value comparison and format integrity check, each version in the version sequence is verified, and the valid data object version that passes the verification is screened out as the third data. The third data includes two parts. One is the complete content of the data object, that is, the binary data of the text file, retaining the original encoding format and content, and the binary data of the image file, including pixel information and metadata; the other is the corresponding structured metadata snapshot, which records the version number, timestamp (update time accurate to seconds), and associated file ID of the current version, and identifies the unique code and content hash value of the associated data object, and the unique fingerprint generated by calculating the complete content of the data object through the SHA-256 algorithm, which is used to quickly verify data integrity and version association.

[0022] Generate a version sequence of the data object based on the structured metadata and match the third data according to the preset strategy, specifically including: Based on the version number and modification timestamp in the structured metadata, a linear version timeline and a tree-like version branch structure are constructed; A linear version timeline, arranged in ascending order by modification timestamp, forms a complete modification history axis of the data object; The tree-like version branch structure takes the initial version as the root node to form the main branch. When a branch modification occurs in a data object, a version node is generated according to the version number to record the branch source and merge relationship. Branch modifications include simultaneous editing by different users; Each version node includes a structured metadata snapshot and content hash value, which is used to quickly verify the complete content consistency of the data object; The verification specifically includes: When generating a version node, a hash value is calculated for the complete content of the data object of the current version corresponding to the node and stored in the structured metadata snapshot; During verification, the hash value of the complete content of the data object actually stored in the current system is recalculated and compared with the content hash value recorded in the version node; If the hash values ​​are consistent, it is determined that the complete content of the data object has not been tampered with or damaged; If the hash values ​​are inconsistent, it is determined that the complete content of the data object has been changed or damaged, and the repair process is triggered.

[0023] In one embodiment, when the version sequence is generated and matched with the third data, a linear version timeline and a tree-like version branch structure are constructed based on the version number and modification timestamp in the structured metadata: the linear version timeline arranges all versions in ascending order by modification timestamp to form a complete modification history axis of the data object; the tree-like version branch structure forms a trunk branch with the initial version (version number 1.0) as the root node. When multiple users edit simultaneously, that is, when a branch modification occurs, a version node is generated according to the version number rules (such as major version and sub-version) and the branch source and merge relationship are recorded. Each version node includes a structured metadata snapshot and a content hash value. When generating the node, the SHA-256 algorithm is used to calculate the hash value of the complete content of the data object and store it in the snapshot. During verification, the hash value of the actual stored content is recalculated and compared with the recorded value. If they are consistent, the content is determined to be untampered with or damaged. If they are inconsistent, the content is determined to be changed or damaged and a repair process is triggered. This mechanism implements systematic version management through the timeline and branch structure, combines hash value verification to ensure data integrity, and meets the version traceability and conflict resolution requirements in multi-user concurrent editing scenarios.

[0024] Verification application scenarios include: After repairing an encrypted data block or a damaged non-encrypted data block, the correctness of the repair result is verified by comparing the hash value; When matching the third data according to the preset strategy, the undamaged version is filtered out by the hash value; When parallel branches in a tree-like version branch structure are merged, content conflicts are identified by comparing hash values.

[0025] In one embodiment, when implementing the verification mechanism, its application scenarios include: after completing the decryption of an encrypted data block or the repair of a damaged non-encrypted data block, recalculating the content hash value of the repaired data object using the SHA-256 hash algorithm and comparing it with the original hash value recorded in the structured metadata snapshot. If they are consistent, the repair is determined to be correct; if they are inconsistent, a secondary repair process is triggered; when matching the third data (valid version) according to a preset strategy, by traversing the version nodes in the version sequence, the version whose content hash value passes the verification (i.e., the hash value is consistent with the recorded value) is selected to ensure that the complete content of the selected version of the data object has not been tampered with or damaged; parallel branch merging is performed in a tree-like version branch structure. For example, when integrating versions edited by multiple users, the content hash values ​​of the version nodes of the branch to be merged are compared. If the hash values ​​are different, it is determined that there is a content conflict, requiring manual intervention or automatic merge rules, such as retaining the latest modification and resolving the conflict before performing the merge operation. This verification mechanism uses the consistency judgment of hash value comparison to achieve efficient verification of data integrity and version accuracy in repair verification, version screening, and branch merging scenarios.

