Two-dimensional code identification-based exit-entry certificate authenticity identification method and system, and medium

Abstract

The invention relates to an exit-entry certificate authenticity identification method and system based on two-dimensional code identification, and a medium. The method comprises the following steps: constructing a cross verification rule base comprising logical relationship information between two-dimensional code information of the entry-exit certificate and signing and issuing information of the entry-exit certificate; acquiring a data page image of an entry-exit certificate to be detected; identifying a two-dimensional code in the preprocessed data page image by using a two-dimensional code identification engine; identifying signing and issuing information in the preprocessed data page image; according to the dynamic two-dimensional code verification library, the dynamic blacklist library and the identified signing and issuing information, uniqueness verification and legality verification are carried out on the identified two-dimensional code information; comparing the corresponding relationship between the identified two-dimensional code information and the identified signing and issuing information according to a cross verification rule base; and judging the authenticity of the entry-exit certificate and generating a report. According to the invention, true and false identification is carried out through three dimensions, the identification accuracy is high, and high-risk false and false situations such as same codes, different certificates and tampering fields can be accurately identified.

Description

Technical Field

[0001] This disclosure relates to the field of document authentication technology, and in particular to methods, systems and media for authenticating entry and exit documents based on QR code recognition. Background Technology

[0002] With the increasing frequency of cross-border travel, passports and other travel documents have become the core basis for identity verification and border control. To effectively prevent document forgery and ensure the authenticity of personal identities, passport verification is gradually developing towards intelligent recognition and multimodal verification. As the demand for global immigration management continues to grow, countries are constantly evolving their technologies for passport and travel document security and authenticity verification. Currently, document anti-counterfeiting and verification methods mainly include physical anti-counterfeiting measures, electronic chip authentication, and artificial intelligence-assisted recognition. To effectively intercept counterfeit passports, existing verification technologies can be broadly categorized into the following three types:

[0003] I. Documents typically employ various material-level and structural-level anti-counterfeiting measures, such as optically variable ink, OVI patterns, watermarks, security threads, embossing, and microtext. These physical features are primarily detected by visual inspection or specialized equipment such as high-speed document scanners (Chinese patent publication number CN117132750A). While these methods offer some anti-counterfeiting capabilities, their ability to identify highly realistic counterfeit documents is limited, and their automation level is low, relying heavily on human experience and posing a risk of misjudgment. Furthermore, the specialized equipment mentioned is difficult to deploy on a large scale in frontline ports of entry and exit, and is also difficult to deploy and apply in immigration management, border inspection, and entry / exit window units. In practice, it still relies on human experience and judgment, which is difficult for the human eye to discern differences in PDF417 QR codes, resulting in high subjectivity, low efficiency, and a high rate of misjudgment.

[0004] Second, for passports equipped with electronic chips (e-passports), encrypted personal information, photos, and signature data can be read using dedicated equipment to complete consistency verification. Taking Chinese patent CN115147884A as an example, it discloses an electronic passport verification and identification method. By extracting facial photos, names, nationalities, and other feature information from the chip and combining it with a digital signature verification mechanism, it achieves multi-dimensional consistency verification with high accuracy. However, this method relies on chips and PKI infrastructure and cannot be applied to traditional entry and exit documents without electronic chips, resulting in significant compatibility gaps in immigration management, border inspection, and entry and exit management scenarios.

[0005] III. In recent years, some studies have attempted to use artificial intelligence technology to recognize document images. For example, the method described in Chinese patent publication number CN112215225A proposes a document forgery recognition method based on image depth features. It uses a convolutional neural network to extract features such as texture, edge, and color from passport images and then uses a classification model to determine whether forgery exists. This method is suitable for detecting image tampering types such as photo replacement and signature synthesis. However, it is generally ineffective for passports with weak anti-counterfeiting features or simple production processes. Furthermore, its recognition accuracy depends on the quality of the training data and lacks a multimodal fusion model that integrates with barcode data, document structure, or issuance rules, making it unsuitable for complex passport structures that span multiple countries and standards.

[0006] Currently, most developed countries have adopted e-passports with embedded encryption chips, using digital signatures and key authentication mechanisms to ensure the authenticity and integrity of the data. However, some countries and regions have not yet widely adopted e-passports, and their non-electronic entry and exit documents still contain PDF417 QR codes on their data pages. These PDF417 QR codes carry document data for anti-counterfeiting verification. These types of documents pose a significant risk of forgery in immigration and border control processes such as international customs clearance, air transfers, and visa-on-arrival applications. Current technology struggles to accurately identify and authenticate non-electronic entry and exit documents with PDF417 QR codes, and the accuracy of identification needs improvement. Summary of the Invention

[0007] Therefore, it is necessary to provide a method, system, and medium for authenticating entry and exit documents based on QR code recognition, addressing the issue that the accuracy of identifying counterfeit non-electronic entry and exit documents with PDF417 QR codes needs to be improved.

