Blockchain-based value flow conversion system for financial supply chain traceability

Abstract

Embodiments of the present disclosure disclose a value flow conversion system based on blockchain financial supply chain traceability. A specific implementation of the system includes: a blockchain node server, an Internet of Things data acquisition device, at least one bill identification upload terminal, a customer traceability terminal, and a cargo flow node, wherein: the bill identification upload terminal is configured to identify bill images; the Internet of Things data acquisition device is configured to acquire cargo state data corresponding to a cargo identifier; the blockchain node server is further configured to, in response to receiving cargo confirmation information corresponding to the cargo identifier sent by the cargo flow node, perform the following steps: generating respective digital value information; sending the digital value information to the bill identification upload terminal; and the customer traceability terminal is configured to, in response to receiving value transfer-in application information containing the digital value information sent by the bill identification upload terminal, perform a value flow conversion operation. This implementation improves user experience.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a value transfer system based on blockchain-based financial supply chain traceability. Background Technology

[0002] With the increasing complexity of global supply chain blockchains and the acceleration of digitalization, supply chain blockchains have become a key link in empowering the real economy. A value transfer system based on blockchain-based financial supply chain traceability provides a system for trusted transaction storage and digital asset transfer for all participants in the supply chain (such as financial institutions, core enterprises, upstream and downstream suppliers, and logistics providers). Currently, the typical approach for value transfer among supply chain participants is as follows: based on a centralized server architecture, each participant maintains an independent information system (such as enterprise ERP or bank credit systems), and completes value transfer through manual verification and point-to-point interfaces.

[0003] However, when using the above methods for value transfer, the following technical problems often arise: Based on a centralized server architecture, each participating party maintains its own independent information system (such as enterprise ERP or bank credit system). Value transfer is completed through manual verification and point-to-point interfaces. In order to ensure the security of value transfer (e.g., reduce fraud risk) and ensure that the value transfer is based on the actual flow of goods, it is often necessary to query the databases of different systems level by level for traceability. Cross-system information query efficiency is low, and the traceability process is time-consuming, resulting in slow response speed of value transfer operations and poor user experience.

[0004] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not form prior art known to those skilled in the art. Summary of the Invention

[0005] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of this disclosure propose a blockchain-based supply chain traceability value transfer system to address one or more of the technical intentions mentioned in the background section above.

[0007] Some embodiments of this disclosure provide a value transfer system based on blockchain-based financial supply chain traceability. The system includes: a blockchain node server, an IoT data acquisition device, at least one document identification and uploading terminal, a customer traceability terminal, and a goods flow node. Specifically: for each of the at least one document identification and uploading terminal, the terminal is configured to identify a document image, obtain document identification information including a goods identifier, and send the document identification information to the blockchain node server. The blockchain node server then performs block generation and on-chain processing on the document identification information to store it in a pre-created blockchain. The IoT data acquisition device is configured to collect goods status data corresponding to the goods identifier and send the goods status data to the blockchain. A blockchain node server is provided for writing cargo status data to the blockchain. The blockchain node server is further configured to, in response to receiving cargo receipt information corresponding to a cargo identifier from a cargo flow node, perform the following steps: generating various digital value information based on the document identification information corresponding to the cargo identifier; for each digital value information, sending it to one of the at least one document identification upload terminals corresponding to the digital value information; and configuring the customer traceability terminal to, in response to receiving a value transfer application information containing digital value information from the document identification upload terminal, perform supply chain process traceability verification processing based on the value transfer application information and the blockchain to execute value transfer operations.

[0008] The various embodiments disclosed above have the following beneficial effects: the value transfer system based on blockchain financial supply chain traceability in some embodiments of this disclosure improves the user experience. Specifically, the reason for the poor user experience is that, based on a centralized server architecture, each participating party maintains an independent information system (such as an enterprise ERP or bank credit system), and completes the value transfer through manual verification and point-to-point interfaces. In order to ensure the security of the value transfer (e.g., reduce the risk of fraud) and ensure that the value transfer is based on the actual flow of goods, it is often necessary to query the databases of different systems level by level for traceability. The efficiency of cross-system information query is low, and the traceability process is time-consuming, resulting in a slow response speed for value transfer operations and a poor user experience. Based on this, some embodiments of the value transfer system for blockchain-based financial supply chain traceability disclosed herein include: a blockchain node server, an IoT data acquisition device, at least one document identification and uploading terminal, a customer traceability terminal, and a goods flow node. Each of the at least one document identification and uploading terminal is configured to identify a document image, obtain document identification information including a goods identifier, and send the document identification information to the blockchain node server. The blockchain node server then performs block generation and on-chain processing on the document identification information to store it in a pre-created blockchain. Thus, document identification information can be obtained through the document identification and uploading terminal and uploaded to the blockchain. Decentralized data storage ensures the immutability and transparency of the document identification information, enhancing its credibility. Next, the IoT data acquisition device is configured to collect goods status data corresponding to the goods identifier and send the goods status data to the blockchain node server, so that the blockchain node server can write the goods status data into the blockchain. Therefore, status data corresponding to the goods identifier can be collected and stored in the blockchain. Next, the blockchain node server is further configured to, in response to receiving goods receipt information corresponding to the goods identifier from the goods flow node, execute the following steps: Based on the document identification information corresponding to the aforementioned goods identifier, generate various digital value information. For each digital value information, send it to one of the at least one document identification upload terminals corresponding to the digital value information. Thus, after goods delivery confirmation, document identification information can be automatically and in real-time converted into transferable digital value vouchers (i.e., digital value information) and distributed to the supplier (i.e., the document identification upload terminal), automating value generation and distribution and shortening the time required for manual rights confirmation and splitting processes.Subsequently, the customer traceability terminal is configured to respond to value transfer requests containing digitized value information sent by the document recognition and uploading terminal. Based on the aforementioned value transfer request information and the blockchain, it performs supply chain process traceability verification to execute value transfer operations. Thus, when value transfer is required, the customer traceability terminal can verify trusted data (documents, logistics, and rights confirmation) across the entire chain based on a unified blockchain, eliminating the need for manual queries and verification across levels and systems, thereby shortening the time required for the traceability process. When the document recognition and uploading terminal initiates a value transfer request, it can process the value transfer operation corresponding to the value transfer request information more quickly, ensuring that the value transfer is based on the actual flow of goods, thus improving the user experience. Attached Figure Description

[0009] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0010] Figure 1 This is an architecture diagram of an exemplary system for a value transfer system based on blockchain-based financial supply chain traceability, as disclosed herein. Detailed Implementation

[0011] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0012] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0013] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0014] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0015] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0016] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0017] Figure 1 An exemplary system architecture 100 of a heterogeneous data exchange system to which some embodiments of the present disclosure may be applied is shown.