[0026] The preset strategy matches the third data, specifically including: A priority matching strategy based on version integrity, including selecting the version of the data object that has passed content hash verification and has the latest modification timestamp; Matching strategies based on version relevance include filtering data object versions based on file relevance and version number continuity in structured metadata; Conflict resolution strategies based on branch structure, including trunk branch priority and multi-user editing merge strategies; Special matching strategies based on data types, including text format feature verification and image metadata verification; Abnormal situation handling strategies, including version rollback and hash value conflict verification; Multi-User Editing merge strategies include: Sort by user permission level, giving priority to versions modified by users with higher permissions; If the permissions are the same, merge them by modification timestamp and keep the latest modification content; Strategies for handling exceptions include: When all candidate versions are damaged, automatically roll back to the last available version of the data object; When the hash value is the same but the content is different, a second verification is performed; Secondary verification verifies data objects by comparing file sizes and modification timestamps.

[0027] In one embodiment, when implementing the preset strategy to match the third data, the following core strategies are included: Version integrity priority strategy: Prioritizing the version with a content hash value that passes verification (i.e., has not been tampered with or damaged) and has the latest modification timestamp, to ensure that the latest valid version of the data object is obtained; Version relevance matching strategy: Based on the file association relationship (such as the primary and attached file association) and version number continuity (such as the incremental logic of 1.0→1.1→1.2) in the structured metadata, the system selects versions that are closely associated with the current data object and have a coherent version evolution. Branch structure conflict resolution strategy: The main branch (the main branch derived from the initial version) is the priority for merging. When multiple users edit, the modified version of the user with higher permissions is prioritized (for example, administrators have higher permissions than ordinary users). If the permissions are the same, the latest content is retained according to the modification timestamp. Data type-specific matching strategies: Verify format features (such as encoding format and paragraph structure integrity) for text files, and verify metadata (such as resolution and color mode consistency) for image files; Abnormal situation handling strategy: When all candidate versions are damaged, they will automatically roll back to the previous available version that passes the hash verification; if a rare conflict occurs with the same hash value but different content, a secondary verification will be performed by comparing the file size and modification timestamp to ensure the accuracy of the version matching.

[0028] Through the combination of the above strategies, intelligent screening and conflict resolution of valid versions in complex scenarios can be achieved, ensuring the integrity and accuracy of data object version matching.

[0029] Use natural language processing technology to identify encoding formats, paragraph structures, and font features for text archive data; Extract resolution, color mode, file format and OCR recognition text layer data from image scan data; Format features include the file extension of the original archive data, byte encoding rules, and original archive metadata field definition specifications.

[0030] In one embodiment, for text archive data, natural language processing based technology is used to perform format analysis: by matching character frequency statistics with encoding features, such as detecting the BOM header of UTF-8 and GBK The system can identify the double-byte character distribution of the document, recognize the file encoding format, parse the paragraph structure and hierarchical relationship based on text delimiter rules (such as line breaks, consecutive blank lines) and hierarchical markers (such as the title prefix "Chapter 1", bold font style), directly extract the document's built-in font metadata (such as the font type field in the .doc file) or identify the font type (such as Songti, Heiti) and font size through the image pixel distribution pattern (such as font stroke width, serif characteristics); for image scan data, extract the underlying metadata of resolution (such as 300DPI), color mode (such as RGB, CMYK), and file format (such as JPEG, PNG) through image processing algorithms, use mature OCR tools (such as Tesseract, Baidu OCR engine) to generate text layer data, and set OCR recognition confidence verification (the threshold is set to 90%, and a value below this value triggers the manual review process); the format characteristics specifically include the file extension of the original archive data (such as .doc and .jpg), byte encoding rules (such as the byte order mark of UTF-8), and the original archive metadata field definition specifications, such as the timestamp using ISO 8601 format, ensuring format compatibility and feature integrity of different types of data through standardized parsing processes.

