A method, apparatus, equipment and medium for content review of airport terminal display screens.

Abstract

This invention discloses a method, apparatus, device, and medium for content review of airport terminal displays, specifically relating to the field of content review technology. Its key technical points are: calculating the hash values ​​of each of the processed multiple content segments; verifying the consistency of the content to be displayed using a Byzantine fault tolerance mechanism based on the hash values ​​of each of the multiple content segments; after the consistency verification is passed, calculating the security probability of the multiple content segments using a security probability calculation function to obtain the content security probability of the content to be displayed, and outputting a broadcast control command after the content security probability meets preset conditions; obtaining the hash value of the broadcast control command, and determining the integrity of the broadcast control command using the hash values ​​of similar commands in a historical database; after passing the determination, signing and encrypting the broadcast control command, and uploading the encrypted command to the blockchain for notarization; if notarization is successful, sending the broadcast control command to the advertising screen execution terminal for playback according to the broadcast control command.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of content review, in particular to an airport terminal display screen content review method, device, equipment and medium. BACKGROUND

[0002] The current airport advertising screen management presents a diversified trend, and different types of advertising screens have their own management problems: third-party managed advertising screens (such as Focus Media) are remotely published by off-site devices with one key, the publishing process is independent of real-time supervision of the airport, the content transmission link has the risk of being hijacked, and the responsibility subject traceability chain is long; although the self-managed advertising screens of the merchants use the airport network, the management authority is independent, the content publishing is disconnected from the airport review system, and it is difficult to identify the responsibility after the illegal content appears; among the self-owned electronic advertising screens of the airport, the single machine screen relies on manual content update, and the networked screen is connected to the airport network but lacks a security review means.

[0003] The prior art has many deficiencies, which cannot cover the differentiated management needs of various types of airport advertising electronic screens, resulting in that the airport cannot uniformly implement safety management, the advertising screen content published by each subject cannot be strictly reviewed, the real-time safety of the advertising screen content cannot be detected and evaluated, there is a lack of pre-audit mechanism for the content published by each subject, the audit response is lagging, the traceability data formats of various screens are not unified, it is difficult to form a standardized responsibility mechanism system, and there is a lack of quantitative model for security content review, resulting in frequent advertising screen content security incidents.

[0004] Therefore, the present application aims to provide an airport terminal display screen content review method, device, equipment and medium to solve the above-mentioned related problems. SUMMARY

[0005] The technical problem to be solved by the present application is that the display content published by each content sending subject lacks a pre-audit mechanism, and the real-time safety of the display content cannot be detected and evaluated, and the purpose is to provide an airport terminal display screen content review method, device, equipment and medium, which verifies the consistency of the display content by using the Byzantine fault tolerance mechanism, determines the content security probability based on the bit similarity score and the real-time credit score, and determines the integrity of the broadcast control instruction, constructs a multiple security content review architecture, realizes multiple review determination of the display content, and completes the pre-examination of the display content; through the credit dynamic weight acquisition of each modality vector of the current playing content, and combining the playing safety value calculation of the current playing content, the safety monitoring of the real-time playing content is realized; by combining the determination results of multiple verification determinations, the real-time credit score is dynamically updated and adjusted, so as to dynamically adjust the content security probability of the next display content and the playing safety value of the next playing content.

[0006] The present application is implemented by the following technical solutions:

[0007] The application discloses an airport terminal display screen content auditing method, and relates to the field of airport terminal display screen content auditing.

[0008] The method comprises the following steps: carrying out fragmentation processing on the to-be-displayed content, and calculating the fragmentation hash values of the processed multiple pieces of fragmented content; verifying the consistency of the to-be-displayed content by using a Byzantine fault tolerance mechanism based on the fragmentation hash values of the multiple pieces of fragmented content.

[0009] After the consistency verification is passed, the safety probability of the multiple pieces of fragmented content is calculated by using a safety probability calculation function in combination with the real-time credit score of the sending subject of the to-be-displayed content, the content safety probability of the to-be-displayed content is obtained, and a broadcast control instruction is outputted after the content safety probability meets a preset condition.

[0010] The hash value of the broadcast control instruction is obtained, the integrity of the broadcast control instruction is judged by using the homologous instruction hash values in a historical database, the broadcast control instruction is signed and encrypted after the judgment is passed, the encrypted instruction is uploaded to a block chain for notarization after the encryption, and the broadcast control instruction is issued to an advertising screen execution terminal for playing according to the broadcast control instruction if the notarization is successful.

[0011] Further, the consistency of the to-be-displayed content is verified by using a Byzantine fault tolerance mechanism based on the fragmentation hash values of the multiple pieces of fragmented content, and the specific steps are as follows.

[0012] The reference content hash values stored in a reference database are obtained, and whether the fragmentation hash values of the multiple pieces of fragmented content are equal to the reference content hash values is calculated respectively.