[0008] To solve the above problems, the present disclosure adopts the following technical solution:

[0009] Firstly, this disclosure provides a method for authenticating entry and exit documents based on QR code recognition, including the following steps:

[0010] A cross-modal rule engine is constructed. The rules of the cross-modal rule engine include a cross-verification rule base, which includes the logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents.

[0011] Obtain an image of the data page of the entry / exit document to be inspected;

[0012] The data page image is preprocessed;

[0013] The QR code information is obtained by using a QR code recognition engine to identify the QR code in the preprocessed data page image.

[0014] The identified issuance information is obtained by identifying visual and machine-readable information in the preprocessed data page image;

[0015] Based on the dynamic QR code verification database and the dynamic blacklist database, the uniqueness and legality of the identified QR code information are verified.

[0016] The cross-modal rule engine compares the correspondence between the identified QR code information and the identified issuance information based on the cross-verification rule base to obtain the comparison result;

[0017] Based on at least one of the uniqueness verification result, the legality verification result, and the comparison result, the authenticity of the entry and exit documents is determined, and an authenticity verification report is generated.

[0018] In a preferred embodiment, the method for comparing the correspondence between the identified QR code information and the identified issuance information includes comparing regular expressions, fuzzy comparison, and logical reasoning.

[0019] In a preferred embodiment, the comparison step uses comparison regular expressions, fuzzy comparison, and logical reasoning methods to determine whether the correspondence between the identified QR code information and the identified issuance information is correct.

[0020] For QR code information with a fixed format, regular expressions are used for comparison;

[0021] For identifiers in the structured data corresponding to non-fixed format QR code information and issuance information, a fuzzy comparison method is used for comparison.

[0022] For fields that have objective logical relationships, the logical relationships between the fields are compared using logical reasoning methods.

[0023] In a preferred embodiment, the cross-verification rule base also includes the correspondence between the issuing authority and the country code of the entry and exit documents, and the method for authenticating the entry and exit documents also includes the cross-modal rule engine comparing the correspondence between the issuing authority and the country code in the identified QR code information and / or the identified issuance information according to the cross-verification rule base to see if it is correct.

[0024] In a preferred embodiment, the step of determining the authenticity of the entry / exit document based on at least one of the uniqueness verification result, the legality verification result, and the comparison result, and generating an authenticity verification report, specifically involves: determining the authenticity of the entry / exit document based on the uniqueness verification result, the legality verification result, and the comparison result to obtain a determination result, and generating an authenticity verification report based on the determination result, the uniqueness verification result, the legality verification result, and the comparison result.

[0025] In a preferred embodiment, the step of performing uniqueness and legality verification on the identified QR code information based on the dynamic QR code verification library and the dynamic blacklist library specifically involves: performing uniqueness verification on the identified QR code information based on the dynamic QR code verification library and the identified issuance information; and performing legality verification on the identified QR code information based on the dynamic QR code verification library, the dynamic blacklist library, and the identified issuance information.

[0026] In a preferred embodiment, the dynamic QR code verification database includes country code information and document number information of entry and exit documents; the identified issuance information includes country code and document number;

[0027] The uniqueness verification specifically includes: searching for information in the dynamic QR code verification database that is identical to the identified QR code information; if the information is successfully found, then the country code information and the document number information of the entry and exit document corresponding to the dynamic QR code verification database are matched one by one with the country code and document number in the identified issuance information. If all the matches are correct, the uniqueness verification result is that it is unique; otherwise, the uniqueness verification result is that it is not unique.

[0028] In a preferred embodiment, the preprocessing specifically includes at least one of image enhancement, alignment processing, and distortion correction processing.

[0029] Secondly, this disclosure provides a system for authenticating entry and exit documents based on QR code recognition, including:

[0030] The module is used to build a cross-modal rule engine. The rules of the cross-modal rule engine include a cross-verification rule library, which includes logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents.

[0031] The image acquisition module is used to acquire images of the data pages of the entry / exit documents to be inspected;

[0032] The image processing module is used to preprocess the image of the data page;

[0033] The QR code information recognition module is used to identify the QR code in the preprocessed data page image using a QR code recognition engine to obtain the recognized QR code information.

[0034] The verification module is used to perform uniqueness and legality verification on the identified QR code information based on the dynamic QR code verification library and the dynamic blacklist library.