[0031] Verifying user authority through the key management module, obtaining the corresponding decryption key according to the level identifier, and decrypting the fourth data specifically includes: The key management module adopts a hierarchical key system, hierarchical identification and decryption keys, which are associated through access control lists; Each level identifier corresponds to the key authority; Key permissions include browsing permissions, editing permissions, and management permissions; User rights are verified using a double-check mechanism. The key acquisition process generates an operation log and stores it encrypted in the blockchain node. The dual verification mechanism includes biometric authentication and dynamic token verification; Biometric authentication, which uses biometric collection equipment to verify the consistency of user identity and biometric information recorded in the authority database; Biometrics include fingerprint recognition, facial recognition, or iris scanning; Dynamic token verification: the user enters a dynamic token based on time synchronization or event triggering, and the system verifies whether the token matches the real-time token generated by the server; If both the biometric authentication and the dynamic token verification are passed, the decryption key acquisition process for the corresponding level identification is triggered; Operation logs are generated during the key acquisition process and stored encrypted through blockchain nodes.

[0032] In one embodiment, when implementing key management and authority verification, a hierarchical key system is constructed through a key management module: the hierarchical identifier (such as file confidentiality level, departmental authority) and the decryption key are associated through an access control list (ACL), and each level corresponds to three types of key permissions: browsing, editing, and management. User authority verification adopts a double verification mechanism: first, the consistency of the user's biometrics (fingerprint, face or iris information) and the records in the authority database are verified through a biometric collection device (such as a fingerprint sensor, face recognition camera). At the same time, the user enters a dynamic token based on time synchronization (such as the TOTP algorithm) or event triggering, and the system matches the user token with the token generated in real time by the server; only when the biometric authentication and dynamic token verification are passed at the same time, the decryption key acquisition process for the corresponding hierarchical identifier is triggered, and the operation log generated during the key acquisition process (including user ID, operation time and key level) is encrypted and stored through the blockchain node. The tamper-proof nature of the blockchain is used to ensure the security and traceability of the operation records, thereby realizing hierarchical authority control and security audit of the decryption of encrypted data blocks.

[0033] If there is an unextracted non-encrypted damaged data block in the first data, generating supplementary repair data; Repair data is repaired using a repair algorithm, including: Use byte stream completion algorithm to repair text archive data; The interpolation repair algorithm is used to repair the image scan data; The repair strategy for non-encrypted damaged data blocks is graded based on the degree of damage, including: Directly repair when the damage rate is less than the first threshold; When the damage rate reaches the second threshold and the third threshold, the historical version difference data is called to complete the task; When the damage rate is greater than a fourth threshold, a version sequence matching is triggered for the data object to which the current damaged data block belongs to obtain an alternative version.

[0034] In one embodiment, when repairing a non-encrypted damaged data block, if it is detected that there is an unextracted damaged data block in the first data, the data block damage rate (i.e., the ratio of the number of damaged bytes to the total number of bytes in the data block) is first calculated, and the damage is graded and processed according to the degree of damage: when the damage rate is less than 10% (a first threshold, based on the conventional standard for determining minor damage in the field of data repair), a byte stream completion algorithm is used for text archive data, such as directly repairing missing bytes by pattern matching of adjacent byte sequences, and bilinear interpolation or neighborhood pixel replication algorithm is used for image scan data to repair local damage; when the damage rate is between 10% and 30% (a second threshold to a third threshold, which is verified to be within the range of version difference), the damage is repaired. When there is a damage rate greater than 30% (the fourth threshold, defined as severe damage with a direct repair success rate of less than 50%), the version sequence matching mechanism is triggered, and the latest valid alternative version of the same type of data object is screened according to the version number and timestamp in the structured metadata to overwrite the current damaged data block. Through the above-mentioned hierarchical repair strategy, combined with text byte stream completion and image interpolation repair technology, efficient repair of data with different degrees of damage can be achieved, thereby ensuring the integrity and availability of electronic archive data.