[0013] When the number of the pieces of fragmented content with equal hash values and the total number of the pieces of fragmented content reach a preset ratio, the consistency verification of the to-be-displayed content is passed.

[0014] Further, the safety probability of the multiple pieces of fragmented content is calculated by using a safety probability calculation function, and the content safety probability of the to-be-displayed content is obtained, and the specific steps are as follows.

[0015] The Hamming distance between the fragmentation hash value of each piece of fragmented content and the reference content hash value is calculated, and the content similarity of each piece of fragmented content is obtained.

[0016] The mean and the standard deviation of each piece of fragmented content and the reference content are obtained, and the image structure similarity of each piece of fragmented content is calculated.

[0017] The bit similarity score of each piece of fragmented content is obtained by using the image structure similarity and the content similarity for weighted calculation.

[0018] The real-time credit score of the sending subject of the to-be-displayed content is obtained, and the content safety probability of the to-be-displayed content is calculated by using the real-time credit score in combination with the bit similarity scores of the multiple pieces of fragmented content.

[0019] Further, the hash value of the broadcast control instruction is obtained, and the integrity of the broadcast control instruction is determined by using the hash value of the homologous instruction in the historical database, specifically:

[0020] The hash value of the broadcast control instruction and the hash value of the homologous instruction in the historical database are obtained, and the difference between the hash value of the broadcast control instruction and the hash value of the homologous instruction is calculated.

[0021] When the difference between the two hash values is less than a preset determination threshold, the integrity determination of the broadcast control instruction is passed.

[0022] Further, after the broadcast control instruction is issued to the advertising screen execution terminal for playing according to the broadcast control instruction, the method further comprises:

[0023] Obtain a plurality of first modal feature vectors of the current playing content and a plurality of second modal feature vectors of the reference content in the reference database;

[0024] Using the plurality of first modal feature vectors and the real-time credit score of the sender of the current playing content, a plurality of credit dynamic weights of the plurality of first modal feature vectors are calculated;

[0025] Using the plurality of first modal feature vectors, the plurality of second modal feature vectors, and the plurality of credit dynamic weights, a playing safety value of the current playing content is calculated.

[0026] When the playing safety value is less than a preset safety threshold, a pre-constructed safety plan is called to process the playing content.

[0027] Further, the method further comprises: when at least one of the consistency verification determination result, the content safety probability determination result, the integrity determination result, and the playing safety value determination result is obtained, the real-time credit score is updated to obtain an updated real-time credit score; the updated real-time credit score is used to dynamically adjust the content safety probability of the next to-be-displayed content and the playing safety value of the next playing content.

[0028] The application also provides an airport terminal display screen content auditing device, which is used in the airport terminal display screen content auditing method.

[0029] The content consistency verification module is used for fragmenting the to-be-displayed content and calculating the hash value of each fragmented content; and the consistency of the to-be-displayed content is verified by using the Byzantine fault tolerance mechanism based on the hash value of each fragmented content.

[0030] A content security probability determination module is configured to calculate the security probability of the plurality of fragmented contents by using a security probability calculation function in combination with the real-time credit score of the sending subject of the to-be-displayed content after the consistency verification passes, to obtain the content security probability of the to-be-displayed content, and to output a broadcast control instruction when the content security probability meets a preset condition.

[0031] An instruction integrity determination module is configured to obtain the hash value of the broadcast control instruction, determine the integrity of the broadcast control instruction by using the hash values of the homologous instructions in the historical database, sign and encrypt the broadcast control instruction after the determination passes, upload the encrypted instruction to the block chain for notarization after the encryption, and issue the broadcast control instruction to the advertising screen execution terminal for playing according to the broadcast control instruction if the notarization is successful.

[0032] The application further provides a computer device comprising a system memory and a processor, wherein the system memory stores a computer program, and the processor implements the steps of the method according to any one of the preceding embodiments when executing the computer program.

[0033] The application further provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method according to any one of the preceding embodiments when executed by a processor.

[0034] The application further provides a computer program product comprising instructions, which, when executed by a computer device cluster, cause the computer device cluster to perform the method according to any one of the preceding embodiments.

[0035] Compared with the prior art, the application has the following advantages and beneficial effects:

[0036] In the application, the consistency verification of the to-be-displayed content by using the Byzantine fault tolerance mechanism, the content security probability determination based on the bit similarity score and the real-time credit score, and the integrity determination of the broadcast control instruction are combined to build a multi-security content auditing architecture, to realize the multi-auditing determination of the to-be-displayed content and complete the pre-examination of the to-be-displayed content. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be considered as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor. In the drawings:

[0038] Figure 1 The method flowchart of the airport terminal display screen content auditing method in the embodiment;

[0039] Figure 2 The module connection diagram of the airport terminal display screen content auditing device in the embodiment;

[0040] Figure 3 The structural diagram of the computer device in the embodiment. DETAILED DESCRIPTION

[0041] The exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to help understanding, and should be considered as merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present application. Also, in order to be clear and concise, the description below omits the description of well-known functions and structures.

