School uniform production full-chain blockchain quality traceability method and system

Abstract

The present application relates to school uniform production quality traceability technical field, especially in kind of school uniform production whole chain block chain quality traceability method and system. Obtain school uniform production link process data, form time sequence evidence chain based on trusted execution environment and carry out hash encapsulation, hash code of fabric batch identification is not copied physical identification and is applied to production batch and sample for examination simultaneously, and batch binding certificate is established. Based on standard knowledge graph, the quality inspection report is automatically analyzed to generate the verification conclusion, the time sequence evidence chain, the batch binding certificate and the verification conclusion are submitted to the alliance block chain consensus verification to generate the evidence storage transaction identification, the identification is retrieved and analyzed in response to the traceability request, the quality inspection report is matched and confirmed to the corresponding production batch through the physical identification characteristic code and the chain certificate, the verification result and the whole link quality data are returned, the school uniform quality credible traceability and the anti-tamper verification are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of school uniform production quality traceability, and particularly relates to a full-chain blockchain quality traceability method and system for school uniform production. Background Art

[0002] Conventional school uniform production quality traceability usually relies on internal paper records of enterprises or stand-alone information management systems, and quality data of each production link are entered and stored separately by different departments. Data such as fabric procurement, production and processing, and quality inspection reports are scattered in multiple independent systems, and there is no effective association and verification mechanism between them. Although some enterprises have adopted barcodes or two-dimensional codes for batch management, the coded information only includes basic production dates and batch numbers, and does not involve a complete quality data chain. The quality inspection link usually only issues test reports for submitted samples, and the correspondence between samples and production batches is maintained through manual registration or simple labels. This connection mode is prone to information breakage or artificial replacement during the circulation process.

[0003] The process data of existing traceability methods are managed by each link itself, and there is no anti-tamper solidification means. When relevant parties conduct traceability, they can only trust the unilateral records provided by the enterprise. In particular, school uniform fabrics involve safety indicators such as formaldehyde content and color fastness, and the reliability of inspection data is directly related to the health of students. Once a quality dispute arises, these records can hardly be used as objective evidence, and consumers and regulatory authorities cannot confirm the authenticity and completeness of the data, resulting in traceability results that are often mere formalities.

[0004] Although the submitted samples and production batches may use the same brand or model, there is a lack of unique and non-transferable identifier binding, and there is doubt as to whether the samples targeted by the test report truly come from the corresponding production batch. As a result, even if the quality inspection report is qualified, it cannot effectively prove the quality status of the entire batch of school uniforms, leaving hidden dangers for unqualified products to enter the market. Summary of the Invention

[0005] The embodiments of the present invention provide a full-chain blockchain quality traceability method and system for school uniform production, which can solve the problems in the prior art.

[0006] In a first aspect of the embodiments of the present invention, a full-chain blockchain quality traceability method for school uniform production is provided, comprising:

[0007] acquiring process data at each link of the school uniform production chain;

[0008] performing hash encapsulation on the process data based on a trusted execution environment, wherein the hash value of the current process node is obtained by combined operation of the hash value of the previous node and the current data, so as to form a time-sequential evidence chain;

[0009] Before the fabric is sent for inspection, the batch identifier hash value of the fabric is encoded as a non-replicable physical identifier and applied simultaneously to both the production batch of fabric and the sample submitted for inspection, establishing a physical hierarchical binding between the quality inspection sample and the production batch, thus obtaining a batch binding certificate;

[0010] Based on a standard knowledge graph, the system automatically analyzes and compares the detection indicators of the quality inspection report data against the limit values, and generates verification conclusions.

[0011] The chronological evidence chain, batch binding certificate, and verification conclusion are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier.

[0012] In response to traceability requests, the system retrieves and parses the notarized transaction identifier from the consortium blockchain, verifies batch consistency by reading the physical identifier of the school uniform, and confirms that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and returns the verification result and full-chain quality data.

[0013] The process data is hashed and encapsulated based on a trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data, forming a time-series evidence chain, including:

[0014] A trusted execution environment module is integrated into the data acquisition equipment at the production site. The trusted execution environment module includes an isolated execution region and a hardware root of trust.

[0015] Within the isolated execution area, process data of the current process is received, and the hash value of the preceding process node is extracted as the parent node identifier;

[0016] The parent node identifier is concatenated with the process data of the current process, and the hash value of the current process node is calculated using a hash function. The hash value of the current process node is then digitally signed by the hardware root of trust.

[0017] The data structure containing the parent node identifier, the hash value of the current process node, and the digital signature is output as the current node record, and the hash value of the current process node in the current node record is used as the parent node identifier of the next process node;

[0018] Multiple node records are generated sequentially according to the process sequence, forming a chain-like data structure with sequential dependencies, thus constituting a time-series evidence chain.

[0019] Before the fabric is sent for inspection, the batch identifier hash value is encoded as a non-replicable physical identifier and simultaneously applied to both the production batch and the submitted sample. This establishes a physical hierarchical binding between the quality inspection sample and the production batch, resulting in a batch binding certificate, including:

[0020] Extract the batch hash value corresponding to the fabric batch identifier from the temporal evidence chain;

[0021] A unique sub-identifier is generated for each individual piece of fabric in the corresponding production batch and the sample to be tested. The sub-identifier is obtained by hashing the batch hash value and the randomly generated individual salt value. The sub-identifier is then encoded to generate a graphic code.

[0022] The graphic code is carried by an identification material that irreversibly bonds with the fabric substrate. During the application process, the identification material chemically bonds with the fabric fibers or causes microstructural modification of the fibers, leaving irreversible physical damage characteristics on the fabric surface after the identification is removed.

[0023] When submitting fabrics for inspection, the identification material bearing the graphic code is simultaneously applied to the same area of ​​both the sample to be inspected and the finished fabric of the corresponding production batch.

[0024] The identification device reads the markings applied to the submitted sample and extracts the sub-markers from the graphic code;

[0025] The batch hash value, sub-identifier, individual salt value and the submitted sample identifier are associated and encapsulated, and the mapping relationship table between the batch hash value and all sub-identifiers in the batch is recorded on the consortium blockchain to generate a batch binding certificate.

[0026] The graphic code is carried by an identification material that irreversibly bonds with the fabric substrate. During application, the identification material chemically bonds with the fabric fibers or causes modification of the fiber's microstructure, leaving irreversible physical damage characteristics on the fabric surface after the identification is removed, including:

[0027] The labeling material is selected to contain reactive groups that can covalently bond with the functional groups on the surface of the fabric fibers;

[0028] During the application of the marking material, the reactive groups are activated by energy excitation, causing the marking material to react chemically with the functional groups on the surface of the fabric fibers to form covalent bonds.

[0029] During the chemical bonding process, the physical response characteristics of the interface between the marker material and the fabric substrate are simultaneously collected. These physical response characteristics include optical reflection characteristics or electromagnetic response characteristics of the interface.

[0030] Feature extraction is performed on the physical response characteristics to generate a combined interface fingerprint. The combined interface fingerprint is determined by the randomness of the distribution of the identifying material and the randomness of the fabric fiber microstructure, and has individual uniqueness.

[0031] The interface fingerprint is associated with the corresponding sub-identifier and recorded in the batch binding certificate;

[0032] During traceability verification, the graphic code of the physical identifier is read to obtain the sub-identifier, and the physical response characteristics of the area surrounding the identifier are simultaneously collected and compared with the interface fingerprint in the batch binding certificate;

[0033] When the identifier is removed or replaced, the physical response characteristics of the bonding interface are irreversibly changed due to the breakage of covalent bonds. This causes the currently collected physical response characteristics to not match the bonding interface fingerprint recorded in the batch binding certificate, thus determining that the identifier has been tampered with.

[0034] Based on a standard knowledge graph, the system automatically parses and compares the detection indicators of the quality inspection report data against the limits, generating verification conclusions, including:

[0035] A standard knowledge graph is constructed, which organizes testing indicators, national standard limits, certification body qualifications, hazardous substance lists, batch sensitivity attributes, and process causal rules in the form of nodes and related edges;

[0036] Extract the names and values ​​of testing indicators from the quality inspection report data;

[0037] The batch sensitivity attributes corresponding to the detection indicators are queried from the standard knowledge graph. Batch correlation indicators whose detection values ​​are affected by production process parameters are filtered out from the batch sensitivity attributes. The process causal rules corresponding to the batch correlation indicators are obtained. The process causal rules describe the causal constraint relationship between the detection values ​​and the production process parameters.

[0038] Extract the production process parameters corresponding to the batch in the quality inspection report from the chronological evidence chain;

[0039] According to the process causality rule, the detection values ​​of batch-related indicators are checked for causal consistency with the production process parameters of the corresponding batch. When the detection value exceeds the preset value range corresponding to the process parameters of the batch, the quality inspection report is marked as having a logical conflict with the batch.

[0040] For non-batch-related indicators, compare them against national standard limits and verify the qualifications of certification bodies;

[0041] When there is a logical conflict between batch-related indicators, it is judged as batch mismatch; when there is a standard exceeding the limit for non-batch-related indicators, it is judged as quality non-compliance; when the testing agency exceeds the authorized scope or prohibited substances are detected, it is judged as compliance abnormality. A verification conclusion is generated that includes batch matching conclusion, quality compliance conclusion, and compliance conclusion.