[0035] The text recognition module is used to recognize the visual and machine-readable information in the preprocessed data page image to obtain the recognized issuance information;

[0036] The comparison module is used to compare the correspondence between the identified QR code information and the identified issuance information based on the cross-modal rule engine and the cross-verification rule library to obtain the comparison result;

[0037] The authenticity verification module is used to determine the authenticity of entry and exit documents based on at least one of the uniqueness verification result, legality verification result, and comparison result, and generate an authenticity verification report.

[0038] Thirdly, this disclosure provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for authenticating entry and exit documents based on QR code recognition as described in the first aspect of the claim.

[0039] The aforementioned method, system, and medium for authenticating entry and exit documents based on QR code recognition identify QR code information and issuance information in the document page image. By constructing a cross-verification rule base, it compares the correspondence between the identified QR code information and the identified issuance information. It also performs uniqueness and legality verification based on a dynamic QR code verification database and a dynamic blacklist database. This three-dimensional approach to document authenticity verification offers high accuracy, enabling the identification of blacklisted QR codes and accurately identifying high-risk forgery scenarios such as different documents with the same code or tampered fields. Furthermore, the generation of an authenticity verification report facilitates rapid access to details during border inspection. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating a method in one embodiment of the present disclosure;

[0041] Figure 2 This is a schematic diagram of the system structure in one embodiment of the present disclosure. Detailed Implementation

[0042] The technical solutions of this disclosure will now be described in detail with reference to the accompanying drawings and preferred embodiments.

[0043] See Figure 1 This embodiment provides a method for authenticating entry and exit documents based on QR code recognition, including:

[0044] A cross-modal rule engine is constructed. The rules of the cross-modal rule engine include a cross-verification rule base, which includes the logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents.

[0045] Obtain an image of the data page of the entry / exit document to be inspected;

[0046] The data page image is preprocessed;

[0047] The QR code information is obtained by using a QR code recognition engine to identify the QR code in the preprocessed data page image.

[0048] The visual and machine-readable information in the preprocessed data page image is identified to obtain the identified issuance information;

[0049] Based on the dynamic QR code verification database and the dynamic blacklist database, as well as the identified issuance information, the uniqueness and legality of the identified QR code information are verified.

[0050] The cross-modal rule engine compares the correspondence between the identified QR code information and the identified issuance information based on the cross-verification rule base to obtain the comparison result;

[0051] Based on at least one of the uniqueness verification result, the legality verification result, and the comparison result, the authenticity of the entry and exit documents is determined, and an authenticity verification report is generated.

[0052] It should be understood that the steps for recognizing QR code information and visual and machine-readable information in the above method may vary depending on the execution process. For example, they may be performed simultaneously, one may be performed first, or they may be performed interspersed. Therefore, the above method does not represent or imply that all steps must be performed in this order. Those skilled in the art can change or modify the execution order of the above steps based on this disclosure. Some embodiments of the above method are illustrated below.

[0053] It is understood that the QR code is a PDF417 QR code.

[0054] Understandably, the cross-verification rule base includes the logical relationship information of entry and exit documents from several countries.

[0055] In this embodiment, an in-depth study of the data pages of entry and exit documents from 42 countries worldwide that contain PDF417 QR code information is conducted. Automated batch recognition of the PDF417 QR codes on the data pages of entry and exit documents is performed, and a cross-verification rule base is constructed using data mining techniques.

[0056] PDF417 QR codes are unique, and in most countries, the parsed QR codes store key information related to the document holder. This information includes, for example, personal identification number, document number, and name. This information is highly correlated with the machine-readable zone (MRZ) and / or visually readable zone (VIZ) information of the document. The machine-readable zone and visually readable zone information are collectively referred to as the issuance information of entry / exit documents. In this field, issuance information includes the country code and document number (the document number refers to the document number of the entry / exit document, not the holder's ID number), but it is understood that it is not limited to only including the country code and document number. In this embodiment, the cross-verification rule base contains 134 logical relationships, which are 134 cross-verification rules for "issuance information + QR code information". Each rule has its own sequence number, which can be understood as an ID. It should be understood that this does not mean that every country is applicable to all 134 logical relationships. Usually, a certain country is applicable to some of the 134 logical relationships. The cross-verification rule base has a correspondence between the logical relationship information and the country information. That is to say, it includes the country information and the logical relationship information corresponding to the country information. The logical relationship information is the information on the logical relationship between the QR code information of the entry and exit document and the issuance information of the entry and exit document.

[0057] In this embodiment, obtaining the image of the entry / exit document information page specifically involves: automatically capturing the image of the entry / exit document information page through a document reader or obtaining the image of the entry / exit document information page manually uploaded by the user.