[0042] In the present application, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, time sequence relationship or importance relationship of these elements. Such terms are only used to distinguish one element from another. In some examples, the first element and the second element can refer to the same instance of the element, and in some cases, based on the context of the description, they can also refer to different instances.

[0043] The terms used in the description of various examples in the present application are only for the purpose of describing the specific examples, and are not intended to be limiting. Unless the number of elements is specifically limited, the element can be one or more. In addition, the term "and / or" used in the present application covers any one of the listed items and all possible combinations.

[0044] As described in the background, the prior art mainly has the following technical problems:

[0045] Content Compliance and Tampering Risks: Airport third-party management agencies and merchants have the authority to publish advertisements at will, and the compliance of the content information needs further consideration. At the same time, their information publishing systems have security vulnerabilities. Hackers may invade the system through network attacks, tamper with the advertising content, and publish false, malicious, or misleading information, such as political, pornographic, or violent content, causing adverse social impact.

[0046] Insecure transmission links: During the transmission of content from the information publishing system to the advertising screen, the data is not protected effectively. Hackers may intrude through the transmission links, directly tamper with the displayed content, or capture sensitive information to attack the system or advertising screen.

[0047] Screen terminals are being hacked: Some advertising screens are installed and deployed in open areas, and the hardware interfaces of the devices, such as USB ports and HDMI interfaces, may be exposed. Criminals can access these interfaces on-site to tamper with the content being played.

[0048] Real-time risks and credit blind spots: Traditional cloud-based review suffers from delays, cannot cope with dynamic content tampering, and lacks real-time security monitoring of content displayed on airport advertising screens. It also lacks quantitative assessment of the historical behavior of merchants and advertisers, making it difficult to dynamically adjust the intensity of the review.

[0049] To address the aforementioned technical problems, this invention provides a method, apparatus, device, and medium for content review on airport terminal displays. Specific technical solutions are detailed in the following embodiments: Example

[0050] See Figure 1 , Figure 1 A flowchart illustrating a method for content review on an airport terminal display screen is provided, wherein the method includes:

[0051] S1: The content to be displayed is segmented, and the segment hash value of each segment is calculated. Based on the segment hash value of each segment, the consistency of the content to be displayed is verified using the Byzantine fault tolerance mechanism.

[0052] Specifically, in this embodiment, the content to be displayed (including images, audio, and text) is converted through an adapter port and then fragmented to obtain N fragments, where N is the number of edge nodes. The N fragments are then sent one-to-one to the N edge nodes. Each edge node uses the SHA-256 algorithm to calculate the hash value of the fragmented content, obtaining the fragment hash value for each fragment. Then, the baseline content stored in the baseline database is retrieved, and its baseline hash value is calculated. The fragment hash values ​​of each of the N fragments are then checked against the baseline hash value. If the number of fragments with equal hash values ​​reaches a preset ratio to the total number of fragments, the consistency verification of the content to be displayed passes. Otherwise, the consistency verification fails, triggering the airport security center to perform node isolation and record traceability information, including the terminal ID, timestamp, and content hash.

[0053] It should be noted that in this embodiment, the baseline database is stored on the cloud node. When the edge nodes perform consistency verification, the cloud node sends the baseline content hash value to each edge node. The baseline database is a standard content library built based on the airport advertising database. The airport advertising database includes general airport materials, approved advertising materials, and the national advertising material filing database. The baseline database contains baseline content of text, audio, and images corresponding to the content to be displayed. In this embodiment, the preset ratio is 85%. In other embodiments, other ratio settings can be used, and no further restrictions are imposed here. The consistency of the content to be displayed is verified using the Byzantine fault tolerance mechanism. The specific calculation function is as follows: ,in, This indicates the total number of edge nodes participating in content fragment verification; Indicates the first Each fragment's content hash value; Indicates the base content hash value; Indicates an indicator function, if Returns 1 if the condition is met, otherwise returns 0.

[0054] If the consistency percentage is greater than 85%, the consistency verification is passed, a temporary token is issued, and the data is transmitted to the cloud node. A judgment result X1 is generated and fed back to the credit federated learning engine for dynamic credit evaluation. If the consistency percentage is less than or equal to 85%, the airport security center is triggered to isolate the execution node and record the traceability information.

[0055] For example, in this embodiment, taking a Class B merchant advertising screen as an example, assume that the merchant "Airport Catering Store A" (current credit score C=82) uploads the advertising video "promo_2025.mp4" (15 seconds long, 1080p resolution); firstly, the video and audio keyframes are automatically extracted (1 frame per second, 15 frames in total), the audio is separated into independent audio tracks (sampling rate 44.1kHz), and the hash value of the base content is calculated: The current number of available edge nodes is 9. The video file is divided into 9 equal segments, each segment being 1.67 seconds long. Assuming that the hashes of segments 1-8 all match (H... edge =0x89b2...), Node 9 returned an abnormal value (H) due to network latency. edge =0x3f7a...), the actual calculation result is: 8 nodes hash match, 1 node is abnormal (e.g., the hash value of shard 3 is 0x3f7a..., which does not match the cloud baseline value 0x9b2e...), the consistency ratio is 8 / 9=88.9%, and a temporary token is generated: Token=AES("B-289b2 / / 20250714T1430", Key=0x5d9c...).