[0042] The chronological evidence chain, batch binding credentials, and verification conclusions are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier, including:

[0043] The chronological evidence chain, batch binding credentials, and verification conclusions are structured and encapsulated to generate an evidence storage data package;

[0044] Data digests are calculated for the chronological evidence chain, batch binding credential, and verification conclusion in the evidence storage data packet. The three data digests are then concatenated in chronological order, and the overall hash value is calculated as the evidence storage fingerprint.

[0045] The evidence storage data packet and evidence storage fingerprint are encapsulated into an evidence storage transaction request and submitted to the consensus node of the consortium blockchain network;

[0046] Consensus nodes verify the notarization transaction request. Verification includes checking the consistency between the notarization fingerprint and the notarization data packet, as well as verifying the identity and permissions of the submitting node. Once the verification is successful, the notarization data packet is written into the candidate block, and consensus voting is conducted on the candidate block. When the candidate block receives confirmation from more than a preset proportion of consensus nodes in the consortium blockchain network, the candidate block is added to the blockchain ledger.

[0047] Extract the transaction record identifier containing the evidence storage data packet from the newly added block as the evidence storage transaction identifier, and associate it with the corresponding fabric batch identifier for subsequent querying of the corresponding blockchain evidence storage transaction identifier through the fabric batch identifier.

[0048] In response to traceability requests, the system retrieves and parses the notified transaction identifier from the consortium blockchain. It verifies batch consistency by reading the physical identifier of the school uniform. When the physical identifier's feature code matches the batch binding certificate on the blockchain, it confirms that the quality inspection report truly corresponds to the production batch. The system then returns the verification results and end-to-end quality data, including:

[0049] Receive a traceability request, which includes the physical identifier or fabric batch identifier of the school uniform to be verified;

[0050] The sub-identifier is obtained by parsing the graphic code on the school uniform using an identification device, or by directly obtaining the fabric batch identifier. The corresponding notarized transaction identifier is then retrieved from the consortium blockchain based on the sub-identifier or the fabric batch identifier.

[0051] Based on the evidence storage transaction identifier, evidence storage data packets are extracted from the consortium blockchain ledger, and the time-series evidence chain, batch binding certificate, and verification conclusion are obtained by parsing the evidence storage data packets;

[0052] Extract the sub-identifier mapping table for the batch from the batch binding certificate, and determine whether the read sub-identifier exists in the mapping table. If the sub-identifier exists in the mapping table, it is determined that the physical identifier matches the batch binding certificate, confirming that the quality inspection report of the school uniform truly corresponds to the production batch. If the sub-identifier does not exist, it is determined that there is a risk of batch replacement.

[0053] The batch matching determination results are correlated with the verification conclusions to generate verification results that include both batch consistency determination and quality compliance determination.

[0054] Process data for each production step in this batch is extracted from the chronological evidence chain and combined with the quality inspection report data in the verification conclusion to generate end-to-end quality data;

[0055] The verification results and end-to-end quality data are returned to the traceability request initiator.

[0056] A second aspect of this invention provides a blockchain-based quality traceability system for the entire school uniform production chain, comprising:

[0057] The data acquisition unit is used to acquire process data at various stages of the school uniform production chain.

[0058] The hash encapsulation unit is used to perform hash encapsulation on the process data based on the trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data to form a time-series evidence chain.

[0059] The batch binding unit is used to encode the fabric batch identifier hash value into an uncopyable physical identifier and apply it simultaneously to the production batch fabric and the submitted sample before the fabric is sent for inspection, thereby establishing a physical hierarchical binding between the quality inspection sample and the production batch and obtaining a batch binding certificate.

[0060] The indicator comparison unit is used to automatically parse and compare the detection indicators of the quality inspection report data with the limit based on the standard knowledge graph, and generate verification conclusions.

[0061] The consensus-based evidence storage unit is used to submit the time-series evidence chain, batch binding certificate and verification conclusion to the consortium blockchain network for consensus verification and generate a storage transaction identifier.

[0062] The traceability verification unit is used to respond to traceability requests, retrieve and parse the evidence transaction identifier from the consortium blockchain, verify batch consistency by reading the physical identifier of the school uniform, and confirm that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and return the verification result and full-chain quality data.

[0063] A third aspect of the present invention provides an electronic device, comprising:

[0064] processor;

[0065] Memory used to store processor-executable instructions;

[0066] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0067] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0068] This invention constructs a trusted traceability mechanism covering the entire production chain. By using a trusted execution environment to hash and encapsulate process data, the data at each process node is closely linked to its predecessor, forming an irreversible chain of chronological evidence. Any data tampering at any stage will be detected by the on-chain data logic, effectively preventing internal data falsification and significantly improving the transparency and auditability of school uniform production.

[0069] Before fabric is sent for testing, the batch identifier is hashed into a unique physical identifier and simultaneously applied to both the production batch and the submitted sample, achieving a physical-level binding between the quality inspection sample and the production batch. During traceability, by reading the physical identifier of the school uniform and comparing the feature code with the batch binding certificate on the blockchain, the correspondence between the quality inspection report and the production batch can be accurately confirmed, avoiding problems such as inconsistencies between samples and actual production batches, and misattribution of quality inspection reports, ensuring the authenticity and reliability of the quality data for each batch of school uniforms.

[0070] By utilizing standard knowledge graphs to automatically parse and compare quality inspection reports against limits, verification conclusions can be generated quickly without manual item-by-item review, significantly improving the efficiency of quality inspection data processing. Combined with the consensus verification of the consortium blockchain network, all key evidence is consistently confirmed among multiple nodes, generating an immutable notarized transaction identifier, thus providing technical assurance for the authority and credibility of quality data.

[0071] When responding to traceability requests, the system retrieves and parses the evidence transaction identifier from the consortium blockchain, combines it with the physical identifier to complete batch consistency verification, and can quickly return end-to-end quality data and verification results. This provides a convenient and reliable verification method for regulatory authorities, schools, and parents, effectively supports the supervision of school uniform quality and safety and consumer rights protection, reduces corporate compliance costs, and enhances brand reputation. Attached Figure Description

[0072] Figure 1 A flowchart illustrating the blockchain-based quality traceability method for the entire school uniform production chain;

[0073] Figure 2 This is a flowchart and example diagram of an intelligent verification process for quality inspection reports based on a standard knowledge graph. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0076] Figure 1 This is a flowchart illustrating the blockchain-based quality traceability method for the entire school uniform production chain, as described in this embodiment of the invention.

[0077] The blockchain-based quality traceability method for the entire school uniform production chain includes:

[0078] Acquire process data at each stage of the school uniform production chain;

[0079] The process data is hash-encapsulated based on a trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data to form a time-series evidence chain.

[0080] Before the fabric is sent for inspection, the batch identifier hash value of the fabric is encoded as a non-replicable physical identifier and applied simultaneously to both the production batch of fabric and the sample submitted for inspection, establishing a physical hierarchical binding between the quality inspection sample and the production batch, thus obtaining a batch binding certificate;

[0081] Based on a standard knowledge graph, the system automatically analyzes and compares the detection indicators of the quality inspection report data against the limit values, and generates verification conclusions.

[0082] The chronological evidence chain, batch binding certificate, and verification conclusion are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier.

[0083] In response to traceability requests, the system retrieves and parses the notarized transaction identifier from the consortium blockchain, verifies batch consistency by reading the physical identifier of the school uniform, and confirms that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and returns the verification result and full-chain quality data.

[0084] The process data is hashed and encapsulated based on a trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data, forming a time-series evidence chain, including:

[0085] A trusted execution environment module is integrated into the data acquisition equipment at the production site. The trusted execution environment module includes an isolated execution region and a hardware root of trust.

[0086] Within the isolated execution area, process data of the current process is received, and the hash value of the preceding process node is extracted as the parent node identifier;

[0087] The parent node identifier is concatenated with the process data of the current process, and the hash value of the current process node is calculated using a hash function. The hash value of the current process node is then digitally signed by the hardware root of trust.

[0088] The data structure containing the parent node identifier, the hash value of the current process node, and the digital signature is output as the current node record, and the hash value of the current process node in the current node record is used as the parent node identifier of the next process node;

[0089] Multiple node records are generated sequentially according to the process sequence, forming a chain-like data structure with sequential dependencies, thus constituting a time-series evidence chain.

[0090] At each stage of the school uniform production process, data acquisition devices are responsible for sensing and reporting raw process data. To ensure that this data possesses tamper-proof and trustworthy attributes during the acquisition phase, a trusted execution environment module is integrated into each data acquisition device. This module divides the operation into two independent operating areas at the hardware level: an isolated execution area and a normal execution area. The isolated execution area operates under the protection of the processor's security extension mechanism, preventing external software from directly accessing or modifying its code and data. The hardware root of trust exists in the form of a chip-embedded key, providing hardware-level endorsement for the computation results generated in the isolated execution area, ensuring that any forgery or tampering can be identified during the verification phase.

[0091] After the process data enters the isolated execution area, it first reads the hash value generated by the preceding process node and uses it as the parent node identifier. For the initial process of the entire production chain, since there are no preceding nodes, a pre-agreed genesis value is used as the... This genesis value is jointly confirmed by all participating parties and solidified in a smart contract during the deployment of the consortium blockchain network, ensuring the uniqueness and traceability of the starting point of the chain.