[0018] like Figure 1 As shown, the system architecture 100 may include: a blockchain node server 103, an IoT data acquisition device 101, at least one document identification and uploading terminal 102, a customer traceability terminal 105, and a goods flow node 104. The IoT data acquisition device 101, at least one document identification and uploading terminal 102, and goods flow node 104 are connected to the blockchain node server 103 via a network. At least one document identification and uploading terminal 102 is connected to the customer traceability terminal 105 via a network. Network connections include various connection types, such as wired, wireless communication links, or fiber optic cables. Each of the at least one document identification and uploading terminal 102 can be a terminal used by the goods supplier (e.g., a mobile phone, industrial tablet computer, computer, or other computing device). The blockchain node server 103 can be a server that stores document identification information and goods status data to the blockchain. The goods flow node 104 can be a terminal used by the purchasing user (e.g., a core enterprise, i.e., an enterprise that connects with suppliers) (e.g., a mobile phone, industrial tablet computer, computer, or other computing device). The aforementioned customer traceability terminal 105 can be a supply chain finance platform that executes value transfer operations (such as payments and loans). The aforementioned document identification information includes identifying the goods, identifying the transferred value, identifying the outgoing user (i.e., the identifier of the goods supplier), and identifying the incoming user (i.e., the identifier of the purchasing user).

[0019] In some embodiments, for each of the at least one document identification and uploading terminal 102, the document identification and uploading terminal is configured to identify a document image, obtain document identification information including a goods identifier, and send the document identification information to the blockchain node server 103. The blockchain node server 103 then performs block generation and on-chain processing on the document identification information to store it in a pre-created blockchain. In practice, the blockchain node server 103 determines the hash value of the document identification information as the document hash value. Next, the current time can be obtained as a timestamp. The document identification hash value and the timestamp can then be combined to obtain constructed data. The constructed data is then encrypted to obtain a digital signature. The document hash value and the digital signature are then used to determine the transaction information. This transaction information is then broadcast to a pre-defined P2P network. After broadcasting, other nodes in the P2P network will receive the document hash value and digital signature. Each node first performs independent verification: it decrypts the digital signature using the sender's public key to obtain the original hash value, and simultaneously recalculates the hash of the received transaction data, comparing the two to verify the integrity and authenticity of the transaction. Verified transactions are placed in the node's "transaction pool" for processing. Next, the nodes responsible for packaging (such as miners or validators) select a batch of transactions from the pool, arrange them in the order they entered the pool, and generate a unique Merkle root hash using the Merkle tree algorithm. Then, the node constructs a new block. The block header contains key information such as the hash of the previous block, timestamp, and Merkle root hash, while the block body contains the list of selected transactions. The consensus phase then begins. Taking proof-of-work as an example, the packaging node needs to continuously adjust the random number and calculate the hash value of the block header until it finds a solution that meets the network difficulty requirements. Once found, the node immediately broadcasts the newly generated block to the entire network. Upon receiving the block, other nodes will verify its proof-of-work validity, the legitimacy of all transactions within the block, and the correctness of the Merkle root. Once verified, all nodes will unanimously append this new block to the end of their respective blockchain replicas. At this point, the transaction information containing the document hash value is permanently recorded in this block, which has achieved network consensus, completing its "on-chain" storage. As subsequent blocks are generated on top of this block, the record receives increasing confirmation, becoming immutable and irreversible. Afterward, any participant can quickly locate their block and transaction position by querying the document hash value and obtain publicly verifiable proof of existence.

[0020] In addressing the technical challenges mentioned above, the application scenario involves identifying and uploading user documents or documents uploaded during low-light periods in areas with outdated equipment (such as old phones or low-resolution cameras) for subsequent supply chain traceability and trade settlement. However, this process often presents the following technical problems: images of user documents or documents uploaded during low-light periods in these areas often suffer from noise, skewed text, and low contrast. Directly performing OCR recognition on these low-quality images results in low accuracy and low reliability of the uploaded information. Therefore, this application scenario requires the following characteristics: it must be suitable for identifying and uploading low-quality documents to ensure accurate and tamper-proof document information in supply chain traceability and trade settlement scenarios. To address these technical challenges, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the aforementioned document recognition and uploading terminal can recognize the document image through the following steps to obtain document recognition information containing the goods identifier: The first step is to receive document images uploaded by users. These document images can be images of documents such as invoices, bills of lading, and purchase orders.

[0021] The second step is to obtain the user's location information and the time the image of the uploaded document was taken. In practice, the user's location information can be obtained through base station positioning technology. This user location information can represent the user's geographical location.

[0022] The third step involves determining if the user's location information falls within the preset list of restricted upload device locations or if the uploaded document image occurs during a preset low-light period. This process performs quality checks on the document image to obtain an image quality score. In practice, the executing entity can use an image sharpness detection algorithm (e.g., the Tenengrad gradient method) to perform sharpness checks on the document image and obtain a sharpness value as the quality score. Each restricted upload device location in the preset list can represent an area with outdated devices, such as older phones or low-resolution cameras (e.g., some African countries or rural areas in South Asia). The low-light period can be from sunset to the early morning of the following day, such as 19:30-05:00.

[0023] Third, in response to determining that the image quality detection score is less than or equal to a preset threshold, the following enhanced recognition steps are performed: The first sub-step involves denoising the aforementioned document image to obtain a denoised document image. In practice, Gaussian filtering can be used to process the document image to obtain the denoised document image.