[0056] S2: After the consistency verification is passed, the security probability of multiple content segments is calculated using the security probability calculation function, based on the real-time credit score of the sending entity of the content to be displayed, to obtain the content security probability of the content to be displayed. Then, the broadcast control command is output after the content security probability meets the preset conditions.

[0057] Specifically, in this embodiment, after consensus is achieved, the cloud node first calculates the Hamming distance between the hash value of each fragment content and the hash value of the base content to obtain the content similarity of each fragment content; secondly, it obtains the mean and standard deviation of each fragment content and the base content to calculate the image structure similarity of each fragment content; then, it uses the image structure similarity and content similarity to perform a weighted calculation to obtain the bit similarity score of each fragment content; finally, it obtains the real-time credit score of the sending entity of the content to be displayed, and uses the real-time credit score combined with the bit similarity scores of multiple fragment contents to calculate the content security probability of the content to be displayed.

[0058] It should be noted that, in this embodiment, the Hamming distance calculation formula is: The 'popcount()' function is used to count the number of '1's in a binary number. For example, for a 256-bit hash value, the result of popcount(X) is the total number of '1' bits in the hash value X. The '⊕' symbol represents the bitwise XOR operation. After performing a bitwise XOR operation on two binary numbers of equal length, only the corresponding bits that are different will be set to 1, and the same bits will be set to 0. Indicates the first The Hamming distance between each segment and the baseline content; the formula for calculating image structural similarity is: ,in, They represent the first The average values ​​of each segment's content and the baseline content; Indicates the first The standard deviation of each fragment and the baseline content; This represents the dynamic range of pixel values; the bit similarity score is calculated as follows: ,in, All represent weights. , ; Indicates the first Bit similarity score of each fragment's content; the specific security probability calculation function is as follows: ,in, Indicates the content security probability of the content to be displayed (value ranges from 0 to 1); This represents the edge computing weight coefficient, in this embodiment ; This represents the credit score adjustment factor, which is dynamically configured based on the type of airport advertising screen. In this embodiment... ; This represents the real-time credit score of the sender of the content to be displayed, calculated by the credit federated learning engine, with a value range of 0-100.

[0059] If the content security probability is ≥0.9, the content security probability judgment is passed, triggering the airport security broadcast control gateway to generate broadcast control instructions; and generating judgment result X2, which is fed back to the credit federated learning engine for dynamic credit evaluation.

[0060] If the content security probability is less than 0.9, the airport security center will be triggered to intercept the content and record the source information.

[0061] For example, in this embodiment, the merchant's current credit score C=82 (no historical violations), the average bit similarity is 0.94 (1.0 for 8 nodes, 0.6 for 1 node), the weighting coefficient α=0.6, the adjustment factor k=0.1, and the security probability is calculated. The judgment result X2: P=0.948>0.9, passed the review.

[0062] S3: Obtain the hash value of the broadcast control instruction, and use the hash values ​​of the same source instructions in the historical database to determine the integrity of the broadcast control instruction; after the determination is passed, sign and encrypt the broadcast control instruction, and upload the encrypted instruction to the blockchain for notarization; if notarization is successful, send the broadcast control instruction to the advertising screen execution terminal to play according to the broadcast control instruction.

[0063] Specifically, in this embodiment, the hash value of the broadcast control command calculated using the SHA-256 algorithm and the hash values ​​of similar commands in the historical database are first obtained, and the difference between the hash value of the broadcast control command and the hash value of the similar commands is calculated. When the difference between two hash values ​​is less than a preset threshold If the broadcast control instruction integrity check is successful, the instruction is deemed to have been tampered with; if the check fails, the instruction is considered to have been altered, and an instruction correction process is initiated, including format verification and signature regeneration. Simultaneously, after the broadcast control instruction integrity check is completed, a judgment result X3 is generated and fed back to the credit federated learning engine for dynamic credit evaluation. Then, after the check passes, the broadcast control instruction is signed and encrypted, and the encrypted instruction is uploaded to a blockchain based on Hyperledger Fabric for notarization. If notarization is successful, the broadcast control instruction is sent to the advertising screen execution terminal for playback according to the instruction. If notarization fails, an instruction rollback is triggered, returning to the airport security broadcast control gateway for correction and notarization re-execution.

[0064] It should be noted that, in this embodiment, a same-origin instruction refers to an instruction that shares the same origin as the broadcast control instruction in terms of the issuing entity, content identifier, spatiotemporal sequence, transmission path, and operation chain; in this embodiment, when signing and encrypting the broadcast control instruction, the airport operator uses the RSA-PSS algorithm, and the advertiser uses the SM2 algorithm; and the blockchain evidence storage module based on Hyperledger Fabric is a conventional technical means in this field, and will not be elaborated on here.