[0092] The process data generated in the current process is recorded as follows: Its content includes process type identifiers, operation timestamps, equipment numbers, operator identifiers, and process-specific quality parameters, such as fabric tension values ​​in the cutting process, stitch density in the sewing process, and temperature and pressure parameters in the ironing process. The parent node identifier is concatenated byte-by-byte with the current process data to form the input string to be hashed. The hash value of the current process node is then calculated using a hash function. ,Right now

[0093]

[0094] in This indicates a byte string concatenation operation. This is a hash algorithm implementation that is fixed within the isolated execution area. The fundamental reason for adopting this chained computation method is that any modification to historical process data will cause its node hash value to change, thereby causing the hash values ​​of all subsequent nodes to become invalid in a chain reaction. Attackers cannot partially tamper with historical records without being detected.

[0095] After the calculation is complete, the hardware root of trust uses its built-in private key pair. Perform digital signature and generate signature value The private key of the hardware root of trust is written during the chip manufacturing stage and never leaves the chip in plaintext throughout its entire lifecycle; the signing operation is completed entirely within the chip. External verifiers hold the public key corresponding to this private key and can verify the signature at any time, thereby confirming its validity. It is indeed generated by a certified trusted execution environment module, rather than being forged by a regular software environment.

[0096] Parent node identifier The hash value of the current process node Digital signature and process data The necessary summary information is encapsulated together into a structured node record. . As a complete credential in the current process's chronological evidence chain, it is output outside the isolated execution area for subsequent on-chain evidence storage. Meanwhile, The data acquisition device passed to the next process serves as the parent node identifier required by that process to calculate its node hash value, thereby establishing hash dependencies between adjacent nodes.

[0097] In actual production processes, the production chain of school uniforms typically includes multiple process nodes such as fabric inspection, cutting, sewing, hemming, ironing, quality sampling, and packaging. After completing their respective process data collection and hash encapsulation, each process node outputs its node record sequentially and passes the hash value of the current process node down. Through the sequential processing of all process nodes, a chain-like data structure composed of multiple node records is formed. Adjacent nodes form a dependency relationship through the correspondence between the parent node identifier and the hash value of the current process node, thus constituting a temporal evidence chain.

[0098] The core value of a temporal chain of evidence lies in its irreversibility of time sequence and its verifiability across the entire chain. For the first... Each process node has a hash value. The calculation depends on the first The hash value of each node ,and And depend on This continues until the beginning of the link. Therefore, to forge the first... The process data of the first node must be recalculated simultaneously. The system generates a chain of chronological evidence by hashing each node and all subsequent nodes, and forging a valid hardware root of trust signature for each node, which is computationally infeasible without access to the chip's built-in private key. This characteristic enables the chronological chain of evidence to provide a legally valid electronic evidentiary basis for the school uniform production process.

[0099] In production scenarios involving multiple parallel processes or brief interruptions between processes, the timestamp field of a node record and the process sequence number together constrain the legitimate insertion position of the node record, preventing attackers from forging the production process by adding historical nodes. The trusted execution environment module internally maintains a monotonically increasing counter, which increments automatically with each generated node record. The counter value is then incorporated into... The components of the system ensure that even if the business data collected twice is exactly the same, the node hash value will necessarily be different, fundamentally eliminating the risk of replaying node records.

[0100] The verification process of the chronological evidence chain corresponds to its generation process: starting from the genesis node at the beginning of the chain, each node record is retrieved sequentially, and the information stored in the node records is used... and Recalculate the hash value and compare the result with the information in the node record. Whether they match; and simultaneously verify the digital signature using the public key corresponding to the hardware root of trust. The validity of the verification is confirmed. Failure of any node verification indicates data anomalies in that node or its preceding nodes, allowing for precise identification of specific processes and providing quantifiable technical evidence for determining responsibility for quality issues.

[0101] Before the fabric is sent for inspection, the batch identifier hash value is encoded as a non-replicable physical identifier and simultaneously applied to both the production batch and the submitted sample. This establishes a physical hierarchical binding between the quality inspection sample and the production batch, resulting in a batch binding certificate, including:

[0102] Extract the batch hash value corresponding to the fabric batch identifier from the temporal evidence chain;

[0103] A unique sub-identifier is generated for each individual piece of fabric in the corresponding production batch and the sample to be tested. The sub-identifier is obtained by hashing the batch hash value and the randomly generated individual salt value. The sub-identifier is then encoded to generate a graphic code.

[0104] The graphic code is carried by an identification material that irreversibly bonds with the fabric substrate. During the application process, the identification material chemically bonds with the fabric fibers or causes microstructural modification of the fibers, leaving irreversible physical damage characteristics on the fabric surface after the identification is removed.

[0105] When submitting fabrics for inspection, the identification material bearing the graphic code is simultaneously applied to the same area of ​​both the sample to be inspected and the finished fabric of the corresponding production batch.

[0106] The identification device reads the markings applied to the submitted sample and extracts the sub-markers from the graphic code;

[0107] The batch hash value, sub-identifier, individual salt value and the submitted sample identifier are associated and encapsulated, and the mapping relationship table between the batch hash value and all sub-identifiers in the batch is recorded on the consortium blockchain to generate a batch binding certificate.

[0108] Extracting the batch hash value corresponding to the fabric batch identifier from the temporal evidence chain is the starting point for establishing a physical binding relationship. The temporal evidence chain sequentially records the hash encapsulation results of process data at each production process node. Before a fabric batch enters the quality inspection process, it is necessary to locate the corresponding process node record on the chain, read the stored batch identifier field, and parse the batch hash value from that field. Batch hash value It is the unique identifier digest of this production batch on the chain. All subsequent sub-identifiers are generated with it as the root value, ensuring that all individual identifiers within the same batch logically belong to the same batch root node, forming a hierarchical identifier system.

[0109] For each individual piece of fabric from the corresponding production batch and the sample to be tested, a unique sub-identifier needs to be generated. Generate a random salt value independently for each individual item. salinity Generated by a cryptographically secure random number generator, with a length of at least 128 bits, ensuring that the salt value is not repeated among different individuals within the same batch. Sub-identifier By batch hash value With individual salinity The result is obtained by performing a hash operation after concatenation. ,in This represents a byte-level concatenation operation, and the hash function uses a collision-resistant algorithm. Due to the salt value... The randomness, even in batch hash values Same, different individual sub-identifiers They are also different from each other, thus achieving unique differentiation at the individual level within a batch. Sub-identifiers Graphical codes are generated through encoding conversion. The encoding method adopts a two-dimensional matrix code format, which maps the binary or hexadecimal representation of the sub-identifier to a module matrix of the graphic code. The graphic code also embeds verification and error correction information to meet the error tolerance requirements of reading when the identifier is locally damaged.

[0110] The selection and application method of the labeling material are crucial for achieving a unique physical bond. The labeling material must meet the requirement of irreversible bonding with the fabric substrate. Specifically, this involves using polymer inks or nano-functional coatings containing active functional groups. These active functional groups can chemically bond with hydroxyl and amino groups on the surface of the fabric fibers under certain temperature and pressure conditions, forming covalent bonds. For natural fiber fabrics such as cotton and linen, the active groups in the labeling material undergo etherification or esterification reactions with the hydroxyl groups on the cellulose molecular chains. For synthetic fiber fabrics such as polyester and nylon, the labeling material penetrates into the micropores of the fiber through a hot-pressing process and forms a mechanical intercalation upon cooling. This process is accompanied by localized modification of the fiber's microstructure, causing irreversible changes in the arrangement of the fiber's crystalline regions. Both of the above-mentioned binding mechanisms result in a strong physical anchor between the labeling material and the fabric. Any forced removal of the label will damage the fiber structure and leave irreversible physical damage features on the fabric surface, including fiber breakage marks, surface roughening areas, or color change areas. These damage features can be identified under ordinary light sources or ultraviolet light sources, thus making it impossible to remove the label without leaving a trace, effectively preventing the label from being transferred to other fabrics or samples after being peeled off.

[0111] When fabrics are submitted for inspection, the marking material carrying the graphic code is simultaneously applied to the same areas of both the sample to be inspected and the finished fabric of the corresponding production batch. The application process is completed at the same time. A dedicated marking application device performs printing or pressing processes on the sample and the production batch fabric according to preset location coordinates, ensuring that the graphic code content on both is completely identical, i.e., the sub-mark. With salinity The correspondence between samples and batches of fabric is maintained one-to-one. The application site is a fixed location on the fabric, such as the lining seam or the back of the collar. This location is unique in the finished school uniform and is not easily affected by daily wear and tear, ensuring the readability of the markings throughout the uniform's lifespan. The application operation is recorded in the equipment log with the application timestamp, application site coordinates, operator identification, and batch number. This log data, along with the batch binding certificate, is stored on the blockchain as supplementary evidence, providing operational-level supporting proof for subsequent traceability.