[0024] The second sub-step involves performing text skewing correction processing on the denoised document image to obtain a corrected document image. In practice, OpenCV-based text skewing correction techniques can be used to correct the text skewing in the denoised document image, resulting in the corrected document image. The corrected document image can be the image with the skewing text corrected.

[0025] The third sub-step involves performing adaptive region contrast stretching on the aforementioned corrected document image to obtain a text-enhanced document image. In practice, text region detection can be used to detect text regions in the corrected document image and obtain text region location information. This text region location information represents the position of the text within the corrected document image. This information may include coordinate points (e.g., the vertex coordinates of a rectangular region). Then, by employing local contrast enhancement techniques, the image corresponding to the text region location information in the corrected document image is enhanced, and the enhanced image is determined as the text-enhanced document image.

[0026] The fourth sub-step involves processing the enhanced text area document image to obtain document identification information containing the goods identifier. In practice, optical character recognition (OCR) is used to process the enhanced text area document image to obtain document identification information containing the goods identifier. The goods identifier can be the goods' number or name.

[0027] The fourth step is to send the document identification information to the aforementioned blockchain node server 103.

[0028] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem of "low accuracy in document information recognition and low reliability of on-chain information." Factors leading to low accuracy in document information recognition and low reliability of on-chain information often include: user documents from areas with outdated equipment (e.g., old mobile phones, low-pixel cameras) or document images uploaded during low-light periods contain noise, skewed text, and low contrast. Directly performing OCR recognition on these low-quality images results in low accuracy in document information recognition and low reliability of on-chain information. Solving these factors can improve both the accuracy of document information recognition and the reliability of on-chain information. To achieve this, firstly, the document image uploaded by the user is received. Then, the user's location information and the upload time of the document image are obtained. Next, in response to determining that the user's location information is in a preset list of restricted upload device locations or that the upload time of the document image is within a preset low-light period, quality detection processing is performed on the document image to obtain an image quality detection score. Therefore, image quality can be detected when the user's location information falls within a restricted upload device area (i.e., an area with outdated devices such as old phones or low-resolution cameras) or when the upload time occurs during a low-light period, resulting in a line quality detection score to assess image clarity. Next, in response to determining that the image quality detection score is less than or equal to a preset threshold, the following enhancement recognition steps are performed: First, the document image is denoised to obtain a denoised document image. This removes noise from the document image, preventing interference with subsequent text recognition. Then, the denoised document image undergoes text tilt correction processing to obtain a corrected document image. This corrects tilted text in the image. Afterward, the corrected document image undergoes adaptive region contrast stretching processing to obtain a text-enhanced document image. This enhances the edge sharpness of text strokes and deepens the difference between text and background, making subsequent accurate text recognition easier. Finally, the text-enhanced document image undergoes recognition processing to obtain document recognition information including goods identification. Therefore, through the above steps, noise removal, text tilt correction, and enhancement of the difference between text and background can be performed even when image quality is poor. Then, the enhanced document image with improved text regions is used for recognition, resulting in document information with high recognition accuracy. Next, the document recognition information is sent to the aforementioned blockchain node server. This ensures that highly accurate document recognition information is sent to the blockchain node server, thereby improving the authenticity and reliability of the information uploaded to the blockchain.

[0029] In addressing the aforementioned technical challenges by employing technological solutions, the application scenario—identifying sensitive documents containing data within cross-border trade or international supply chains and executing value transfer operations based on blockchain technology—often presents the following technical issues: Documents collected from different countries and regions (such as invoices, bills of lading, and certificates of origin) often contain diverse languages ​​and a large amount of sensitive information, including supplier and customer identification and bank account details. Directly sending the identified document information to the blockchain node server for on-chain storage may fail to generate digitized value information due to language differences, leading to the failure of subsequent value transfer operations. Furthermore, directly sending documents containing sensitive information to the blockchain node server or using centralized key storage is susceptible to single-point failures (such as server breaches) that could invalidate all encrypted data. The risk of key reuse or leakage is high, easily leading to the leakage of sensitive information and resulting in low information transmission security. Therefore, this application scenario requires the following characteristics: language standardization (e.g., uniform conversion to the contract's primary language or English) and desensitization of sensitive information during document information sending and uploading. Faced with these technical challenges, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the above-mentioned document recognition and uploading terminal can proceed through the following steps: The first step is to perform language type recognition processing on the aforementioned document identification information to obtain language type recognition information. In practice, the document identification information can be input into a preset language detector (such as an OpenL language detection tool) for language type recognition processing to obtain language type recognition information. This language type recognition information can represent the language type (e.g., Japanese, Korean, etc.).

[0030] The second step involves performing language conversion processing on the document identification information based on the aforementioned language type identification information to obtain converted document identification information. In practice, in response to the determination that the language type represented by the aforementioned language type identification information is not a preset language type, the executing entity can use neural network machine translation technology to convert the document identification information from the language type represented by the language type identification information to a preset language type, and then identify the converted document identification information as the converted document identification information.

[0031] The third step involves performing sensitive information detection processing on the aforementioned conversion document identification information to obtain sensitive detection information. In practice, this can be done by using a regular expression library to match sensitive information (such as ID card numbers, passport numbers, bank account numbers, phone numbers, etc.) to obtain sensitive detection information. Optionally, text-based sensitive information identification technology can be used to perform sensitive information detection processing on the conversion document identification information to obtain sensitive detection information. The aforementioned sensitive detection information can be the start and end positions for locating sensitive information (e.g., characters 120 to 137).

[0032] The fourth step is to identify the text information at the location specified by the sensitive detection information as the sensitive information to be de-identified; The fifth step involves dividing the sensitive information to be de-identified into individual strings based on a preset delimiter. For example, the sensitive information to be de-identified could be "Zhang San, 150×××00××3". The preset delimiter can be a comma, and the individual strings could be "Zhang San" or "150×××00××3".