[0065] For example, in this embodiment, the content of the broadcast control instruction is {"play": "promo_2025.mp4", "duration": 15, "hash": "0x89b2..."}, and the hash value of the broadcast control instruction is Hcmd=SHA-256(instruction)=0x3a7d...; then, the hash value H_history = 0x5e1a... of a similar instruction (e.g., a similar broadcast control instruction {"play": "promo_2024.mp4", "duration": 15, "hash": "0x5e1a..."} recently successfully executed by the same merchant "Airport Catering A Store") is retrieved from the historical database. Next, the difference between the two hash values ​​is calculated, assuming...

[0066] Preset judgment threshold The calculated difference is much smaller than Therefore, the integrity check of the broadcast control instruction passed, and the result X3 is passed. The broadcast control instruction is then signed and encrypted; the merchant SM2 signature is "Sign".SM2 =0x8e2f...; Airport RSA-PSS signature is Sign RSA-PSS =0x4b6a...; Uploaded to a Hyperledger Fabric-based blockchain for evidence storage. The Hyperledger Fabric transaction hash is 0x3a7d..., and the block height is #7821.

[0067] As one possible implementation, after the broadcast control command is sent to the advertising screen execution terminal and the video is played according to the broadcast control command, the method further includes:

[0068] Obtain multiple first-modal feature vectors of the currently playing content and multiple second-modal feature vectors of the benchmark content in the benchmark database;

[0069] It should be noted that, in this embodiment, the multiple first modal feature vectors include the text modal feature vector, image modal feature vector, and audio modal feature vector of the currently playing content; the multiple second modal feature vectors include the text modal feature vector, image modal feature vector, and audio modal feature vector of the reference content; the modal feature vector extraction method adopts conventional technical means in this field, and will not be described in detail here.

[0070] By using multiple first-modal feature vectors and the real-time credit score of the sender of the currently playing content, the credit dynamic weights of multiple first-modal feature vectors are calculated.

[0071] It should be noted that, in this embodiment, the formula for calculating the dynamic credit weight is as follows: ,in This represents the credit-weighted confidence level of the m-th modality. ; These represent the text modal feature vector, image modal feature vector, and audio modal feature vector of the currently playing content, respectively. This represents the base confidence score for the k-th modality; This represents the credit dynamic weights of the first modality feature vector. .

[0072] By combining multiple first-modal feature vectors, multiple second-modal feature vectors, and multiple credit dynamic weights, the playback safety value of the currently playing content is calculated.

[0073] It should be noted that, in this embodiment, the function for calculating the playback safety value is: ,in, This indicates the playback safety value for the currently playing content; This represents the first modal feature vector of the currently playing content; The second modal feature vector representing the baseline content; The standard deviation of the first modality eigenvector is used to normalize feature differences and is derived from historical data statistics.

[0074] When the playback safety value is lower than the preset safety threshold, a pre-built safety plan is invoked to process the playback content.

[0075] It should be noted that in this embodiment, the preset security threshold is set to 80. Other threshold settings can also be used in other embodiments, and no further restrictions are imposed here. When the playback security value is less than 80, the pre-built security plan is invoked to process the playback content. The processing methods include blocking illegal content and temporarily locking the terminal. When the playback security value is not less than 80, data collection and monitoring continue. After the playback security value is determined, the determination result X4 is output and the determination result X4 is fed back to the credit federated learning engine for dynamic credit evaluation.

[0076] It should also be noted that this embodiment includes a three-level emergency response mechanism, the triggering conditions of which are related to the judgment result:

[0077] Level 1 Response (Immediately Stop Playback): Triggered when the consistency verification result X1 ≤ 85%, the content security probability result X2 < 0.9, or the playback security value result X4 < 70.

[0078] Level 2 response (downgraded to static image): triggered when the playback safety value judgment result X4 is in the range [70, 80).

[0079] Level 3 Response (Switch Security Cache): Triggered when the broadcast control instruction fails to be stored (judgment result X3 is failure) or when an abnormality is detected in the transmission link but the playback security value X4 is ≥80.

[0080] For example, in this embodiment, modal feature vectors are first extracted, and the modal feature vector difference is calculated with the baseline content. The OCR then recognizes the keyword "limited-time 50% off". YOLOv5 detected a blurred region in frame #7. Background sound abrupt change detection; Multimodal weight dynamics Similarly, we can calculate... , Then, the playback safety value Y = 1 − (0.42 × 0.15 + 0.33 × 0.4 + 0.25 × 0.25) = 75.45 is calculated; the judgment result X4: 70 < Y = 75.45 < 80, then the security center calls the airport advertising screen security plan database to implement a level-two emergency response according to the risk level.