[0112] The identification device reads the identifier applied to the submitted sample and extracts the sub-identifier from the graphic code. The recognition device employs a high-resolution image sensor combined with a graphic code decoding algorithm to scan and parse the graphic code, outputting the original encoded value of the sub-identifier. During the extraction process, the device verifies the integrity of the graphic code; if local damage is found, it is restored using embedded error correction information to ensure the accuracy of sub-identifier extraction. After extraction, the sub-identifier is... The physical number of the submitted sample is associated with the sample to form a sample-level identification extraction record, which includes the extraction time, identification device number, and extraction result.

[0113] Batch hash value Sub-identifier Individual salinity The data is associated with the sample identifier and encapsulated to construct a structured binding data packet. . Includes the following field: Batch hash value Used to identify the root node and sub-identifier of the batch. Used to locate specific individuals within a batch; individual salt value. Used to verify the legality of the generated sub-identifier (the verifier can recalculate it). and (Confirmation is achieved through comparison). The sample identifier is used to associate the test report number with the quality inspection agency. The batch hash value is recorded on the consortium blockchain. A mapping table relating all sub-identifiers within the batch is provided, organized in key-value pairs, where the key is the sub-identifier. The value is the corresponding individual salt value. Including application site information, the entire mapping table is hashed and digested before being included in the bound data packet. The data is submitted to the consortium blockchain for consensus verification, generating batch binding credentials.

[0114] The batch binding certificate is stored on the blockchain in the form of a single document containing... The transaction records in the summary and mapping relationship table are written into the block by the consortium blockchain nodes after consensus is reached. The transaction hash serves as the unique on-chain reference identifier for the batch binding certificate. Subsequently, during the tracing phase, the sub-identifier is extracted by reading the physical identifier on the school uniform. Then query the on-chain mapping table. Corresponding batch hash value With salinity Re-execute Calculation and extraction By comparing the data, it can be verified whether the school uniform belongs to the corresponding production batch. At the same time, by combining the inspection of physical damage characteristics, it can be determined whether the markings have been removed or replaced, thus completing the batch consistency verification at the physical level and ensuring that the quality inspection report truly corresponds to the fabric of the production batch.

[0115] The graphic code is carried by an identification material that irreversibly bonds with the fabric substrate. During application, the identification material chemically bonds with the fabric fibers or causes modification of the fiber microstructure, leaving irreversible physical damage features on the fabric surface after the identification is removed, including:

[0116] The labeling material is selected to contain reactive groups that can covalently bond with the functional groups on the surface of the fabric fibers;

[0117] During the application of the marking material, the reactive groups are activated by energy excitation, causing the marking material to react chemically with the functional groups on the surface of the fabric fibers to form covalent bonds.

[0118] During the chemical bonding process, the physical response characteristics of the interface between the marker material and the fabric substrate are simultaneously collected. These physical response characteristics include optical reflection characteristics or electromagnetic response characteristics of the interface.

[0119] Feature extraction is performed on the physical response characteristics to generate a combined interface fingerprint. The combined interface fingerprint is determined by the randomness of the distribution of the identifying material and the randomness of the fabric fiber microstructure, and has individual uniqueness.

[0120] The interface fingerprint is associated with the corresponding sub-identifier and recorded in the batch binding certificate;

[0121] During traceability verification, the graphic code of the physical identifier is read to obtain the sub-identifier, and the physical response characteristics of the area surrounding the identifier are simultaneously collected and compared with the interface fingerprint in the batch binding certificate;

[0122] When the identifier is removed or replaced, the physical response characteristics of the bonding interface are irreversibly changed due to the breakage of covalent bonds. This causes the currently collected physical response characteristics to not match the bonding interface fingerprint recorded in the batch binding certificate, thus determining that the identifier has been tampered with.

[0123] School uniform fabrics undergo multiple washes, frictions, and chemical treatments during production. Traditional printed or adhesive labels are prone to detachment or being manually peeled off and replaced, disrupting the quality control and traceability process. To address this issue, a graphic code is carried by a labeling material that irreversibly bonds to the fabric substrate. Through a chemical bonding mechanism, the label is permanently physically bound to the fabric fibers. The bonding process simultaneously collects unique physical characteristics of the bonding interface, creating an unforgeable physical anti-tampering mechanism.

[0124] The selection of labeling materials is a prerequisite for achieving irreversible bonding. For different school uniform fabric substrates such as cotton, polyester, and nylon, labeling materials containing corresponding reactive groups must be matched. Taking cotton fabric as an example, the surface of cotton fibers is rich in hydroxyl groups (-OH). Functional inks or coating materials containing epoxy groups (-CH2OCH-) or isocyanate groups (-NCO) can be selected. These reactive groups can undergo ring-opening addition or condensation reactions with hydroxyl groups under specific conditions to form stable covalent bonds. For polyester fabric, its surface ester bond structure is relatively inert. Labeling materials containing amino (-NH2) or carboxyl (-COOH) groups can be selected, and reacted with the pre-treated active sites on the polyester surface under high temperature or ultraviolet light conditions with a catalyst. The selection of labeling materials must be verified for substrate compatibility to ensure that there is a clear chemical reaction pathway between the reactive groups and the functional groups on the surface of the target fabric fibers, avoiding bonding failure or insufficient bonding strength due to material incompatibility.

[0125] During the application of marking materials, reactive groups are activated through energy excitation, driving chemical bonding reactions. The energy excitation method can be thermal excitation, ultraviolet (UV) excitation, or laser excitation, depending on the marking material system. Thermal excitation is suitable for thermosetting functional inks. The fabric area printed with the graphic code is placed in a controlled temperature field, typically between 120°C and 180°C, for a duration determined by the material formulation. This activates the reactive groups and allows them to form covalent bonds with functional groups on the fiber surface. UV excitation is suitable for photoinitiated marking materials. Under UV light of a specific wavelength, the photoinitiator decomposes to generate free radicals, initiating free radical polymerization or grafting reactions between the acrylate or vinyl functional groups in the marking material and the fiber surface. Laser excitation offers the advantage of high spatial precision, allowing selective activation of the graphic code area and avoiding thermal damage to non-marking areas of the fabric. Regardless of the excitation method used, precise control of the excitation energy parameters is essential to ensure sufficient activation of reactive groups and complete covalent bond formation, while avoiding excessive energy input that could lead to fiber degradation and affect the quality of the school uniform.

[0126] During the chemical bonding process, the physical response characteristics of the interface between the labeling material and the fabric substrate are simultaneously acquired. The timing of acquisition is crucial; the physical state of the interface best reflects the true interaction between the material distribution and the fiber structure when the bonding reaction has reached a stable state but is not yet fully cured. Optical reflectance characteristics are acquired using multispectral imaging equipment, scanning the labeling area pixel-by-pixel in the visible and near-infrared bands to record reflectance distribution images at different wavelengths. Due to the randomness of the penetration depth, lateral diffusion range, and bonding degree of the labeling material in the fiber gaps with different fiber bundles, the reflectance distribution images exhibit unpredictable spatial variations at the microscale. Electromagnetic response characteristics are acquired using near-field detection equipment in the microwave or terahertz bands. Utilizing the difference in dielectric constant between the labeling material and the fabric substrate, the dielectric distribution characteristics of the bonding interface are detected. This characteristic is highly sensitive to the depth and uniformity of the material's embedding in the fiber structure.

[0127] Feature extraction is performed on the acquired physical response characteristics to generate a bonding interface fingerprint. The feature extraction process first preprocesses the original acquired images or signals, including denoising, normalization, and coordinate alignment, to eliminate the influence of system errors in the acquisition equipment. Then, multi-scale texture features are extracted, and local binary patterns, gray-level co-occurrence matrices, or wavelet transform coefficients are calculated at different spatial resolutions to capture the feature information of the bonding interface at both the macroscopic distribution and microscopic detail levels. The extracted multidimensional feature vectors are then dimensionality-reduced and encoded into a fixed-length bonding interface fingerprint string. The uniqueness of the bonding interface fingerprint is guaranteed by two independent sources of randomness: the randomness of the distribution of the marker material during application due to rheological properties, surface tension, and environmental disturbances; and the inherent microstructural randomness of the fabric fibers themselves in terms of textile structure, fiber diameter distribution, and surface morphology. The superposition of these two randomnesses ensures that even when markers are applied to different locations of the same batch of fabric under the same process parameters, each bonding interface fingerprint is statistically unique, and the same result cannot be obtained through repeated preparation or artificial replication.

[0128] The interface fingerprint is associated with the corresponding sub-identifier and recorded in the batch binding certificate. The sub-identifier is provided by the individual unique identifier in the batch binding data packet generated in the previous process. Combined with the interface fingerprint, it serves as the biometric anchor of the physical entity corresponding to the sub-identifier, forming a complete physical identity record together with the sub-identifier. The record structure stored in the batch binding certificate includes the sub-identifier code, combined with the interface fingerprint data, the collection timestamp, the collection device identifier, and the collection parameter configuration, ensuring that feature comparison can be performed under the same conditions during subsequent verification. The entire recording process is completed in a trusted execution environment to prevent the fingerprint data from being tampered with before storage.

[0129] During the traceability verification phase, after obtaining the sub-identifier by reading the graphic code of the physical identifier, physical response features of the area surrounding the identifier are simultaneously collected, with the collection parameters remaining consistent with those used in the initial recording. The currently collected features are then processed through the same feature extraction workflow and compared with the interface fingerprint stored in the batch binding voucher for similarity. The comparison employs metrics such as normalized Hamming distance or cosine similarity, setting a judgment threshold to distinguish between minor feature changes caused by normal wear and tear and significant feature shifts caused by human tampering.