[0033] Step 6: For each of the above string data, perform the following steps: The first sub-step involves extracting key features and performing pattern analysis on the aforementioned string data to obtain data fingerprint feature information. In practice, the length and structure of the string data can be extracted to obtain data fingerprint feature information. For example, if the data length and structure of "150×××00××3" are "length 11, with a fixed number segment 15", then the data fingerprint feature information can be "length 11, with a fixed number segment 15".

[0034] The second sub-step involves inputting the aforementioned data fingerprint feature information into the type classification rule engine to obtain the type identifier corresponding to the aforementioned string data. The type classification rule engine can be a rule engine that receives raw data (such as a string) and its features, analyzes and judges it according to a series of predefined logical rules, and finally outputs the classification result of the data (such as "phone number" or "ID number").

[0035] The third sub-step involves querying the encryption interval information corresponding to the aforementioned type identifier from a preset encryption interval mapping table. This preset encryption interval mapping table can be a mapping table that records a one-to-one correspondence between the type identifier and the encryption interval information. The encryption interval information can represent a string range, such as the entire string or the first four characters of a string.

[0036] The fourth sub-step is to determine the substring corresponding to the encrypted interval information mentioned above in the string data; The fifth sub-step involves obtaining the master key and fragmenting it to obtain individual key fragments. For example, the master key could be "123456", and the individual key fragments could be "12", "34", and "56".

[0037] The sixth sub-step involves reversibly encrypting the substring based on the master key to obtain the encrypted substring. In practice, a symmetric encryption algorithm can be used to encrypt the substring based on the master key to obtain the encrypted substring.

[0038] The seventh sub-step involves replacing the substrings in the string data with encrypted substrings, and identifying the replaced string data as the de-identified document identification sub-information.

[0039] The eighth sub-step involves destroying the aforementioned master key, randomly selecting three key fragments from each fragment, and storing the three key fragments respectively in the document identification upload terminal, the blockchain node server, and the customer traceability terminal corresponding to the document identification information. When all three key fragments exist simultaneously, the master key is restored using a preset reconstruction algorithm to decrypt the anonymized document identification sub-information. Step 7: Identify each de-identified document identification sub-information as the de-identified document identification information.

[0040] In practice, the aforementioned executing entity can encrypt the text information at the location specified by the sensitive detection information in the conversion document identification information to obtain encrypted text information. Then, the encrypted text information replaces the aforementioned text information, and the replaced conversion document identification information is determined as the desensitized document identification information.

[0041] Step 8: Send the above-mentioned de-identified document identification information to the above-mentioned blockchain node server 103.

[0042] The above technical solution, combining the steps executed by the blockchain node server and the value transfer operations and related content performed by the customer traceability terminal, serves as an inventive point in this disclosure, solving the technical problem of "value transfer operation failure and low information transmission security." Factors leading to value transfer operation failure and low information transmission security often include: documents collected from different countries and regions (such as invoices, bills of lading, and certificates of origin) contain diverse languages ​​and a large amount of sensitive information such as supplier and customer identification and bank account details. Directly sending the identified document information to the blockchain node server for on-chain storage may fail to generate digitized value information due to language differences, leading to subsequent value transfer operation failures. Furthermore, directly sending documents containing sensitive information to the blockchain node server, or using centralized key storage, is prone to single-point failures (such as server breaches) that could invalidate all encrypted data. The risk of key reuse or leakage is high, easily leading to the leakage of sensitive information and low information transmission security. Solving these factors can reduce value transfer operation failures and improve information transmission security. To achieve this effect, firstly, language type recognition processing is performed on the aforementioned document identification information to obtain language type recognition information. This allows for the detection of the language type of the document identification information. Then, based on the language type recognition information, language conversion processing is performed on the document identification information to obtain converted document identification information. Multilingual documents are uniformly converted into a system-preset standard language (such as English or Chinese), eliminating semantic barriers for blockchain node servers when processing multilingual data. Next, sensitive information identification and detection processing is performed on the converted document identification information to obtain sensitive detection information. This allows for the location of privacy and sensitive data that need protection within the document identification information. Next, the text information at the location specified by the sensitive detection information is identified as sensitive information to be de-identified. Then, according to a preset desegmenter, the sensitive information to be de-identified is segmented to obtain individual string data. Then, for each string data, the following steps are performed: First, key feature extraction and pattern analysis are performed on the string data to obtain data fingerprint feature information. This data fingerprint feature information is input into the type classification rule engine to obtain the type identifier corresponding to the string data. The second sub-step involves querying the encryption interval information corresponding to the aforementioned type identifier from a preset encryption interval mapping table. The third sub-step involves determining the substrings corresponding to the aforementioned encryption interval information within the string data. The master key is obtained, and it is fragmented to obtain individual key fragments. The fourth sub-step involves performing reversible encryption on the substrings based on the master key to obtain encrypted substrings. The substrings in the string data are then replaced with the encrypted substrings, and the replaced string data is identified as the de-identified document identification sub-information.The fifth sub-step involves destroying the aforementioned master key, randomly selecting three key fragments from each fragment, and storing these three fragments respectively at the document identification upload terminal, the blockchain node server, and the customer traceability terminal corresponding to the document identification information. If all three key fragments exist simultaneously, a preset reconstruction algorithm is used to recover the master key, thereby decrypting the anonymized document identification sub-information. Each anonymized document identification sub-information is then identified as the anonymized document identification information. Therefore, reversible technologies such as field-level encryption are used to replace sensitive content with secure ciphertext. By dividing the sensitive information to be de-identified into string data and performing a refined de-identification process on each string (including type identification, encryption range location, fragmented storage of the master key, and dynamic reconstruction), the de-identified document identification information sent to the blockchain and various terminals achieves decentralized key control (ensuring that if any node is compromised, the attacker cannot obtain the complete master key, preventing all encrypted data from becoming invalid due to a single point of failure). Combined with the design of distributed key fragment storage on the document identification upload terminal, blockchain node server, and customer traceability terminal, and the threshold mechanism of "the master key can only be reconstructed if all three fragments exist simultaneously," global risks caused by key reuse or leakage from a single node are prevented. Simultaneously, the encrypted substrings within each string data are independently and reversibly encrypted, and the original master key is destroyed, retaining only the fragmented key fragments, reducing the probability of key leakage during transmission and storage. The aforementioned de-identified document identification information is then sent to the aforementioned blockchain node server. Because of the pre-processing involving language standardization and reversible desensitization of sensitive information, the data sent to the blockchain is semantically unified (ensuring that subsequent smart contracts can accurately parse and generate digital value information). Combined with the steps executed by the blockchain node server and the content of the value transfer operations performed by the customer traceability end, digital value information can be generated, reducing the failure rate of subsequent value transfer operations. Simultaneously, reversible desensitization of document identification information reduces the leakage of sensitive information during the transmission of document identification information, improving the security of document identification information transmission.