[0058] In this embodiment, preprocessing the data page image involves improving the accuracy of the identified QR code information and the identified issuance information to obtain a preprocessed data page image. Specifically, the preprocessing includes at least one of the following: image enhancement, alignment processing, and distortion correction, to ensure accurate and reliable extraction of subsequent QR code and visual information.

[0059] In this embodiment, the identification of visual and machine-readable information in the preprocessed data page image to obtain the identified issuance information specifically involves:

[0060] Using OCR text recognition technology, printed text in the preprocessed document page image is identified to obtain the recognized issuance information, including but not limited to key information such as the certificate holder's name, gender, date of birth, certificate number, issuance date, validity period, and issuing country. The OCR text recognition technology possesses capabilities such as image interference removal, field location optimization, and semantic error correction, ensuring the accuracy and structural consistency of the extracted data. The extracted issuance information is then organized into a structured data format, such as storing it in a table, to facilitate field-by-field comparison with the QR code information in subsequent steps.

[0061] In this embodiment, the specific steps of using a QR code recognition engine to recognize the QR code information in the preprocessed data page image are as follows:

[0062] The PDF417 QR code recognition engine is used to automatically locate and recognize the QR code area in the pre-processed document page image, obtaining the recognized QR code information. For some countries, the recognized QR code information typically includes multiple items, such as the document number of the entry / exit document, the holder's ID card number, and the holder's name. Furthermore, it may also include date of birth, machine-readable code, and tracking number information. For some countries, the recognized QR code information is in hexadecimal format, and the PDF417 QR code recognition engine cannot directly recognize the document number, holder's ID card number, and holder's name.

[0063] The process involves using a dynamic QR code verification database and a dynamic blacklist database, along with the identified issuance information, to perform uniqueness and legality checks on the QR code information. By comparing the QR code information against the dynamic QR code verification database and the dynamic blacklist database, uniqueness and legality checks are achieved. This process can identify fraudulent situations such as "tampering with fields while maintaining the QR code structure," effectively detecting QR code forgery.

[0064] In this embodiment, the dynamic QR code verification library is used to verify uniqueness, and the dynamic blacklist library is used to verify legitimacy. In some embodiments, the dynamic QR code verification library is also used to verify legitimacy. In this embodiment, verifying the uniqueness of the QR code information includes: according to the dynamic QR code verification library, verifying whether there is a unique (and only) QR code information in the dynamic QR code verification library that is completely identical to the identified QR code information, that is, every piece of information in the QR code information completely corresponds to it.

[0065] In this embodiment, the dynamic QR code verification database includes QR code information provided by several countries, specifically QR code information provided by national entry and exit document production and / or management agencies. The function of the dynamic QR code verification database is to verify whether the identified QR code information is provided by the country, i.e., for uniqueness verification. In some embodiments, the dynamic QR code verification database is used for uniqueness verification and legality verification. If all the contents of the identified QR code information are completely identical to all the contents of a single QR code information in the dynamic QR code verification database, then the uniqueness verification and legality verification based on the dynamic QR code verification database are passed; otherwise, they are not passed. Here are some examples of situations that fail the verification: For instance, if a recognized QR code only includes name, gender, and date of birth, and the dynamic QR code verification database contains more than one completely matching QR code, then the uniqueness requirement is not met. Similarly, if a particular element of a recognized QR code results in no unique corresponding QR code in the dynamic QR code verification database, then the uniqueness requirement is not met. In this case, the entry / exit document to be verified may be invalid, or it may be directly determined to be invalid.

[0066] In this embodiment, the dynamic blacklist database includes several QR code information entries, including several invalid QR code entries. The function of the dynamic blacklist database is to determine whether the QR code information is valid. The identification of QR code information is validated against the dynamic blacklist database. If the identified QR code information has a matching QR code entry, the validation result is invalid. Here, the dynamic blacklist database includes QR code information corresponding to individuals who are prohibited from entering or leaving the country, and QR code information of problematic QR codes identified in prior authentication.

[0067] In a preferred embodiment, the dynamic QR code verification database includes country code information and document number information of entry / exit documents. In a specific embodiment, the dynamic QR code verification database also includes QR code information; in another specific embodiment, the QR code information includes country code information and / or document number information of entry / exit documents, then the dynamic QR code verification database also includes several other pieces of information in the QR code information.

[0068] The uniqueness verification specifically includes: searching for information in the dynamic QR code verification database that is identical to the identified QR code information; if the information is successfully found, then the country code information and the document number information of the entry and exit document corresponding to the dynamic QR code verification database are matched one by one with the country code and document number in the identified issuance information. If all the matches are correct, the uniqueness verification result is that it is unique; otherwise, the uniqueness verification result is that it is not unique.