[0130] When the label is physically removed, the covalent bond breaking process inevitably damages the microstructure of the fabric fibers, leaving irreversible physical damage characteristics such as fiber tears, surface etching, or chemical residues in the label area. Even if a similar-looking label is subsequently applied to the same location, the fingerprint of the interface formed between the new label and the damaged fiber surface will differ significantly from the original record, and the similarity comparison result will be below the judgment threshold, thus triggering a tampering detection. When the label is completely replaced, that is, when label material containing counterfeit sub-labels is applied to different fabric locations, the fingerprint of the interface at the new location will also fail to match the original fingerprint in the batch binding certificate due to the difference in the microstructure of the fabric fibers, and the tampering behavior can still be detected. This mechanism physically blocks the possibility of label forgery and replacement, and together with the digital layer anti-tampering mechanism of blockchain evidence storage, it forms a dual protection system to ensure the integrity and credibility of the school uniform quality inspection traceability link.

[0131] Based on a standard knowledge graph, the system automatically parses and compares the detection indicators of the quality inspection report data against the limits, generating verification conclusions, including:

[0132] A standard knowledge graph is constructed, which organizes testing indicators, national standard limits, certification body qualifications, hazardous substance lists, batch sensitivity attributes, and process causal rules in the form of nodes and related edges;

[0133] Extract the names and values ​​of testing indicators from the quality inspection report data;

[0134] The batch sensitivity attributes corresponding to the detection indicators are queried from the standard knowledge graph. Batch correlation indicators whose detection values ​​are affected by production process parameters are filtered out from the batch sensitivity attributes. The process causal rules corresponding to the batch correlation indicators are obtained. The process causal rules describe the causal constraint relationship between the detection values ​​and the production process parameters.

[0135] Extract the production process parameters corresponding to the batch in the quality inspection report from the chronological evidence chain;

[0136] According to the process causality rule, the detection values ​​of batch-related indicators are checked for causal consistency with the production process parameters of the corresponding batch. When the detection value exceeds the preset value range corresponding to the process parameters of the batch, the quality inspection report is marked as having a logical conflict with the batch.

[0137] For non-batch-related indicators, compare them against national standard limits and verify the qualifications of certification bodies;

[0138] When there is a logical conflict between batch-related indicators, it is judged as batch mismatch; when there is a standard exceeding the limit for non-batch-related indicators, it is judged as quality non-compliance; when the testing agency exceeds the authorized scope or prohibited substances are detected, it is judged as compliance abnormality. A verification conclusion is generated that includes batch matching conclusion, quality compliance conclusion, and compliance conclusion.

[0139] Combination Figure 2 The intelligent verification process for quality inspection reports based on standard knowledge graphs is illustrated with an example diagram:

[0140] The construction process of the standard knowledge graph includes the following specific steps: First, extract the mandatory standard text from the national standard database, perform structured parsing on the standard text, and extract fields such as the name of the testing indicator, limit value, scope of application, and standard number of the testing method. Each testing indicator is modeled as a node, with node attributes including indicator name, unit, and a list of synonyms. Second, the national standard limit values ​​corresponding to each testing indicator are modeled as independent nodes, with node attributes including upper limit value, lower limit value, and applicable product category. An "applicable limit value" association edge is established between the testing indicator nodes and the limit value nodes. Third, extract information on testing institutions with CMA or CNAS qualifications from the certification body directory published by the market supervision department, including the institution name, certificate number, list of authorized testing items, and validity period. Each institution is modeled as a certification body qualification. The graph is constructed by first establishing quality nodes and creating "authorized testing" association edges between institution nodes and corresponding authorized testing item nodes. Then, based on the statistical analysis of textile production process principles and historical quality inspection data, indicators significantly affected by production process parameters are labeled as batch-sensitive indicators. Batch sensitivity attribute tags are added to these indicator nodes, and a process causal rule node is constructed for each batch-sensitive indicator. This node stores a mapping table of process parameter types, process parameter value ranges, and preset test value ranges. This mapping table is obtained through regression analysis or expert knowledge rule extraction based on historical production batch process parameter records and corresponding quality inspection report data. Finally, all nodes and association edges are imported into a graph database, an index is created to support fast querying, and version numbers and update date attributes are set for the graph. After the graph is constructed, nodes and association edges are maintained through a version management mechanism. When national standards are updated, only the affected nodes and edges need to be partially updated, and historical versions of the graph are retained to support traceability and verification of historical batch quality inspection reports.

[0141] When extracting test indicator names and test values ​​from quality inspection report data, a combination of rule-based structured parsing and semantic matching is used. Before entering the parsing process, the quality inspection report data has undergone OCR or electronic format conversion to form structured text. The parser identifies the test item column, test result column, and test basis column based on the report layout template, splitting each record into a quadruple (indicator name, test value, unit, test basis standard). Since reports issued by different testing institutions may differ in the description of indicator names (e.g., "formaldehyde content" and "free formaldehyde" both refer to the same test indicator), the parser matches the indicator name with the synonym list of indicator nodes in the standard knowledge graph to complete the standardized mapping of indicator names, ensuring a one-to-one correspondence between the extracted results and graph nodes. After extraction, all quadruples are temporarily stored in the form of structured records for subsequent batch sensitivity screening and limit comparison steps.

[0142] When querying batch sensitivity attributes corresponding to testing indicators from a standard knowledge graph, the normalized indicator name is used as the search key, and the corresponding batch sensitivity attribute nodes are retrieved along the "batch-sensitive" association edge. A batch sensitivity attribute refers to the characteristic that the measured value of a certain testing indicator will change regularly with variations in the process parameters of a production batch. Specifically, for testing indicators of school uniform fabrics, the measured values ​​of some indicators, such as pH value, formaldehyde content, and color fastness, are directly affected by dyeing process parameters (such as dyeing temperature, dyeing time, and auxiliary agent concentration) or finishing process parameters (such as setting temperature and setting time). When process parameters differ between different batches, the measured values ​​of these indicators will also fluctuate within a certain range; these indicators are labeled as batch-related indicators. However, some indicators, such as fiber composition and fabric structure, are mainly determined by the characteristics of raw materials and remain relatively stable between different production batches of the same fabric series, unaffected by significant fluctuations in conventional production process parameters; these indicators are not batch-related indicators. The batch sensitivity attribute node stores a Boolean flag indicating whether the indicator belongs to the batch-related indicators, as well as a list of production process parameter types that affect the indicator value, such as dyeing temperature, dyeing time, setting temperature, and auxiliary agent concentration. After filtering out the batch-related indicators, the corresponding process causality rule node is obtained along the "constrained by process causality rule" association edge. The process causality rule node stores a mapping table of process parameter names, process parameter value ranges, and preset value ranges of detection values. This mapping table is determined based on historical production data and standard limits, and is expressed as: when process parameter... Falling into the range At that time, the detection value of the corresponding detection indicator should fall within the preset value range. ,in , These are the lower and upper limits of the range of process parameters. , The lower and upper limits of the range for the measured values ​​are preset. If the measured value... satisfy If so, the indicator passes the causal consistency check; if or If so, the indicator is marked as having a logical conflict.

[0143] When extracting the production process parameters corresponding to the batch in the quality inspection report from the time-series evidence chain, the batch hash value recorded in the batch binding certificate is used as an index to locate the corresponding process node records on the time-series evidence chain. Each process node record contains a process type label and a process data field. The process data field stores process parameter key-value pairs according to a predefined data pattern. All process node records for the batch are traversed and aggregated by process parameter name to form a complete process parameter file for the batch. For the process parameter types involved in the batch-related indicators, the corresponding actual record values ​​are extracted from the process parameter file as input for causal consistency verification. The process node records in the time-series evidence chain have been protected by hash encapsulation and digital signatures within a trusted execution environment. The extracted process parameter values ​​possess tamper-proof reliability and can be used as a verification benchmark.

[0144] After causal consistency verification is completed, national standard limit comparison and certification body qualification verification are performed on non-batch-related indicators. Limit comparison uses the standardized indicator name as the key, queries the corresponding compliance threshold from the national standard limit node in the standard knowledge graph, compares the extracted test values ​​with the thresholds, and marks indicators exceeding the thresholds as exceeding the standard limit. Certification body qualification verification uses the testing organization name and testing items recorded in the quality inspection report as input, queries the authorized testing scope of the organization from the certification body qualification node in the graph, and determines whether all testing items covered in this report are within the authorized scope. Simultaneously, all substances detected in the report are compared with the hazardous substance list node. If any prohibited substance is detected, a compliance anomaly flag is triggered, regardless of whether the detected amount exceeds a certain threshold; detection is considered an anomaly.