[0043] In some embodiments, the IoT data acquisition device 101 is configured to acquire cargo status data corresponding to cargo identifiers and send the cargo status data to the blockchain node server 103 so that the blockchain node server 103 can write the cargo status data to the blockchain.

[0044] In some optional implementations of certain embodiments, the IoT data acquisition device 101 described above can acquire cargo status data corresponding to the cargo identifier through the following steps, and send the cargo status data to the blockchain node server 103 described above: The first step is to collect the location information of the goods corresponding to the goods identification using a positioning device. In practice, the implementing entity can determine the dynamic collection frequency based on the goods identification, and then use the positioning device to collect the location information of the goods corresponding to the goods identification based on the dynamic collection frequency. For example, when the type of goods identification belongs to a first preset type (e.g., valuable goods), the first preset collection frequency (e.g., collecting once every ten minutes) can be determined as the dynamic collection frequency. When the type of goods identification belongs to a second preset type (e.g., ordinary goods), the second preset collection frequency (e.g., collecting once every hour) can be determined as the dynamic collection frequency. The first preset collection frequency is faster than the second preset collection frequency.

[0045] The second step involves using video surveillance devices to collect video footage of the goods corresponding to their identification tags. In practice, the implementing entity can determine the dynamic acquisition frequency based on the goods identification tags, and then collect video footage of the goods corresponding to those tags according to that frequency.

[0046] The third step is to store the aforementioned cargo video to a preset cloud storage device to generate cloud storage location information corresponding to the cargo video. This cloud storage location information indicates the storage location of the cargo video within the preset cloud storage device. In practice, the cargo video can be stored on a preset cloud storage device, and the storage location returned by the preset cloud storage device can be used as the cloud storage location information. The preset cloud storage device can be a cloud storage server.

[0047] The fourth step is to determine the above cargo location information and the above cloud storage location information as cargo status data.

[0048] The fifth step is to send the aforementioned cargo status data to the blockchain node server 103.

[0049] In some optional implementations of certain embodiments, the IoT data acquisition device 101 described above can store the cargo video to a preset cloud storage device through the following steps: The first step is to compress the above cargo video to obtain a compressed cargo video. In practice, H.264 compression can be used to compress the cargo video to obtain a compressed cargo video.

[0050] The second step is to encrypt the compressed cargo video and send it to a pre-set cloud storage device for storage. In practice, symmetric encryption technology can be used to encrypt the compressed cargo video. Then, the encrypted compressed cargo video is sent to the pre-set cloud storage device for storage.

[0051] In some optional implementations of certain embodiments, the blockchain node server 103 can be configured to write cargo status data to the blockchain through the following steps: First, the cargo identifier corresponding to the above cargo status data is determined as the target cargo identifier.

[0052] Second, the device identifier of the positioning device corresponding to the target cargo identifier is used as the first device identifier.

[0053] Third, obtain the device identifier of the video surveillance device corresponding to the aforementioned target cargo identifier as the second device identifier.

[0054] Fourth, the first device identifier and the second device identifier are spliced ​​together to obtain the data acquisition device identifier.

[0055] Fifth, generate a device fingerprint based on the data acquisition device identifier. In practice, the aforementioned execution entity can use a hash algorithm to generate a hash value corresponding to the data acquisition device identifier as the device fingerprint.

[0056] Sixth, bind the device fingerprint and the preset key to generate a derived session key. In practice, the device fingerprint and the preset key can be input into a preset key derivation function (e.g., the KDF algorithm) to obtain the derived session key.

[0057] Seventh, based on the derived session key, the above-mentioned cargo status data is encrypted to obtain encrypted cargo status data. In practice, a preset symmetric encryption algorithm (such as the DES encryption algorithm) can be used to encrypt the above-mentioned cargo status data based on the derived session key to obtain encrypted cargo status data.

[0058] Eighth, generate a tamper-proof hash value corresponding to the aforementioned cargo status data. In practice, the cargo status data can be input into a preset hash function to obtain a hash value, which can then be used as the tamper-proof hash value.

[0059] Ninth, write the aforementioned tamper-proof hash value and the aforementioned encrypted cargo status data into the aforementioned blockchain.

[0060] In some embodiments, the blockchain node server 103 is further configured to perform the following steps in response to receiving goods receipt information corresponding to the goods identifier sent by the goods flow node 104: First, based on the document identification information corresponding to the aforementioned goods identifier, various digital value information is generated. Each digital value information corresponds to one document identification information among the various document identification information. The digital value information includes the goods identifier, circulation value, outgoing user identifier, and incoming user identifier. The document identification information includes identifying the goods identifier, identifying the circulation value, identifying the outgoing user identifier, and identifying the incoming user identifier. The identified circulation value can represent the value of the goods corresponding to the identified goods identifier. The identified outgoing user identifier can be the identifier of the user supplying the aforementioned goods. The identified incoming user identifier can be the identifier of the purchasing user who purchases the aforementioned goods. For each document identification information, the document identification information can be input into a smart contract to generate digital value information. Specifically, the smart contract can map document identification information (such as identifying the goods identifier and identifying the circulation value) to digital value information (goods identifier and circulation value) through preset rules. The digital value information can be a data structure generated after standardizing the document identification information according to the preset rules of the blockchain smart contract; the digital value information can also be accounts receivable. The aforementioned preset rules can be used to extract and identify goods identifiers, circulation value, outflow user identifiers, and inflow user identifiers from text using regular expressions or keyword matching, and map them to the goods identifiers, circulation value, outflow user identifiers, and inflow user identifiers in the digital value information, respectively.