[0035] During the integration process of step S4, the hierarchical identifiers correspond to the tree structure of the volume directory, and the directory index is rebuilt using a depth-first traversal algorithm; The hierarchical identification corresponds to the tree structure of the file directory, with the root node being the file number and the version nodes being the year label, organization label, and confidentiality level label; The structured metadata association mechanism includes converting the file association relationship in the structured metadata into a directory hyperlink, and the version number and modification timestamp generate the unique storage path of the data object.

[0036] In one of the embodiments, when implementing data integration and directory reconstruction, a tree structure of the file directory is constructed based on hierarchical identification: with the file number as the root node, the annual label (such as "2023"), institutional label (such as "Finance Department"), and confidentiality label (such as "Confidential") are used as version nodes to expand step by step to form a multi-level directory system. By traversing nodes at each level through a depth-first traversal algorithm and recursively generating directory index files, the integrity and traceability of the file structure are ensured, and the structured metadata association mechanism is executed simultaneously: the file association relationships recorded in the structured metadata, such as the subordinate relationship between the main file and the attachment, are converted into directory hyperlinks, and users can jump to the associated file by clicking on the directory node; the version number (such as "1.0") and the modification timestamp (a time series accurate to the second) are combined to generate a unique storage path for the data object (the format is "file number / year / institution / classification level / version number-timestamp. Extension") to ensure that each version of the data object has a unique identifier in the storage system, and the first data, third data, fifth data and repair data are integrated according to the hierarchical structure to complete the association reconstruction of the file directory and metadata, and finally generate a logically clear and quickly searchable electronic archive file.

[0037] After the complete dossier is generated, the integrity verification process is automatically executed: Recalculate the content hash values ​​of all data objects in the file and compare them with the version node record values. If the verification pass rate falls below the fifth threshold, a secondary repair is triggered. At the same time, a digital fingerprint of the file is generated and matched with the creation fingerprint in the original archive metadata for verification. After verification, a timestamp certificate is added and the file is archived in a read-only storage area.

[0038] In one embodiment, after a complete file is generated, the system automatically triggers an integrity verification process: first, the SHA-256 hash value of the complete content of all data objects in the file (including text binary data and image binary data) is recalculated and compared one by one with the hash value recorded in the structured metadata snapshot of the corresponding version node. If the verification pass rate is lower than 95% (the fifth threshold is preset, based on the high requirements for data integrity in the archive management industry), it is determined that there is a risk of batch data damage, and a secondary repair process is automatically triggered (calling historical version difference data or alternative versions for completion). At the same time, a unique digital fingerprint is calculated for the entire file data using the SHA-256 algorithm, and matched and verified with the creation fingerprint recorded in the original archive metadata. After verification, the timestamp server adds a timestamp certificate that complies with the RFC 3161 standard to the file, and migrates the file to a read-only storage area (such as a CD library or a cloud storage read-only bucket) for archiving to ensure that the file data cannot be tampered with and is traceable.