[0145] The verification conclusion is generated based on the combined logic of three types of judgment results. When any batch-related indicator is marked as logically conflicting, the judgment conclusion is batch mismatch, indicating a logical contradiction at the process parameter level between the quality inspection report and the production batch, suggesting the quality inspection sample may not originate from that production batch. When any non-batch-related indicator exceeds the standard limit, the judgment conclusion is quality non-compliance, indicating the batch of fabric did not meet national standard requirements for the corresponding testing item. When the testing institution conducts testing beyond its authorized scope or detects prohibited substances in the report, the judgment conclusion is compliance anomaly, indicating a problem with the compliance of the quality inspection report itself. These three types of conclusions are independent and can coexist in the same verification conclusion. The final output verification conclusion is organized in structured data format, including batch matching conclusion fields, quality compliance conclusion fields, and compliance conclusion fields. Each field contains a judgment result label and a list of specific testing indicators supporting the judgment result, along with their corresponding values. This allows the verification conclusion, along with the chronological evidence chain and batch binding certificate, to be submitted to the consortium blockchain network for consensus verification and notarization.

[0146] The structured output design of the verification conclusions allows for precise identification of the specific testing indicators and process parameters leading to a particular judgment during traceability queries, rather than simply returning a Boolean value of pass or fail. When the traceability requester needs to understand the root cause of a quality problem in a batch of school uniforms, they can directly trace back to the specific process parameter values ​​recorded at the corresponding process node in the time-series evidence chain through the indicator list and process causal rules in the verification conclusions. This forms a complete causal traceability path from the final school uniform product to the original production process parameters, supporting the identification of responsibility for quality problems and process improvement.

[0147] The chronological evidence chain, batch binding credentials, and verification conclusions are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier, including:

[0148] The chronological evidence chain, batch binding credentials, and verification conclusions are structured and encapsulated to generate an evidence storage data package;

[0149] Data digests are calculated for the chronological evidence chain, batch binding certificate, and verification conclusion in the evidence storage data packet. The three data digests are then concatenated in chronological order, and the overall hash value is calculated as the evidence storage fingerprint.

[0150] The evidence storage data packet and evidence storage fingerprint are encapsulated into an evidence storage transaction request and submitted to the consensus node of the consortium blockchain network;

[0151] Consensus nodes verify the notarization transaction request. Verification includes checking the consistency between the notarization fingerprint and the notarization data packet, as well as verifying the identity and permissions of the submitting node. Once the verification is successful, the notarization data packet is written into the candidate block, and consensus voting is conducted on the candidate block. When the candidate block receives confirmation from more than a preset proportion of consensus nodes in the consortium blockchain network, the candidate block is added to the blockchain ledger.

[0152] Extract the transaction record identifier containing the evidence storage data packet from the newly added block as the evidence storage transaction identifier, and associate it with the corresponding fabric batch identifier for subsequent querying of the corresponding blockchain evidence storage transaction identifier through the fabric batch identifier.

[0153] After completing the construction of the temporal evidence chain, the generation of batch binding certificates, and the formation of quality inspection verification conclusions, the three types of core data need to be uniformly incorporated into the consortium blockchain for immutable and permanent notarization. The starting point for notarization on the blockchain is the structured encapsulation of the three types of data to form a standardized notarization data package. Specifically, the temporal evidence chain exists in the form of an ordered sequence of node records; the batch binding certificate contains the physical layer binding information between the fabric batch identifier and the submitted sample; and the verification conclusion is a structured judgment result output after limit comparison based on a standard knowledge graph. These three types of data are organized into a unified notarization data package according to a predefined data pattern, ensuring that the type, length, and encoding format of each field meet the parsing specifications of the consortium blockchain nodes. During the encapsulation process, the source identifier, generation timestamp, and data version number of each type of data are included in the metadata field so that the generation background of each type of data can be accurately restored during subsequent parsing.

[0154] After the evidence storage data packet is encapsulated, its integrity fingerprint needs to be calculated to support the consistency verification of subsequent consensus nodes. Data digests are calculated independently for the chronological evidence chain, batch binding credential, and verification conclusion in the evidence storage data packet. Cryptographic hash functions are used to perform digest operations on the serialized byte streams of each type of data, resulting in three independent digest values. These three digest values ​​are then concatenated byte-by-byte in chronological order—that is, the chronological evidence chain digest, the batch binding credential digest, and the verification conclusion digest—and the hash value is calculated on the concatenated byte string as a whole to obtain the evidence storage fingerprint. The significance of this hierarchical summary design lies in the fact that even a slight change in any single category of data will cause a change in the corresponding summary, thereby affecting the overall evidence fingerprint. The change allows data tampering to be detected during the consensus verification phase without requiring a full comparison of the evidence data packets, significantly reducing the computational overhead of verification.

[0155] fingerprint evidence After the calculation is completed, the evidence storage data packet and evidence storage fingerprint will be stored. The request is encapsulated into a single notarization transaction request. In addition to the core payload mentioned above, the notarization transaction request also includes the identity credentials of the submitting node. These credentials are typically represented by a digital certificate or an identity token derived from a pre-shared key, used by the consensus nodes to verify the submitter's legitimacy. After being serialized and encoded according to the transaction format specified by the consortium blockchain network, the notarization transaction request is broadcast to the consensus nodes in the consortium blockchain network via a peer-to-peer network protocol. Before broadcasting, the submitting node records the timestamp and request sequence number of this submission to trigger a retry mechanism if the transaction fails to be uploaded to the blockchain within a preset timeout period, ensuring the reliable delivery of the notarization request.

[0156] After receiving a notarization transaction request, the consensus node performs two independent verification operations. The first verification is a consistency check of the notarized fingerprint: the consensus node recalculates the digests of the chronological evidence chain, batch binding certificate, and verification conclusion in the received notarization data packet according to the same hierarchical digest rule as the submitting end. After concatenating them in chronological order, it recalculates the overall hash value and compares the recalculated hash value with the notarized fingerprint carried in the notarization transaction request. The first step involves comparing the two data points. If they don't match, it's determined that the data was corrupted or tampered with during transmission, and the transaction request is rejected. The second verification step checks the identity and permissions of the submitting node: the consensus node queries the access list maintained by the consortium blockchain network to verify whether the submitting node's identity credentials belong to an authorized and legitimate production link node. Transaction requests submitted by unauthorized nodes will also be rejected to prevent malicious external data from being injected into the consortium ledger.

[0157] After both verifications pass, the consensus node includes the evidence storage data packet from the evidence storage transaction request into the candidate block. Multiple evidence storage transactions arriving within a certain time window can be aggregated in the candidate block to improve block utilization. Once the candidate block is constructed, the initiating node broadcasts the candidate block to other consensus nodes in the consortium blockchain network. Each consensus node independently performs the two verifications on each transaction in the candidate block and casts a vote of approval or disapproval based on its own verification results. When the number of approval votes exceeds a pre-set threshold in the consortium blockchain network, the candidate block is considered to have passed consensus, and the master node appends it to the end of the blockchain ledger, becoming an irreversible official block. This consensus voting mechanism ensures that even if a few faulty or malicious nodes exist in the consortium, the consistency and correctness of the evidence storage data can still be guaranteed, meeting the basic requirements of the fault-tolerant design of consortium blockchains.

[0158] After a candidate block is successfully written to the blockchain ledger, the transaction record identifier containing the evidence storage data package is extracted from the newly added block and used as the evidence storage transaction identifier for this evidence storage operation. Evidence-based transaction identifier Typically, the identifier is composed of the block height and the transaction sequence number within the block, or it may directly use the transaction hash value as a globally unique identifier, enabling precise location of the corresponding evidence storage data packet within the consortium blockchain network. The submitting node obtains the evidence storage transaction identifier. Then, the record is associated with the corresponding fabric batch identifier and written to the local batch index database. The association record uses the fabric batch identifier as the primary key and stores the corresponding evidence-based transaction identifier. It also includes supplementary information such as the on-chain time and block height, so that when a traceability request arrives, the corresponding blockchain evidence location can be quickly located through the fabric batch identifier without the need for a full scan on the blockchain.

[0159] This linked recording mechanism has significant practical value in traceability scenarios. When regulatory agencies, school purchasers, or parents requesting traceability scan the physical tags on school uniforms to obtain fabric batch identifiers, they can retrieve the corresponding documented transaction identifiers from the batch index database using these fabric batch identifiers. Then, based on the evidence of the transaction, The system precisely extracts evidence data packages from the consortium blockchain ledger, analyzes the chronological evidence chain, batch binding certificates, and verification conclusions, and completely reconstructs the entire quality data of the batch from raw materials to finished products. The entire retrieval process does not require trust in any single intermediary; all data originates from the consortium blockchain ledger, and its immutability is jointly endorsed by multiple consensus nodes, fundamentally solving the problems of traditional paper-based quality inspection reports being easy to forge and difficult to verify.

[0160] The batch index database itself can also be periodically snapshotted and uploaded to the blockchain in the form of a Merkle tree to prevent the fabric batch identifier and the evidence-based transaction identifier from being compromised if the local index data is tampered with. The mapping relationship between them is distorted. The snapshot on-chain cycle can be flexibly configured according to the frequency of production batches. For example, an index snapshot can be generated after summarizing the number of production batches per day. The root hash value of the snapshot is submitted to the consortium blockchain as an auxiliary evidence transaction, which makes the index mapping relationship itself verifiable and further strengthens the credibility of the overall traceability system.