[0061] Second, for each piece of digital value information, the aforementioned digital value information is sent to one of the document recognition and upload terminals corresponding to the digital value information in the at least one document recognition and upload terminal 102.

[0062] In some optional implementations of certain embodiments, the blockchain node server 103 may send the digital value information to one of the document recognition and uploading terminals corresponding to the digital value information in the at least one document recognition and uploading terminal 102 through the following steps: The first step is to identify the document identification information that corresponds to the aforementioned digital value information from the document identification information above as the target document identification information.

[0063] The second step is to identify the document image corresponding to the above-mentioned target document identification information as the target document image.

[0064] The third step is to identify at least one document recognition and uploading terminal 102 that has uploaded the target document image as the target document recognition and uploading terminal.

[0065] The third step is to send the aforementioned digital value information to the target document recognition and uploading terminal.

[0066] In some embodiments, the database server 103 may be further configured such that the customer traceability terminal 105 is configured to, in response to receiving a value transfer application information containing digital value information sent by the document identification and uploading terminal, perform supply chain process traceability verification processing based on the value transfer application information and the blockchain, in order to execute value transfer operations. The value transfer application information may represent a value inflow application (e.g., a financing application) containing digital value information.

[0067] In some optional implementations of certain embodiments, the aforementioned customer traceability terminal 105 can perform supply chain process traceability verification based on the aforementioned value transfer application information and the aforementioned blockchain through the following steps to execute value transfer operations: The first step is to transfer the aforementioned value into the application information, including the goods identifier, and identify it as the goods to be traced.

[0068] The second step is to query the document identification information corresponding to the above-mentioned traceable goods identifier from the blockchain to obtain the query information.

[0069] The third step involves freezing the transfer operation corresponding to the aforementioned value transfer application information upon confirming that the query information is empty, and sending a warning message indicating abnormal transfer to the preset approval terminal. This warning message can be text-based.

[0070] Fourth step: In response to determining that the query information includes at least one document identification information, the at least one document identification information is determined as the document identification information set to be verified.

[0071] The fifth step is to determine the outflowing user identifier and the inflowing user identifier included in the digital value information in the value transfer application information.

[0072] The sixth step is to identify the document identification information that is the same as both the outgoing user identifier and the incoming user identifier in the above-mentioned document identification information set as the source document identification information.

[0073] Step 7: Responding to the fact that the digitized value information in the value transfer application information, including the transfer value, is the same as the identification transfer value included in the traceability document identification information, a value transfer operation is performed based on the aforementioned value transfer application information. In practice, the aforementioned executing entity can modify the credit limit or fund balance corresponding to the outgoing user identifier in the preset ledger database (such as Oracle DB, mainframe database) based on the transfer value, outgoing user identifier, and incoming user identifier included in the value transfer application information, and simultaneously modify the credit limit or fund balance corresponding to the incoming user identifier in the preset ledger database (such as Oracle DB, mainframe database). As an example, the transfer value can be reduced by decreasing the credit limit or fund balance corresponding to the outgoing user identifier, and the transfer value can be increased by increasing the credit limit or fund balance corresponding to the incoming user identifier.

[0074] In addressing the technical problems mentioned above, and considering the specific application scenario: The fixed field of view of IoT data acquisition devices deployed on goods corresponding to cargo tags results in continuous background interference (such as the interior walls of the vehicle, warehouse shelves, and the floor) in the video feed. This presents further challenges when receiving fragile items in a warehouse. Traditional methods, such as manual unpacking or video recording of the goods in the warehouse for damage detection, are often used during the receiving process. However, warehouses struggle to accurately determine if fragile items were damaged during transport and find it difficult to record and trace the inspection process. Furthermore, the fixed field of view of the IoT data acquisition devices on the goods, coupled with continuous background interference, leads to low accuracy in damage detection. To address these issues, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the aforementioned cargo flow node 104 is further configured to perform the following steps: The first step involves retrieving at least one cargo status data point corresponding to a preset cargo identifier from the blockchain as at least one anomaly detection cargo status data point. Each anomaly detection cargo status data point includes a cargo video and cargo location information, and each anomaly detection cargo status data point has a corresponding collection time. The aforementioned at least one cargo status data point can represent the status of the cargo corresponding to the preset cargo identifier during transportation.

[0075] The second step involves identifying the anomaly detected cargo status data from at least one of the aforementioned cargo status data sets whose acquisition time meets a preset condition as the target anomaly detected cargo status data. The preset condition can be that the time interval between the acquisition time and the current time is the shortest. The target anomaly detected cargo status data represents the latest status of the cargo.

[0076] The third step is to determine the cargo location information included in the target anomaly detection cargo status data as the location information to be detected.

[0077] Fourth step: In response to determining that the location represented by the location information to be detected is the warehouse location corresponding to the above-mentioned preset cargo identifier, at least one cargo video included in at least one abnormal cargo status data is determined as at least one cargo video to be detected.

[0078] Fifth, in response to determining that the preset cargo identifier type is a fragile item identifier, frame extraction processing is performed on the at least one cargo video to be detected to obtain a cargo video frame sequence. In practice, for each of the at least one cargo video to be detected, one video frame can be extracted as the extracted video frame. Then, the extracted video frames are sorted according to the chronological order of their corresponding acquisition times to obtain the extracted video frame sequence as the cargo video frame sequence to be detected.

[0079] The sixth step involves performing background removal processing on each of the above-mentioned video frames of the goods to be detected to obtain the target video frame. In practice, the video frames of the goods to be detected can be input into an image background removal model to obtain the background-removed video frames of the goods to be detected as the target video frames.

[0080] The seventh step is to arrange the obtained target cargo video frames to obtain the target cargo video frame sequence.