[0081] As one possible implementation, the method further includes: after at least one of the following results is generated—a consistency verification result, a content security probability result, an integrity result, and a playback security value result—updating the real-time credit score to obtain an updated real-time credit score; the updated real-time credit score is used to dynamically adjust the content security probability of the next content to be displayed and the playback security value of the next content to be played.

[0082] It should be noted that in this embodiment, when the credit federated learning engine receives a judgment result from X1, X2, X3, or X4, it initiates a real-time credit score update mechanism. Where β is the violation penalty coefficient, which is adjusted according to the violation type, for X3 (instruction integrity not passed). Set to -3; for X1 (consistency verification failed) and X4 (low playback security value). -2; for X2 (low probability of content security). Set to -1; V represents the number of violations of the same type in the past 30 days, calculated using a time sliding window; R k This is the k-th compliance reward, which accumulates as the index decays. For compliance reward coefficient, Set the value to 1; dynamically update the real-time credit score using the above update mechanism.

[0083] For example, in this embodiment, it is assumed that the merchant's current credit score The score is 82. During this review and playback period, the following events were triggered:

[0084] First, its content undergoes pre-approval, generating broadcast control instructions.

[0085] Subsequently, the real-time playback safety monitoring module determined that its playback safety value X4 = 75.45 (less than the safety threshold of 80), and the determination result X4 was "not passed".

[0086] The credit federated learning engine received the X4 result and initiated an update. A check revealed that the merchant had not committed any other X4 type violations in the past 30 days; therefore, V = 1 (this is a new occurrence).

[0087] Select the penalty coefficient based on the violation type x4. = -2. No compliance record, FedAvg({R_k}) = 0.

[0088] Credit score updates are calculated as follows: .

[0089] In this embodiment, an airport terminal display screen content review system is also provided, including an advertising screen terminal layer, a unified access layer, a content security review layer, an anti-tampering detection layer, a security center, an airport security broadcast control gateway, and a credit update and emergency response layer;

[0090] The advertising screen terminal layer is the physical carrier layer of the system, configured to carry and play advertising content, and includes three types of heterogeneous terminals. Among them, type A terminals are advertising screens managed by third parties entrusted by the airport, which have remote content publishing interfaces and realize content updates through third-party systems; type B terminals are advertising screens managed by merchants in the airport's commercial and catering areas, which are connected to the airport network but have independent management permissions and support local content uploads; type C terminals are advertising screens owned by the airport, which need to achieve standardized access through network upgrades; the terminal layer establishes a data connection with the unified access layer, receives broadcast control instructions, and provides feedback on real-time playback status;

[0091] The unified access layer is configured to achieve standardized and secure access for three types of heterogeneous advertising screens. Specifically, for Type A terminals (third-party managed terminal displays), a protocol converter and API security gateway are deployed. The protocol converter adapts to the private communication protocols of the third-party system (such as custom TCP / IP protocols) and converts them to system-wide protocols (such as HTTPS). The API security gateway implements access authentication, encrypted data transmission, and access control. For Type B terminals (merchant-managed terminal displays), a lightweight audit SDK is integrated. This SDK is embedded in the terminal operating system's frame buffer and automatically injects content hash values ​​based on the SHA-256 algorithm during content rendering, while also supporting lightweight data interaction with the unified access layer. For Type C terminals (airport-owned terminal displays), 5GMEC private network access is achieved through network module upgrades, and a TEE (Trusted Execution Environment) is deployed to ensure isolation between the terminal operating system and the application layer, preventing unauthorized access. The unified access layer transmits content data and status information from the terminal layer to the content security audit layer, while simultaneously forwarding instructions from the broadcast control gateway to the terminal.

[0092] The content security review layer is the core detection layer of the system, configured to achieve collaborative processing of pre-release content review and real-time playback content monitoring. It includes two parallel sub-modules: a content security review sub-module and a real-time playback content security detection sub-module. The content security review sub-module is used to segment the content to be displayed and calculate the segment hash value for each of the processed segments. Based on the segment hash values ​​of each segment, the consistency of the content to be displayed is verified using a Byzantine fault tolerance mechanism. It is also used to segment the content to be displayed and calculate the segment hash value for each of the processed segments. Based on the segment hash values ​​of each segment, the consistency of the content to be displayed is verified using a Byzantine fault tolerance mechanism. The hash value is used to verify the consistency of the content to be displayed using a Byzantine fault tolerance mechanism. The real-time playback content security detection submodule is used to obtain multiple first modal feature vectors of the currently playing content and multiple second modal feature vectors of the benchmark content in the benchmark database. Using the multiple first modal feature vectors and the real-time credit score of the sending entity of the currently playing content, the credit dynamic weights of the multiple first modal feature vectors are calculated. Using the multiple first modal feature vectors, multiple second modal feature vectors and multiple credit dynamic weights, the playback security value of the currently playing content is calculated. When the playback security value is less than a preset security threshold, a pre-built security plan is invoked to process the playback content.