[0161] In response to traceability requests, the system retrieves and parses the notified transaction identifier from the consortium blockchain. It verifies batch consistency by reading the physical identifier of the school uniform. When the physical identifier's feature code matches the batch binding certificate on the blockchain, it confirms that the quality inspection report truly corresponds to the production batch. The system then returns the verification results and end-to-end quality data, including:

[0162] Receive a traceability request, which includes the physical identifier or fabric batch identifier of the school uniform to be verified;

[0163] The sub-identifier is obtained by parsing the graphic code on the school uniform using an identification device, or by directly obtaining the fabric batch identifier. The corresponding notarized transaction identifier is then retrieved from the consortium blockchain based on the sub-identifier or the fabric batch identifier.

[0164] Based on the evidence storage transaction identifier, evidence storage data packets are extracted from the consortium blockchain ledger, and the time-series evidence chain, batch binding certificate, and verification conclusion are obtained by parsing the evidence storage data packets;

[0165] Extract the sub-identifier mapping table for the batch from the batch binding certificate, and determine whether the read sub-identifier exists in the mapping table. If the sub-identifier exists in the mapping table, it is determined that the physical identifier matches the batch binding certificate, confirming that the quality inspection report of the school uniform truly corresponds to the production batch. If the sub-identifier does not exist, it is determined that there is a risk of batch replacement.

[0166] The batch matching determination results are correlated with the verification conclusions to generate verification results that include both batch consistency determination and quality compliance determination.

[0167] Process data for each production step in this batch is extracted from the chronological evidence chain and combined with the quality inspection report data in the verification conclusion to generate end-to-end quality data;

[0168] The verification results and end-to-end quality data are returned to the traceability request initiator.

[0169] Traceability requests are typically triggered by random inspections by education authorities, quality inquiries from parents regarding purchased school uniforms, or compliance confirmations by school procurement personnel during the acceptance process. The information carried in traceability requests falls into two categories: one is the physical identifier obtained by scanning the graphic code on the school uniform, and the other is the directly possessed fabric batch identification, such as the batch number recorded in the procurement contract. These two input paths differ in their subsequent processing logic and need to be processed separately before being integrated into a unified retrieval process.

[0170] When a traceability request carries a physical identifier, the identification device (such as a mobile phone camera with a decoding application or a dedicated handheld scanning terminal) optically reads the graphic code on the school uniform label or hangtag, decoding it to obtain an encoded string. This encoded string is then parsed to extract a sub-identifier. The sub-identifier is a composite identifier generated by superimposing a unique individual sub-identifier and a random salt value on the hash value of the fabric batch identifier, ensuring uniqueness within the batch. When the traceability request directly carries the fabric batch identifier, the graphic code reading step is skipped, and the batch identifier is used directly as the search key. Regardless of the input format, the final search request for the notarized transaction identifier is initiated to the query interface of the consortium blockchain network using either the sub-identifier or the fabric batch identifier as an index. The consortium blockchain node locates the corresponding notarized transaction identifier in the on-chain index structure based on the index value and returns it to the requester.

[0171] After obtaining the notarized transaction identifier, the complete notarized data package is extracted from the consortium blockchain ledger. The notarized data package has already undergone overall hash fingerprint verification when uploaded to the chain. After extraction, the overall hash value of the data package needs to be recalculated and compared with the notarized fingerprint recorded on the chain to confirm that the data package has not been tampered with during transmission and storage. After passing the integrity verification, the notarized data package is structured and parsed, extracting three components: the chronological evidence chain, the batch binding certificate, and the verification conclusion. The chronological evidence chain records the process data and hash chain structure of each step from raw material warehousing to finished product delivery; the batch binding certificate contains the physical hierarchical binding information between the fabric batch and the submitted sample, as well as a mapping table of all individual sub-identifiers within that batch; the verification conclusion records the automatic parsing results and limit comparison conclusions of each test indicator in the quality inspection report.

[0172] The core operation of batch consistency verification is to match the sub-identifiers obtained from parsing the graphic code with the sub-identifier mapping table stored in the batch binding certificate. The mapping table is organized in a hash table structure and stores a set of all legal individual sub-identifiers within the production batch. A set member query is performed on the read sub-identifiers: if the query result is true, meaning the sub-identifier exists in the mapping table, the physical identifier of the school uniform is determined to be consistent with the on-chain batch binding certificate, confirming that the quality inspection report corresponding to the school uniform truly belongs to this production batch, and the batch consistency determination conclusion is "match passed"; if the query result is false, meaning the sub-identifier does not exist in the mapping table, a batch replacement risk is determined, meaning the school uniform may not come from the production batch corresponding to the quality inspection report, the batch consistency determination conclusion is "batch replacement risk exists," and an exception flag is triggered, recording the timestamp of the query and the requester information for subsequent manual review.

[0173] After batch consistency determination is completed, the determination conclusion and the verification conclusion are linked and integrated to generate the final verification result. The verification result includes two dimensions of determination: the batch consistency determination describes whether the school uniform physical product comes from the production batch corresponding to the quality inspection report; the quality compliance determination comes from the verification conclusion, describing whether all test indicators of the fabric in this batch meet the corresponding national or industry standard limit requirements. When the batch consistency determination is "match passed" and the quality compliance determination is "fully compliant", the verification result is a qualified conclusion with the highest confidence level; when there is an anomaly in either dimension, the verification result will include a specific description of the anomaly, making it easier for the request initiator to understand the problem.

[0174] The generation of end-to-end quality data relies on process data from each node in the time-series evidence chain. Records for each node are extracted from the time-series evidence chain in sequence according to the process number. Each node record includes the process name, execution timestamp, operator identifier, equipment number, actual recorded values ​​of key process parameters, and their compliance status. For key processes such as fabric pretreatment, cutting, sewing, ironing, and finished product inspection, corresponding process data snapshots can be obtained from the time-series evidence chain. The process data for each of these processes are arranged according to production timeline, and the quality inspection report data from the verification conclusion (including the name of the testing institution, testing date, names of each testing item, actual test values, and comparison results with standard limits) is appended to the quality inspection link in the end-to-end quality data, forming a quality data view covering the entire process from raw material warehousing to finished product delivery.

[0175] The end-to-end quality data view is structurally organized chronologically along the production chain. The party initiating the traceability request can clearly view the execution status and compliance status of each process, as well as the test results of the fabric samples. For supervision and verification scenarios by education authorities, the end-to-end quality data provides a complete chain of evidence to support compliance audits. For parent inquiries, key quality nodes are presented in a simplified summary format, lowering the barrier to information interpretation.

[0176] Finally, the verification results and end-to-end quality data are encapsulated in the response format specified in the traceability request protocol and returned to the traceability request initiator via a secure channel. The response data packet also carries the on-chain notarized transaction identifier for this traceability query, allowing the requesting party to verify the authenticity of the response data with the consortium blockchain node, preventing tampering of the response data by the traceability service's intermediate layer. Throughout the entire traceability verification process, from receiving the request to returning the result, all key operations are completed within the trusted environment of the consortium blockchain. The execution of the verification logic does not rely on the credit endorsement of any single centralized institution, thus ensuring that the traceability results have equal credibility for all participants.

[0177] A second aspect of this invention provides a blockchain-based quality traceability system for the entire school uniform production chain, comprising:

[0178] The data acquisition unit is used to acquire process data at various stages of the school uniform production chain.

[0179] The hash encapsulation unit is used to perform hash encapsulation on the process data based on the trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data to form a time-series evidence chain.

[0180] The batch binding unit is used to encode the fabric batch identifier hash value into an uncopyable physical identifier and apply it simultaneously to the production batch fabric and the submitted sample before the fabric is sent for inspection, thereby establishing a physical hierarchical binding between the quality inspection sample and the production batch and obtaining a batch binding certificate.

[0181] The indicator comparison unit is used to automatically parse and compare the detection indicators of the quality inspection report data with the limit based on the standard knowledge graph, and generate verification conclusions.

[0182] The consensus-based evidence storage unit is used to submit the time-series evidence chain, batch binding certificate and verification conclusion to the consortium blockchain network for consensus verification and generate a storage transaction identifier.

[0183] The traceability verification unit is used to respond to traceability requests, retrieve and parse the evidence transaction identifier from the consortium blockchain, verify batch consistency by reading the physical identifier of the school uniform, and confirm that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and return the verification result and full-chain quality data.

[0184] A third aspect of the present invention provides an electronic device, comprising:

[0185] processor;

[0186] Memory used to store processor-executable instructions;

[0187] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0188] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0189] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

Claims

1. A blockchain-based quality traceability method for the entire school uniform production chain, characterized in that: include: Acquire process data at each stage of the school uniform production chain; The process data is hash-encapsulated based on a trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data to form a time-series evidence chain. Before the fabric is sent for testing, the batch identifier hash value is encoded as a non-replicable physical identifier and applied simultaneously to both the production batch and the submitted sample, establishing a physical hierarchical binding between the quality inspection sample and the production batch, thus obtaining a batch binding certificate; Based on a standard knowledge graph, the system automatically analyzes and compares the detection indicators of the quality inspection report data against the limit values, and generates verification conclusions. The chronological evidence chain, batch binding certificate, and verification conclusion are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier. In response to traceability requests, the system retrieves and parses the notarized transaction identifier from the consortium blockchain, verifies batch consistency by reading the physical identifier of the school uniform, and confirms that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and returns the verification result and full-chain quality data.