[0069] Specifically, for some countries, the identified QR code information is in hexadecimal format. The PDF417 QR code recognition engine cannot directly identify information such as document number, holder's ID number, or holder's name. Therefore, the dynamic QR code verification database for these countries includes hexadecimal information, as well as the country code and entry / exit document number for each piece of hexadecimal information. Similarly, the dynamic blacklist database also includes hexadecimal information, country code information, and entry / exit document number information, with each piece of hexadecimal information having a corresponding country code and entry / exit document number. Obviously, for QR code information in text format (non-hexadecimal format), both the dynamic QR code verification database and the dynamic blacklist database include text-formatted QR code information. Both databases include country code information, document number information for entry / exit documents, and other relevant information, such as the holder's ID card number, name, and machine-readable code. For QR code information in hexadecimal format, the uniqueness check is as follows: the identified QR code information is compared with the dynamic QR code verification database. When information corresponding to the identified QR code information is detected in the dynamic QR code verification database, the corresponding country code information and document number information for entry / exit documents in the dynamic QR code verification database are checked to see if they match the country code and document number of the identified issuance information. If both the country code and document number are correct, the uniqueness check passes; otherwise, the uniqueness check result is not unique. This method can also be used for text-formatted QR code information. In one embodiment, the dynamic blacklist database includes country code information and document number information of entry and exit documents, and the uniqueness verification method is referenced according to the verification of the dynamic blacklist database.

[0070] In this embodiment, the cross-modal rule engine compares the correspondence between the identified QR code information and the identified issuance information according to the cross-verification rule base to obtain the comparison result, specifically:

[0071] Depending on the country, the logical relationship between the data contained in the QR code of the entry and exit documents of the corresponding country (QR code information) and the issuance information of the entry and exit documents is extracted into a rule base. As a cross-verification rule base, it can be understood that this rule base is an extensible rule engine architecture.

[0072] The cross-verification rule base includes the correspondence between the QR code information of entry / exit documents and the issuance information of those documents, i.e., the logical relationship. The logical relationship between the QR code information and the issuance information of entry / exit documents may differ for documents from different countries; therefore, this logical relationship is predicated on the country to which the entry / exit document corresponds. The rules cover information such as name, date of birth, document number, validity period, machine-readable code, and tracking number.

[0073] The cross-verification rule base also includes the correspondence between issuing authorities and country codes. The method for authenticating entry and exit documents based on QR code recognition also includes the cross-modal rule engine comparing the correspondence between issuing authorities and country codes in the recognized QR code information and / or the recognized issuance information to see if it is correct, based on the logical relationship between issuing authorities and country codes in the cross-verification rule base.

[0074] In this comparison step, the methods of comparison regular expression, fuzzy comparison, and logical reasoning are used to determine whether the correspondence between the identified QR code information and the identified issuance information is correct.

[0075] For QR code information with a fixed format, regular expressions are used for comparison: For QR code information with a fixed format, such as machine-readable area information, ID number, date of birth, name, etc., regular expressions are used for rigid comparison and verification to effectively identify problems such as incorrect field order and format tampering;

[0076] For identifiers in structured data corresponding to non-fixed format QR code information and identifiers in structured data corresponding to issuance information, such as encrypted information, image information, fingerprint information, etc., fuzzy comparison method is used for comparison: For non-fixed format issuance information and identifiers in structured data corresponding to QR codes, fuzzy comparison technology is used to handle slight deviations in issuance information based on string similarity, and to accurately detect forgery traces.

[0077] For fields with objective logical relationships, the logical relationship between the fields is compared through logical reasoning methods to determine whether the logical relationship between the fields is correct. For issuance information data items with multiple fields having objective logical relationships, cross-field logical reasoning is used for consistency verification, which can discover complex forgery problems such as factual contradictions. That is, the logical reasoning method is cross-field logical reasoning. Typically, "inter-field" refers to the relationship between fields in the QR code information and fields in the issuance information. For example, a certain field in the QR code information and two fields in the issuance information should have an objective logical relationship. The method involves analyzing whether the actual logical relationship is correct using logical reasoning. However, this is not limited to this. For instance, in some embodiments, the cross-verification rule base includes the logical relationships within the QR code information of the entry / exit document and the logical relationship information within the issuance information of the entry / exit document. The identification method further includes a cross-modal rule engine comparing the logical relationships within the identified QR code information and the logical relationships within the QR code information of the entry / exit document using logical reasoning. The cross-modal rule engine also compares the logical relationships within the identified issuance information and the logical relationships within the issuance information of the entry / exit document using logical reasoning.