[0039] This invention constructs a structured metadata-driven version sequence, extracts and standardizes original archival metadata, generates structured metadata containing version numbers, timestamps, and file associations, builds a linear timeline and tree-like branching structure, records the history of multi-user editing branches, and implements systematic management of data object versions to meet the version traceability requirements of the National Archives Administration. A hierarchical strategy is designed for heterogeneous data repair, using direct repair, version difference completion, and alternative version matching based on the degree of damage. Text semantic repair, image interpolation, and OCR text-layer association techniques are combined to address formatting errors, pixel distortion, and metadata association breaks. The security system integrates blockchain evidence storage and dual verification. Through dual authentication using biometrics and dynamic tokens, key permissions are bound to the file level, and decryption logs are encrypted and stored on the chain in real time to ensure that audit records cannot be tampered with, meeting network security level protection requirements. The integration of these technologies creates a three-in-one system of "version management, intelligent repair, and security auditing," overcoming the integrity, efficiency, and security bottlenecks of traditional solutions. This provides an innovative, full-process approach for electronic archive recovery in the judicial and government sectors, from underlying algorithms to upper-layer applications.

[0040] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0041] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit 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 spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for recovering electronic information data of archives, characterized in that: The following steps are involved: Step S1, obtaining original archive data, extracting first data and generating second data according to format features of file extension and byte encoding rules; The second data includes a hierarchical identifier and structured metadata; Step S2, generating a version sequence of the data object based on the version number and timestamp of the structured data, and matching the third data according to a preset strategy; Step S3: The encrypted data blocks not extracted from the first data are used as fourth data, and are decrypted, and then fifth data is generated according to the format characteristics of the file extension and byte encoding rules; If there is an unextracted non-encrypted damaged data block in the first data, generating supplementary repair data; Step S4: Integrate the first data, the third data, the fifth data and the repair data according to the level identifier of the second data, and rebuild the directory and metadata association to generate a complete volume.

2. The method for recovering electronic archive information data according to claim 1, wherein: The original archival data consists of two or more data objects; The first data includes text file data, image scan data and original file metadata; The structured metadata is obtained by unifying the format of the original archive metadata and completing missing fields; The structured metadata includes version number, modification timestamp and file association relationship; The version sequence is generated based on the version number and the modification timestamp; The third data is a valid data object version that has passed verification; The third data includes the complete content of the data object and the corresponding structured metadata snapshot; The complete content of the data object includes binary data of the text file and binary data of the image file; The structured metadata snapshot corresponding to the complete content of the data object includes a version number, a timestamp, an associated file ID, and a content hash value.

3. The method for recovering electronic archive information data according to claim 2, wherein: Generating a version sequence of the data object based on the structured metadata and matching the third data according to a preset strategy specifically includes: Based on the version number and modification timestamp in the structured metadata, a linear version timeline and a tree-like version branch structure are constructed; The linear version timeline is arranged in ascending order of modification timestamps to form a complete modification history axis of the data object; The tree-like version branch structure takes the initial version as the root node to form the main branch. When a branch modification occurs in a data object, a version node is generated according to the version number to record the branch source and merge relationship. The branch modification includes simultaneous editing by different users; Each version node includes a structured metadata snapshot and a content hash value, which is used to quickly verify the complete content consistency of the data object; The verification specifically includes: When generating a version node, a hash value is calculated for the complete content of the data object of the current version corresponding to the node and stored in the structured metadata snapshot; During verification, the hash value of the complete content of the data object actually stored in the current system is recalculated and compared with the content hash value recorded in the version node; If the hash values ​​are consistent, it is determined that the complete content of the data object has not been tampered with or damaged; If the hash values ​​are inconsistent, it is determined that the complete content of the data object has been changed or damaged, and a repair process is triggered.

4. The method for recovering electronic archive information data according to claim 3, wherein: The application scenarios of the verification specifically include: After repairing an encrypted data block or a damaged non-encrypted data block, the correctness of the repair result is verified by comparing the hash value; When matching the third data according to the preset strategy, filtering out the undamaged version by the hash value; When parallel branches in the tree-like version branch structure are merged, content conflicts are identified by comparing hash values.