[0081] Step 8: Input the aforementioned target cargo video frame sequence into a pre-trained item damage detection model to obtain damage detection information. This item damage detection model can be a Convolutional Neural Network (CNN) that takes video frame sequences as input and damage detection information as output. The damage detection information can be represented by Boolean values ​​(True / False), for example, "True" indicates the item is undamaged, and "False" indicates the item was damaged en route. As an example, at least one sample cargo video frame sequence and the corresponding sample damage detection information for each sample cargo video frame sequence can be obtained first. Then, using each sample cargo video frame sequence as input and the corresponding damage detection information as the desired output, the item damage detection model is trained.

[0082] Step 9: In response to the determination that the damage detection information indicates the item is undamaged, the goods receipt information corresponding to the preset goods identifier is sent to the blockchain node server 103. The goods receipt information can be pre-set text information indicating that the goods have been received by the node 104.

[0083] Step 10: In response to determining that the above-mentioned damage detection information indicates that the goods are damaged in transit, the transportation damage information corresponding to the above-mentioned preset cargo identifier is sent and stored to the blockchain node server 103.

[0084] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem mentioned in the background art: "difficulty in determining whether goods have been damaged during transportation and low accuracy of damage detection." Factors leading to difficulty in determining whether goods have been damaged during transportation and low accuracy of damage detection often include: in the goods receiving and inspection stage, especially for fragile items, traditional manual unpacking and inspection or obtaining video of the items in the warehouse for damage detection are commonly used. When receiving fragile items, the warehouse finds it difficult to accurately determine whether the items have been damaged during transportation, and it is difficult to record and trace the inspection process. Furthermore, the goods video data may contain a large amount of background interference, resulting in low accuracy of damage detection. If the above factors are solved, the effect of detecting whether goods have been damaged during transportation and improving the accuracy of damage detection can be achieved. To achieve this effect, firstly, at least one goods status data corresponding to a preset goods identifier is obtained from the blockchain as at least one anomaly detection goods status data. Each anomaly detection goods status data includes goods video and goods location information, and each anomaly detection goods status data has a corresponding collection time. Thus, at least one anomaly detection goods status data representing the status of the goods corresponding to the preset goods identifier during transportation can be obtained. Then, the abnormal cargo status data whose acquisition time meets the preset condition from the above at least one cargo status data is determined as the target abnormal cargo status data. Thus, the target abnormal cargo status data representing the latest cargo status can be obtained. Next, the cargo location information included in the target abnormal cargo status data is determined as the location information to be detected. Then, in response to determining that the location represented by the location information to be detected is the warehouse location corresponding to the above preset cargo identifier, at least one cargo video included in the at least one abnormal cargo status data is determined as at least one cargo video to be detected. Thus, at least one cargo video to be detected during the transportation of the cargo can be determined when the cargo arrives at the warehouse. Next, in response to determining that the identifier type of the above preset cargo identifier is a fragile item identifier, frame extraction processing is performed on the above at least one cargo video to be detected to obtain a cargo video frame sequence to be detected. Thus, a cargo video frame sequence to be detected during the transportation of the cargo can be extracted. Next, background removal processing is performed on each cargo video frame in the cargo video frame sequence to obtain the target cargo video frame. Thus, background removal of the cargo video frame to be detected can reduce background interference during subsequent damage detection. Next, the obtained target cargo video frames are arranged to obtain a target cargo video frame sequence. Then, this target cargo video frame sequence is input into a pre-trained item damage detection model to obtain damage detection information. Thus, item damage detection can be performed on the video frame sequence of the goods to be detected during transportation, realizing the detection of whether the item has been damaged during transportation.Next, in response to determining that the damage detection information indicates the item is undamaged, the goods receipt information corresponding to the preset goods identifier is sent to the blockchain node server, and the damage detection information corresponding to the preset goods identifier is sent and stored on the blockchain node server. Then, in response to determining that the damage detection information indicates the item is damaged, the in-transit damage information corresponding to the preset goods identifier is sent and stored on the blockchain node server. Because by obtaining spatiotemporally tagged data from the blockchain, performing video frame extraction and background removal on fragile items, and obtaining a target goods video frame sequence with the background removed, which is then input into the item damage detection model for analysis, background interference is reduced, thereby improving the accuracy of judging whether goods (especially fragile items) have been damaged during transportation. Simultaneously, the entire judgment process is based on data on the blockchain, enabling the detection process and results to be accurately recorded and traced.

[0085] The various embodiments disclosed above have the following beneficial effects: the value transfer system based on blockchain financial supply chain traceability in some embodiments of this disclosure improves the user experience. Specifically, the reason for the poor user experience is that, based on a centralized server architecture, each participating party maintains an independent information system (such as an enterprise ERP or bank credit system), and completes the value transfer through manual verification and point-to-point interfaces. In order to ensure the security of the value transfer (e.g., reduce the risk of fraud) and ensure that the value transfer is based on the actual flow of goods, it is often necessary to query the databases of different systems level by level for traceability. The efficiency of cross-system information query is low, and the traceability process is time-consuming, resulting in a slow response speed for value transfer operations and a poor user experience. Based on this, some embodiments of the value transfer system for blockchain-based financial supply chain traceability disclosed herein include: a blockchain node server, an IoT data acquisition device, at least one document identification and uploading terminal, a customer traceability terminal, and a goods flow node. Each of the at least one document identification and uploading terminal is configured to identify a document image, obtain document identification information including a goods identifier, and send the document identification information to the blockchain node server. The blockchain node server then performs block generation and on-chain processing on the document identification information to store it in a pre-created blockchain. Thus, document identification information can be obtained through the document identification and uploading terminal and uploaded to the blockchain. Decentralized data storage ensures the immutability and transparency of the document identification information, enhancing its credibility. Next, the IoT data acquisition device is configured to collect goods status data corresponding to the goods identifier and send the goods status data to the blockchain node server, so that the blockchain node server can write the goods status data into the blockchain. This allows for the real-time collection of status data corresponding to the goods identifier and its storage on the blockchain. Next, the blockchain node server is further configured to, in response to receiving goods receipt information corresponding to the goods identifier from the goods flow node, execute the following steps: Based on the document identification information corresponding to the aforementioned goods identifier, generate various digital value information. For each of the digital value information, send it to one of the at least one document identification upload terminals corresponding to the digital value information. Thus, after goods delivery confirmation, document identification information can be automatically and in real-time converted into transferable digital value vouchers (i.e., digital value information) and distributed to the supplier (i.e., the document identification upload terminal), automating value generation and distribution and shortening the time required for manual rights confirmation and splitting processes.Subsequently, the customer traceability terminal is configured to respond to value transfer requests containing digitized value information sent by the document recognition and uploading terminal. Based on the aforementioned value transfer request information and the blockchain, it performs supply chain process traceability verification to execute value transfer operations. Thus, when value transfer is required, the customer traceability terminal can verify trusted data (documents, logistics, and rights confirmation) across the entire chain based on a unified blockchain, eliminating the need for manual queries and verification across levels and systems, thereby shortening the time required for the traceability process. When the document recognition and uploading terminal initiates a value transfer request, it can process the value transfer operation corresponding to the value transfer request information more quickly, ensuring that the value transfer is based on the actual flow of goods, thus improving the user experience.