[0078] Based on cross-verification rule base comparison, it can more accurately identify high-risk counterfeit situations such as "same code but different certificate" and "tampering with fields but maintaining QR code structure", and more effectively and accurately detect QR code copying, cloning, forgery and other behaviors.

[0079] In this embodiment, the basis for analyzing and judging the authenticity of entry and exit documents is to generate and output an authenticity identification report based on at least one of the uniqueness verification result, the legality verification result, and the comparison result.

[0080] For example, in one embodiment, if the legality verification result is invalid, then the uniqueness verification and the comparison are not performed / stopped, and the authenticity authentication report only highlights that the legality verification result is invalid. In another embodiment, the comparison is performed regardless of the verification result, and uniqueness verification and legality verification are performed regardless of the comparison result, resulting in a comprehensive and detailed authentication report. In yet another embodiment, a determination is made based on the uniqueness result and the legality verification result to determine whether to perform the comparison. It is understood that the above are merely examples and are not exhaustive.

[0081] In this embodiment, the authenticity verification report includes uniqueness verification results, legality verification results, comparison results, and authenticity verification conclusions.

[0082] In one specific embodiment, the authenticity verification report includes the validity of the QR code structure (determined based on legality or based on legality and uniqueness), uniqueness status, field consistency comparison results, whether there is information suspected of forgery, suspicious fields, rule hit rate, and comparison matching failures (e.g., showing the rule number corresponding to the comparison matching failure).

[0083] The authenticity verification report provides clear and effective support for immigration management, border inspection, and entry-exit management processes.

[0084] In a preferred embodiment, the method for authenticating entry and exit documents is applied to the server side.

[0085] Here, the implementation process of the method for authenticating entry and exit documents based on QR code recognition is illustrated, including:

[0086] Step 1: Construct a cross-modal rule engine. The rules of the cross-modal rule engine include a cross-verification rule base, which includes the logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents. The cross-verification rule base includes the logical relationship information of 42 countries, with a total of 134 unique rules.

[0087] Step 2: Obtain an image of the data page of the entry / exit document to be tested;

[0088] Step 3: Automatically preprocess the data page image to obtain the data page image of the entry / exit document;

[0089] Step 4: Use the QR code recognition engine to automatically recognize the QR code in the preprocessed data page image to obtain the recognized QR code information; automatically recognize the issuance information in the preprocessed data page image to obtain the recognized issuance information;

[0090] Step 5: Based on the dynamic QR code verification library and the dynamic blacklist library, as well as the identified issuance information, the identified QR code information is automatically verified. The verification includes uniqueness verification and legality verification, and uniqueness verification results and legality verification results are obtained. The cross-modal rule engine compares the correspondence between the identified QR code information and the identified issuance information according to the cross-verification rule library to obtain the comparison result.

[0091] Step 6: Based on the uniqueness verification result, the legality verification result, and the comparison result, determine the authenticity of the entry and exit documents and obtain the judgment result; based on the judgment result, the uniqueness verification result, the legality verification result, and the comparison result, generate an authenticity verification report.

[0092] See Figure 2 This disclosure provides a system for authenticating entry and exit documents based on QR code recognition, including:

[0093] The module is used to build a cross-modal rule engine. The rules of the cross-modal rule engine include a cross-verification rule library, which includes logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents.

[0094] The image acquisition module is used to acquire images of the data pages of the entry / exit documents to be inspected;

[0095] The image processing module is used to preprocess the image of the data page;

[0096] The QR code information recognition module is used to identify the QR code in the preprocessed data page image using a QR code recognition engine to obtain the recognized QR code information.

[0097] The text recognition module is used to recognize the issuance information in the preprocessed data page image and obtain the recognized issuance information;

[0098] The verification module is used to perform uniqueness and legality verification on the identified QR code information based on the dynamic QR code verification library and the dynamic blacklist library, as well as the identified issuance information.

[0099] The comparison module is used to compare the correspondence between the identified QR code information and the identified issuance information based on the cross-modal rule engine and the cross-verification rule library to obtain the comparison result;

[0100] The authenticity verification module is used to determine the authenticity of entry and exit documents based on at least one of the uniqueness verification result, legality verification result, and comparison result, and generate an authenticity verification report.

[0101] In specific implementation, the QR code recognition-based entry and exit document authenticity verification system can refer to the QR code recognition-based entry and exit document authenticity verification method in any of the above embodiments to achieve authenticity verification. The specific implementation steps will not be repeated here.