2. The method according to claim 1, characterized in that, The process data is hashed and encapsulated based on a trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data, forming a time-series evidence chain, including: A trusted execution environment module is integrated into the data acquisition equipment at the production site. The trusted execution environment module includes an isolated execution region and a hardware root of trust. Within the isolated execution area, process data of the current process is received, and the hash value of the preceding process node is extracted as the parent node identifier; The parent node identifier is concatenated with the process data of the current process, and the hash value of the current process node is calculated using a hash function. The hardware root of trust then digitally signs the hash value of the current process node. The data structure containing the parent node identifier, the hash value of the current process node, and the digital signature is output as the current node record, and the hash value of the current process node in the current node record is used as the parent node identifier of the next process node; Multiple node records are generated sequentially according to the process sequence, forming a chain-like data structure with sequential dependencies, thus constituting a time-series evidence chain.

3. The method according to claim 1, characterized in that, Before the fabric is sent for inspection, the batch identifier hash value is encoded as a non-replicable physical identifier and simultaneously applied to both the production batch and the submitted sample. This establishes a physical hierarchical binding between the quality inspection sample and the production batch, resulting in a batch binding certificate, including: Extract the batch hash value corresponding to the fabric batch identifier from the temporal evidence chain; A unique sub-identifier is generated for each individual piece of fabric in the corresponding production batch and the sample to be tested. The sub-identifier is obtained by hashing the batch hash value and the randomly generated individual salt value. The sub-identifier is then encoded to generate a graphic code. The graphic code is carried by an identification material that irreversibly bonds with the fabric substrate. During the application process, the identification material chemically bonds with the fabric fibers or causes microstructural modification of the fibers, leaving irreversible physical damage characteristics on the fabric surface after the identification is removed. When submitting fabrics for inspection, the identification material bearing the graphic code is simultaneously applied to the same area of ​​both the sample to be inspected and the finished fabric of the corresponding production batch. The identification device reads the markings applied to the submitted sample and extracts the sub-markers from the graphic code; The batch hash value, sub-identifier, individual salt value and the submitted sample identifier are associated and encapsulated, and the mapping relationship table between the batch hash value and all sub-identifiers in the batch is recorded on the consortium blockchain to generate a batch binding certificate.

4. The method according to claim 3, characterized in that, The graphic code is carried by a marking material that is irreversibly bonded to the fabric substrate, including: The labeling material is selected to contain reactive groups that can covalently bond with the functional groups on the surface of the fabric fibers; During the application of the marking material, the reactive groups are activated by energy excitation, causing the marking material to react chemically with the functional groups on the surface of the fabric fibers to form covalent bonds. During the chemical bonding process, the physical response characteristics of the interface between the marker material and the fabric substrate are simultaneously collected. These physical response characteristics include optical reflection characteristics or electromagnetic response characteristics of the interface. Feature extraction is performed on the physical response characteristics to generate a combined interface fingerprint. The combined interface fingerprint is determined by the randomness of the distribution of the identifying material and the randomness of the fabric fiber microstructure, and has individual uniqueness. The interface fingerprint is associated with the corresponding sub-identifier and recorded in the batch binding certificate; During traceability verification, the graphic code of the physical identifier is read to obtain the sub-identifier, and the physical response characteristics of the area surrounding the identifier are simultaneously collected and compared with the interface fingerprint in the batch binding certificate; When the identifier is removed or replaced, the physical response characteristics of the bonding interface are irreversibly changed due to the breakage of covalent bonds. This causes the currently collected physical response characteristics to not match the bonding interface fingerprint recorded in the batch binding certificate, thus determining that the identifier has been tampered with.

5. The method according to claim 1, characterized in that, Based on a standard knowledge graph, the system automatically parses and compares the detection indicators of the quality inspection report data against the limits, generating verification conclusions, including: A standard knowledge graph is constructed, which organizes testing indicators, national standard limits, certification body qualifications, hazardous substance lists, batch sensitivity attributes, and process causal rules in the form of nodes and related edges; Extract the names and values ​​of testing indicators from the quality inspection report data; The batch sensitivity attributes corresponding to the detection indicators are queried from the standard knowledge graph. Batch correlation indicators whose detection values ​​are affected by production process parameters are filtered out from the batch sensitivity attributes. The process causal rules corresponding to the batch correlation indicators are obtained. The process causal rules describe the causal constraint relationship between the detection values ​​and the production process parameters. Extract the production process parameters corresponding to the batch in the quality inspection report from the chronological evidence chain; According to the process causality rule, the detection values ​​of batch-related indicators are checked for causal consistency with the production process parameters of the corresponding batch. When the detection value exceeds the preset value range corresponding to the process parameters of the batch, the quality inspection report is marked as having a logical conflict with the batch. For non-batch-related indicators, compare them against national standard limits and verify the qualifications of certification bodies; When there is a logical conflict between batch-related indicators, it is judged as batch mismatch; when there is a standard exceeding the limit for non-batch-related indicators, it is judged as quality non-compliance; when the testing agency exceeds the authorized scope or prohibited substances are detected, it is judged as compliance abnormality. A verification conclusion is generated that includes batch matching conclusion, quality compliance conclusion, and compliance conclusion.

6. The method according to claim 1, characterized in that, The chronological evidence chain, batch binding credentials, and verification conclusions are submitted to the consortium blockchain network for consensus verification and to generate a notarized transaction identifier, including: The chronological evidence chain, batch binding credentials, and verification conclusions are structured and encapsulated to generate an evidence storage data package; Data digests are calculated for the chronological evidence chain, batch binding credential, and verification conclusion in the evidence storage data packet. The three data digests are then concatenated in chronological order, and the overall hash value is calculated as the evidence storage fingerprint. The evidence storage data packet and evidence storage fingerprint are encapsulated into an evidence storage transaction request and submitted to the consensus node of the consortium blockchain network; Consensus nodes verify the notarization transaction request. Verification includes checking the consistency between the notarization fingerprint and the notarization data packet, as well as verifying the identity and permissions of the submitting node. Once the verification is successful, the notarization data packet is written into the candidate block, and consensus voting is conducted on the candidate block. When the candidate block receives confirmation from more than a preset proportion of consensus nodes in the consortium blockchain network, the candidate block is added to the blockchain ledger. Extract the transaction record identifier containing the evidence storage data packet from the newly added block as the evidence storage transaction identifier, and associate it with the corresponding fabric batch identifier for subsequent querying of the corresponding blockchain evidence storage transaction identifier through the fabric batch identifier.

7. The method according to claim 1, characterized in that, In response to traceability requests, the system retrieves and parses the notified transaction identifier from the consortium blockchain. It verifies batch consistency by reading the physical identifier of the school uniform. When the physical identifier's feature code matches the batch binding certificate on the blockchain, it confirms that the quality inspection report truly corresponds to the production batch. The system then returns the verification results and end-to-end quality data, including: Receive a traceability request, which includes the physical identifier or fabric batch identifier of the school uniform to be verified; The sub-identifier is obtained by parsing the graphic code on the school uniform using an identification device, or by directly obtaining the fabric batch identifier. The corresponding notarized transaction identifier is then retrieved from the consortium blockchain based on the sub-identifier or the fabric batch identifier. Based on the evidence storage transaction identifier, evidence storage data packets are extracted from the consortium blockchain ledger, and the time-series evidence chain, batch binding certificate, and verification conclusion are obtained by parsing the evidence storage data packets; Extract the sub-identifier mapping table for the batch from the batch binding certificate. If the sub-identifier exists in the mapping table, it is determined that the physical identifier matches the batch binding certificate, confirming that the quality inspection report of the school uniform truly corresponds to the production batch. If the sub-identifier does not exist, it is determined that there is a risk of batch substitution. The batch matching determination results are correlated with the verification conclusions to generate verification results that include both batch consistency determination and quality compliance determination. Process data for each production step in this batch is extracted from the chronological evidence chain and combined with the quality inspection report data in the verification conclusion to generate end-to-end quality data; The verification results and end-to-end quality data are returned to the traceability request initiator.

8. A blockchain-based quality traceability system for the entire school uniform production chain, used to implement the method as described in any one of claims 1-7, characterized in that, include: The data acquisition unit is used to acquire process data at various stages of the school uniform production chain. The hash encapsulation unit is used to perform hash encapsulation on the process data based on the trusted execution environment. The hash value of the current process node is obtained by combining the hash value of the previous node with the current data to form a time-series evidence chain. The batch binding unit is used to encode the fabric batch identifier hash value into an uncopyable physical identifier and apply it simultaneously to the production batch fabric and the submitted sample before the fabric is sent for inspection, thereby establishing a physical hierarchical binding between the quality inspection sample and the production batch and obtaining a batch binding certificate. The indicator comparison unit is used to automatically parse and compare the detection indicators of the quality inspection report data with the limit based on the standard knowledge graph, and generate verification conclusions. The consensus-based evidence storage unit is used to submit the time-series evidence chain, batch binding certificate and verification conclusion to the consortium blockchain network for consensus verification and generate a storage transaction identifier. The traceability verification unit is used to respond to traceability requests, retrieve and parse the evidence transaction identifier from the consortium blockchain, verify batch consistency by reading the physical identifier of the school uniform, and confirm that the quality inspection report truly corresponds to the production batch when the physical identifier feature code matches the batch binding certificate on the chain, and return the verification result and full-chain quality data.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images