[0086] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of technical features, but should also cover other technical solutions formed by arbitrary combinations of technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A value transfer system based on blockchain for financial supply chain traceability, comprising: The system includes a blockchain node server, IoT data acquisition equipment, at least one document recognition and upload terminal, a customer traceability terminal, and a goods flow node, among which: For each of at least one document recognition and uploading terminal, the document recognition and uploading terminal is configured to recognize a document image, obtain document recognition information containing a goods identifier, and send the document recognition information to the blockchain node server, so that the blockchain node server can perform block generation and on-chain processing on the document recognition information, and store the document recognition information in a pre-created blockchain. The IoT data acquisition device is configured to collect cargo status data corresponding to cargo identifiers and send the cargo status data to the blockchain node server so that the blockchain node server can write the cargo status data to the blockchain. The blockchain node server is further configured to perform the following steps in response to receiving a goods receipt confirmation message corresponding to a goods identifier sent by the goods flow node: Based on the document identification information corresponding to the goods identifier, generate various digital value information; For each piece of digital value information, the digital value information is sent to the document recognition and upload terminal corresponding to the digital value information in the at least one document recognition and upload terminal; The customer traceability terminal is configured to respond to a value transfer application information containing digital value information sent by the document recognition and uploading terminal, and to perform supply chain process traceability verification based on the value transfer application information and the blockchain to execute value transfer operations.

2. The value transfer system based on blockchain financial supply chain traceability according to claim 1, wherein, The digital value information includes goods identification, circulation value, outgoing user identification, and incoming user identification; the document identification information includes identifying goods identification, circulation value, outgoing user identification, and incoming user identification; and the customer traceability terminal is further configured to: The value transferred into the application information includes digital value information such as the goods identifier, which is identified as the goods identifier to be traced. The query information is obtained by querying the document identification information corresponding to the identifier of the goods to be traced from the blockchain; In response to the determination that the query information is empty, the transfer operation corresponding to the value transfer application information is frozen, and an early warning message indicating abnormal transfer is sent to the preset approval terminal. In response to determining that the query information includes at least one document identification information, the at least one document identification information is identified as a set of document identification information to be verified. The digital value information in the value transfer application information includes the outflow user identifier and the inflow user identifier; The document identification information in the document identification information set to be verified that is identical to both the outgoing user identifier and the incoming user identifier is identified as the source document identification information. In response to the fact that the digital value information in the value transfer application information, including the transfer value, is the same as the identification transfer value included in the traceability document identification information, a value transfer operation is performed based on the value transfer application information.

3. The value transfer system based on blockchain financial supply chain traceability according to claim 1, wherein, The goods corresponding to the cargo identification are equipped with the IoT data acquisition device, which includes a positioning device and a video monitoring device. The IoT data acquisition device is further configured to: The location information of the goods is collected by the positioning device and is used as the location information of the goods corresponding to the goods identification. Video of the goods corresponding to their identification tags is collected using video surveillance devices; The cargo video is stored in a preset cloud storage device to generate cloud storage location information corresponding to the cargo video; The cargo location information and the cloud storage location information are determined as cargo status data; The cargo status data is sent to the blockchain node server.

4. The value transfer system based on blockchain financial supply chain traceability according to claim 1, wherein, The IoT data acquisition device is further configured to: The cargo video is compressed to obtain a compressed cargo video; The compressed cargo video is encrypted and sent to a preset cloud storage device for storage.

5. The value transfer system based on blockchain financial supply chain traceability according to claim 1, wherein, The goods corresponding to the cargo identifier are equipped with the IoT data acquisition device, which includes a positioning device and a video monitoring device. The blockchain node server is further configured to: The cargo identifier corresponding to the cargo status data is determined as the target cargo identifier; The device identifier of the positioning device corresponding to the target cargo identifier is used as the first device identifier; Obtain the device identifier of the video surveillance device corresponding to the target cargo identifier as the second device identifier; The first device identifier and the second device identifier are concatenated to obtain the data acquisition device identifier; Generate a device fingerprint based on the collected device identifier; Bind the device fingerprint and the preset key to generate a derived session key; Based on the derived session key, the cargo status data is encrypted to obtain encrypted cargo status data. Generate a tamper-proof hash value corresponding to the cargo status data; The tamper-proof hash value and the encrypted cargo status data are written to the blockchain.

6. The value transfer system based on blockchain financial supply chain traceability according to claim 1, wherein, The digitized value information corresponds to one document identification information among various document identification information, and the blockchain node server is further configured to: The document identification information that corresponds to the digital value information in each of the document identification information is determined as the target document identification information; The identification document image corresponding to the target document identification information is determined as the target document image; The document recognition and uploading terminal that uploads the target document image in at least one document recognition and uploading terminal is identified as the target document recognition and uploading terminal; The digitized value information is sent to the target document recognition and uploading terminal.

Images