[0102] In this embodiment, the cross-verification rule base also includes the correspondence between the issuing authority and the country code of the entry and exit documents. The comparison module is also used to use the cross-modal rule engine to compare the correspondence between the issuing authority and the country code in the identified QR code information and / or the identified issuance information according to the cross-verification rule base to see if it is correct.

[0103] In this embodiment, the image acquisition module is specifically used to obtain the data page image of the entry / exit document to be inspected automatically collected by the document reader or to obtain the data page image of the entry / exit document to be inspected manually uploaded by the user; the text recognition module is used to recognize the issuance information in the preprocessed data page image using OCR text recognition technology.

[0104] This disclosure also provides a storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the method for authenticating entry and exit documents based on QR code recognition.

[0105] An electronic device can be implemented according to the method of this disclosure. The electronic device includes: a memory; one or more processors; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing the method for authenticating entry and exit documents based on QR code recognition according to any of the above embodiments.

[0106] This disclosure also provides a cloud server for implementing the steps of the method for authenticating entry and exit documents based on QR code recognition.

[0107] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0108] This disclosure discloses a method, system, and medium for authenticating entry and exit documents based on QR code recognition. It identifies QR code information and issuance information in the document page image, performs uniqueness and legality verification through a dynamic QR code verification library and a dynamic blacklist library, and verifies the correspondence between the identified QR code information and the identified issuance information by constructing a cross-verification rule library. The authenticity of the entry and exit document is determined based on at least one of the uniqueness verification result, legality verification result, and comparison result, generating an authenticity verification report for quick access to details at the border inspection stage. This design, through three dimensions—uniqueness verification, legality verification, and verification of the correspondence between the identified QR code information and the identified issuance information—identifies the authenticity of paper entry and exit documents with PDF417 format QR codes. The accuracy is high, capable of identifying blacklisted QR codes, accurately identifying high-risk counterfeit situations such as different documents with the same code, and tampered fields, and effectively detecting QR code copying, cloning, and forgery.

[0109] Specifically, this disclosure systematically analyzes the encoded information carried by the PDF417 format QR codes in paper entry and exit documents and performs automated cross-verification with the associated issuance information to effectively identify documents with forged QR code information. The structured parsing and field-level comparison mechanism can detect potential forgery, field misalignment, or encoding tampering in the QR codes.

[0110] In reality, criminals may clone or tamper with QR code information to implant the same data into different documents, resulting in the phenomenon of "same code, different documents" where the same identity information appears on multiple documents held by different people. Existing verification methods are unable to identify counterfeit documents with "same code, different documents". However, this disclosure verifies the authenticity of documents through two aspects: uniqueness verification and comparison. In particular, this disclosure designs a mapping comparison between QR code information and machine-readable zone (MRZ) / visually readable zone (VIZ) information, which can accurately identify and confirm counterfeit documents.

[0111] Specifically, this disclosure can automatically achieve identification by deeply integrating image recognition, QR code structure parsing, rule engine comparison, uniqueness verification, and comprehensive analysis to form a full-process, automated, highly accurate, and efficient document verification solution. It significantly fills the technological gap in existing methods for verifying documents without chips and solves the problem that QR code information without encryption protection is difficult to verify independently.

[0112] Specifically, this disclosure is based on a QR code and historical data verification mechanism, combined with the comparison of the correspondence between QR code information and issuance information according to the cross-verification rule base. The verification and comparison can be carried out in parallel, realizing the rapid identification of typical forgery methods such as QR code copying and image replacement to retain the code. It significantly enhances the system's ability to prevent duplicate documents and cloning behavior, and can provide effective early warning even in offline environments.

[0113] Specifically, this disclosure improves its applicability and lifecycle by adjusting the cross-verification rule base, using a dynamic QR code verification base, and a dynamic blacklist base, thus meeting the ever-changing needs of international entry and exit management and giving it good generalization and adaptability.

[0114] Specifically, based on this disclosure, a service center + terminal application deployment method can be adopted. Specifically, the cloud server performs the authentication, and the staff terminal obtains the authenticity authentication report. This implementation can reduce deployment costs and is easy to promote and apply.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] The embodiments described above are merely illustrative of several implementations of this disclosure, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the disclosure. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this disclosure, and these all fall within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the appended claims.