[0093] The anti-tampering detection layer is used to verify the integrity of content throughout the entire publishing and playback process. It generates a unique hash value for the approved content, which is then double-signed by the airport operator (RSA-PSS algorithm) and the advertiser (SM2 algorithm) and stored in the Hyperledger Fabric blockchain; it outputs the anti-tampering result X3.

[0094] The security center is configured to receive abnormal results (X1 / X2 / X3 / X4) from the content security review layer, and perform emergency intervention and information recording. Specifically, this includes: when X1 < 85% or X2 < 0.90, triggering a node isolation mechanism to cut off the connection between the abnormal edge node and the system, and intercepting the content; when X3 is determined to be a failure of evidence storage, rolling back the instruction to the broadcast control gateway to call the built-in instruction correction module to verify and correct the format, signature information, or encryption parameters of the original instruction (such as regenerating the signature or adjusting the blockchain write node); when X4 < 80, calling the preset security contingency plan library; and recording the source information of all abnormal events in real time (including terminal ID, timestamp, content hash, and abnormal type), and synchronizing it to the blockchain evidence storage module.

[0095] The airport security broadcast control gateway is the system's command center, configured to receive the pass results of the content security review submodule, generate standardized broadcast control instructions (including playback time period and content index); receive adjustment instructions from the credit and emergency response layer, dynamically modify broadcast control parameters (such as pausing playback and switching content); and send the final instructions to the corresponding advertising screen terminal through an encrypted channel (AES-256 algorithm), and receive instruction execution feedback from the terminal.

[0096] The credit and emergency response layer is configured to achieve dynamic credit updates and tiered emergency handling, including a credit federated learning engine: the credit federated learning engine receives X1 / X2 / X3 / X4 result data and combines it with historical credit scores. Real-time credit scores are calculated using a federated learning algorithm (aggregating parameters from high-credit nodes every 24 hours). The credit score is then fed back to the content security review layer, and the review threshold is dynamically adjusted (e.g., the P threshold for high-credit subjects is reduced to 0.8).

[0097] Specifically, in this embodiment, a multi-layered secure content review architecture is constructed by utilizing Byzantine fault tolerance mechanism for consistency verification of the content to be displayed, content security probability determination based on bit similarity score and real-time credit score, and integrity determination of broadcast control instructions. This achieves multi-layered review and determination of the content to be displayed, completing the pre-screening of the content to be displayed. By obtaining the dynamic credit weights of each modality vector of the currently playing content and combining them with the calculation of the playback security value of the currently playing content, security monitoring of the real-time playing content is achieved. By combining the determination results of multiple verification judgments, the real-time credit score is dynamically updated and adjusted to achieve the purpose of dynamically adjusting the content security probability of the next content to be displayed and the playback security value of the next content to be played.

[0098] Example 2

[0099] See Figure 2 The present invention also provides an airport terminal display screen content review device, which is used in the airport terminal display screen content review method described in any one of the above claims, the device comprising:

[0100] The content consistency verification module 100 is used to perform fragmentation processing on the content to be displayed and calculate the fragment hash value of each of the multiple fragments after processing; based on the fragment hash value of each of the multiple fragments, the consistency of the content to be displayed is verified using the Byzantine fault tolerance mechanism.

[0101] The content security probability determination module 200 is used to calculate the security probability of multiple content segments by combining the real-time credit score of the sending entity of the content to be displayed with the security probability calculation function after the consistency verification is passed, so as to obtain the content security probability of the content to be displayed, and output the broadcast control instruction after the content security probability meets the preset conditions.

[0102] The instruction integrity determination module 300 is used to obtain the hash value of the broadcast control instruction, and use the hash values ​​of the same source instructions in the historical database to determine the integrity of the broadcast control instruction; after the determination is passed, the broadcast control instruction is signed and encrypted, and then the encrypted instruction is uploaded to the blockchain for notarization; if the notarization is successful, the broadcast control instruction is sent to the advertising screen execution terminal to play according to the broadcast control instruction.

[0103] It should be noted that the modules in the system of Embodiment 2 correspond to the steps in the method of Embodiment 1. The steps in the method of Embodiment 1 have been described in detail in Embodiment 1, and the module content in the system will not be described in detail in this Embodiment 2.

[0104] Example 3

[0105] See Figure 3 This embodiment also provides a computer device, including a system memory 1005 and a processor 1001. The system memory 1005 stores a computer program, and the processor 1001 executes the computer program to implement the steps of any of the methods described above.

[0106] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, when the processor 1001 executes the computer program, it implements the functions of each module / unit in the above system / device embodiments.