5. The method for recovering electronic archive information data according to claim 4, wherein: The preset strategy matches the third data, specifically including: A priority matching strategy based on version integrity, including selecting the version of the data object that has passed content hash verification and has the latest modification timestamp; Matching strategies based on version relevance include filtering data object versions based on file relevance and version number continuity in structured metadata; Conflict resolution strategies based on branch structure, including trunk branch priority and multi-user editing merge strategies; Special matching strategies based on data types, including text format feature verification and image metadata verification; Abnormal situation handling strategies, including version rollback and hash value conflict verification; The multi-user editing merge strategy includes: Sort by user permission level, giving priority to versions modified by users with higher permissions; If the permissions are the same, merge them by modification timestamp and keep the latest modification content; The abnormal situation handling strategy includes: When all candidate versions are damaged, automatically roll back to the last available version of the data object; When the hash value is the same but the content is different, a second verification is performed; The secondary check verifies the data object by comparing the file size and modification timestamp.

6. The method for recovering electronic archive information data according to claim 5, wherein: Using natural language processing technology to identify encoding format, paragraph structure and font features of the text file data; Extract resolution, color mode, file format and OCR recognition text layer data from image scan data; The format features include the file extension of the original archive data, byte encoding rules and original archive metadata field definition specifications.

7. The method for recovering electronic archive information data according to claim 6, wherein: Verifying user authority through a key management module, obtaining a corresponding decryption key according to the hierarchical identifier, and decrypting the fourth data specifically includes: The key management module adopts a hierarchical key system, hierarchical identification and decryption keys, which are associated through access control lists; Each of the hierarchical identifiers corresponds to a key authority; The key permissions include browsing permissions, editing permissions, and management permissions; The verification of user rights adopts a double verification mechanism, and the key acquisition process generates an operation log and stores it in encrypted form on the blockchain node; The dual verification mechanism includes biometric authentication and dynamic token verification; The biometric authentication verifies the consistency between the user's identity and the biometric information recorded in the authority database through the biometric collection device; The biometric feature includes fingerprint recognition, facial recognition or iris scanning; The dynamic token verification, the user inputs a dynamic token based on time synchronization or event triggering, and the system verifies whether the token matches the real-time token generated by the server; If the biometric authentication and dynamic token verification are passed at the same time, the decryption key acquisition process of the corresponding hierarchical identification is triggered; During the key acquisition process, an operation log is generated and encrypted and stored through the blockchain node.

8. The method for recovering electronic archive information data according to claim 7, wherein: If there is an unextracted non-encrypted damaged data block in the first data, generating supplementary repair data; The repair data is repaired by a repair algorithm, specifically including: Use byte stream completion algorithm to repair text archive data; The interpolation repair algorithm is used to repair the image scan data; The repair strategy for non-encrypted damaged data blocks is graded based on the degree of damage, including: Directly repair when the damage rate is less than the first threshold; When the damage rate reaches the second threshold and the third threshold, the historical version difference data is called to complete the task; When the damage rate is greater than a fourth threshold, a version sequence matching is triggered for the data object to which the current damaged data block belongs to obtain an alternative version.

9. The method for recovering electronic archive information data according to claim 8, wherein: During the integration process of step S4, the hierarchical identifiers correspond to the tree structure of the volume directory, and the directory index is rebuilt using a depth-first traversal algorithm; The hierarchical identifier corresponds to the tree structure of the file directory, the root node is the file number, and the version nodes are the year label, institution label and confidentiality level label; The structured metadata association mechanism includes converting the file association relationship in the structured metadata into a directory hyperlink, and the version number and modification timestamp generate the unique storage path of the data object.

10. The method for recovering electronic archive information data according to claim 9, wherein: After the complete file is generated, the integrity verification process is automatically executed: Recalculate the content hash values ​​of all data objects in the file and compare them with the version node record values. If the verification pass rate falls below the fifth threshold, a secondary repair is triggered. At the same time, a digital fingerprint of the file is generated and matched with the creation fingerprint in the original archive metadata for verification. After verification, a timestamp certificate is added and the file is archived in a read-only storage area.

Images