Claims

1. A method for authenticating entry and exit documents based on QR code recognition, characterized in that, Includes the following steps: A cross-modal rule engine is constructed. The rules of the cross-modal rule engine include a cross-verification rule base, which includes the logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents. Obtain an image of the data page of the entry / exit document to be inspected; The data page image is preprocessed; The QR code information is obtained by using a QR code recognition engine to identify the QR code in the preprocessed data page image. The identified issuance information is obtained by identifying visual and machine-readable information in the preprocessed data page image; Based on the dynamic QR code verification database and the dynamic blacklist database, as well as the identified issuance information, the uniqueness and legality of the identified QR code information are verified. The cross-modal rule engine compares the correspondence between the identified QR code information and the identified issuance information based on the cross-verification rule base to obtain the comparison result; Based on at least one of the uniqueness verification result, the legality verification result, and the comparison result, the authenticity of the entry and exit documents is determined, and an authenticity verification report is generated.

2. The method for authenticating entry and exit documents based on QR code recognition according to claim 1, characterized in that, The method for comparing the correspondence between the identified QR code information and the identified issuance information includes comparing regular expressions, fuzzy comparison, and logical reasoning.

3. The method for authenticating entry and exit documents based on QR code recognition according to claim 2, characterized in that, In this comparison step, the methods of comparison regular expression, fuzzy comparison, and logical reasoning are used to determine whether the correspondence between the identified QR code information and the identified issuance information is correct. For QR code information with a fixed format, regular expressions are used for comparison; For identifiers in the structured data corresponding to non-fixed format QR code information and issuance information, a fuzzy comparison method is used for comparison. For fields that have objective logical relationships, the logical relationships between the fields are compared using logical reasoning methods.

4. The method for authenticating entry and exit documents based on QR code recognition according to claim 1, characterized in that, The cross-verification rule base also includes the correspondence between the issuing authority and the country code of the entry and exit documents. The method for authenticating the entry and exit documents also includes the cross-modal rule engine comparing the correspondence between the issuing authority and the country code in the identified QR code information and / or the identified issuance information according to the cross-verification rule base to see if it is correct.

5. The method for authenticating entry and exit documents based on QR code recognition according to claim 1, characterized in that, The step of determining the authenticity of an entry / exit document and generating an authenticity verification report based on at least one of the uniqueness verification result, the legality verification result, and the comparison result specifically involves: determining the authenticity of the entry / exit document based on the uniqueness verification result, the legality verification result, and the comparison result to obtain a determination result; and generating an authenticity verification report based on the determination result, the uniqueness verification result, the legality verification result, and the comparison result.

6. The method for authenticating entry and exit documents based on QR code recognition according to claim 1, characterized in that, The step of performing uniqueness and legality verification on the identified QR code information based on the dynamic QR code verification library and the dynamic blacklist library specifically involves: performing uniqueness verification on the identified QR code information based on the dynamic QR code verification library and the identified issuance information; and performing legality verification on the identified QR code information based on the dynamic QR code verification library, the dynamic blacklist library, and the identified issuance information.

7. The method for authenticating entry and exit documents based on QR code recognition according to claim 1 or 6, characterized in that, The dynamic QR code verification database includes country code information and document number information of entry and exit documents; the identified issuance information includes country code and document number; The uniqueness verification specifically includes: searching for information in the dynamic QR code verification database that is identical to the identified QR code information; if the information is successfully found, then the country code information and the document number information of the entry and exit document corresponding to the dynamic QR code verification database are matched one by one with the country code and document number in the identified issuance information. If all the matches are correct, the uniqueness verification result is that it is unique; otherwise, the uniqueness verification result is that it is not unique.

8. The method for authenticating entry and exit documents based on QR code recognition according to claim 1, characterized in that, The preprocessing specifically includes at least one of image enhancement, alignment processing, and distortion correction processing.

9. A system for authenticating entry and exit documents based on QR code recognition, characterized in that, include: The module is used to build a cross-modal rule engine. The rules of the cross-modal rule engine include a cross-verification rule library, which includes logical relationship information between the QR code information of entry and exit documents and the issuance information of entry and exit documents. The image acquisition module is used to acquire images of the data pages of the entry / exit documents to be inspected; The image processing module is used to preprocess the image of the data page; The QR code information recognition module is used to identify the QR code in the preprocessed data page image using a QR code recognition engine to obtain the recognized QR code information. The text recognition module is used to identify visual and machine-readable information in the preprocessed document page image to obtain the identified issuance information; The verification module is used to perform uniqueness and legality verification on the identified QR code information based on the dynamic QR code verification library and the dynamic blacklist library, as well as the identified issuance information. The comparison module is used to compare the correspondence between the identified QR code information and the identified issuance information based on the cross-modal rule engine and the cross-verification rule library to obtain the comparison result; The authenticity verification module is used to determine the authenticity of entry and exit documents based on at least one of the uniqueness verification result, legality verification result, and comparison result, and generate an authenticity verification report.

10. A storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for authenticating entry and exit documents based on QR code recognition as described in any one of claims 1-8.

Images