[0107] Specifically, in this embodiment, the computer program can be divided into one or more modules / units. One or more modules / units are stored in the system memory 1005 and executed by the processor 1001 to complete this application. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0108] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will understand that this does not constitute a limitation on the terminal device, which may include more or fewer components than shown, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0109] The processor 1001 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0110] System memory 1005 can be an internal storage unit of the terminal device, such as a hard drive or RAM. System memory 1005 can also be a storage device 1004 of the terminal device, such as an external hard drive, SmartMedia Card (SMC), Secure Digital (SD) card, or FlashCard. Furthermore, system memory 1005 can include both internal storage units and storage device 1004. System memory 1005 is used to store computer programs and other programs and data required by the terminal device. System memory 1005 can also be used to temporarily store data that has been output or will be output.

[0111] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0112] Example 4

[0113] This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0114] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof, or any other form of computer-readable storage medium in the art.

[0115] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside within an application-specific integrated circuit (ASIC). In embodiments of the invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device.

[0116] Example 5

[0117] This embodiment also provides a computer program product containing instructions that, when executed by a cluster of computer devices, cause the cluster of computer devices to perform the method described in Embodiment 1.

[0118] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An airport terminal display screen content review method, characterized by, The method comprises: The method comprises: The method comprises: When the number of hash values that are equal reaches a preset ratio of the total number of hash values, the consistency verification of the to-be-displayed content is passed; After the consistency verification is passed, the Hamming distance between the hash value of each piece of content and the hash value of the reference content is calculated to obtain the content similarity of each piece of content; The mean and standard deviation of each piece of content and the reference content are obtained, and the image structure similarity of each piece of content is calculated; the image structure similarity and the content similarity are weighted to obtain the bit similarity score of each piece of content; The real-time credit score of the sender of the to-be-displayed content is obtained, and the content security probability of the to-be-displayed content is calculated by combining the bit similarity score of the plurality of pieces of content with the real-time credit score; and when the content security probability meets a preset condition, a broadcast control instruction is outputted; The hash value of the broadcast control instruction is obtained, and the integrity of the broadcast control instruction is judged by using the homologous instruction hash value in the historical database; after the judgment is passed, the broadcast control instruction is signed and encrypted, and the encrypted instruction is uploaded to the block chain for notarization after the encryption; If the notarization is successful, the broadcast control instruction is issued to the advertising screen execution terminal to play according to the broadcast control instruction.

2. The method of claim 1, wherein, The hash value of the broadcast control instruction is obtained, and the integrity of the broadcast control instruction is judged by using the homologous instruction hash value in the historical database, specifically: The hash value of the broadcast control instruction and the homologous instruction hash value in the historical database are obtained, and the difference between the hash value of the broadcast control instruction and the homologous instruction hash value is calculated; When the difference between the two hash values is less than a preset judgment threshold, the integrity judgment of the broadcast control instruction is passed.

3. The method of claim 1, wherein, After the broadcast control instruction is issued to the advertising screen execution terminal to play according to the broadcast control instruction, the method further comprises: A plurality of first modal feature vectors of the current playing content and a plurality of second modal feature vectors of the reference content in the reference database are obtained; The credit dynamic weight of the plurality of first modal feature vectors is calculated by using the plurality of first modal feature vectors and the real-time credit score of the sender of the current playing content; The play safety value of the current playing content is calculated by using the plurality of first modal feature vectors, the plurality of second modal feature vectors, and the plurality of credit dynamic weights; When the play safety value is less than a preset safety threshold, a pre-constructed safety plan is called to process the playing content.

4. The method of claim 3, wherein, The method further comprises: when one of the consistency verification result, the content security probability result, the integrity result and the play safety value result is generated, updating the real-time credit score to obtain an updated real-time credit score; and the updated real-time credit score is used to dynamically adjust the content security probability of the next content to be displayed and the play safety value of the next content to be played.

5. An apparatus for screening the contents of an airport terminal display screen, characterized by The device is used in the content review method of the airport terminal display screen according to any one of claims 1-4, and the device comprises: a content consistency verification module configured to perform fragmentation processing on the content to be displayed, and calculate a plurality of fragment hash values of the processed fragment contents; and verify the consistency of the content to be displayed based on the plurality of fragment hash values of the fragment contents by using a Byzantine fault tolerance mechanism; a content security probability determination module configured to, after the consistency verification passes, calculate the security probability of the plurality of fragment contents by using a security probability calculation function in combination with the real-time credit score of the sending subject of the content to be displayed, to obtain the content security probability of the content to be displayed, and output a play control instruction when the content security probability meets a preset condition; an instruction integrity determination module configured to obtain a hash value of the play control instruction, determine the integrity of the play control instruction by using a same-source instruction hash value in a historical database, sign and encrypt the play control instruction after the determination passes, upload the encrypted instruction to a block chain for notarization after the encryption, and if the notarization is successful, issue the play control instruction to an advertising screen execution terminal for playing according to the play control instruction.

6. A computer device comprising a system memory having stored thereon a computer program and a processor, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1-4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps of the method of any one of claims 1-4.

8. A computer program product comprising instructions, characterized in that, When the instructions are run by the computer device cluster, the computer device cluster is caused to execute the method of any one of claims 1-4.

Images