Indoor inspection position verification method and device based on code scanning positioning

By collecting environmental feature fingerprints around the inspection points and comparing them with real-time fingerprint similarity, the accuracy problem of indoor inspection location verification is solved, and effective verification and anti-counterfeiting of inspection locations are achieved.

CN123075366APending Publication Date: 2026-11-10BEIJING LEHU FUTURE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611361076.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-03
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing inspection methods have low positioning accuracy in indoor scenarios, making it difficult to accurately verify the inspection location, and are easily falsified, resulting in poor accuracy in inspection location verification.

Method used

The environmental fingerprints around the inspection points are collected as standard fingerprints. Real-time environmental fingerprints are collected simultaneously when the inspection personnel scan the code. The legality of the inspection location is verified by similarity calculation. Multimodal feature fusion and verification are performed by combining wireless signals, Bluetooth beacons, geomagnetism, air pressure and acoustic features.

Benefits of technology

It effectively prevents remote virtual positioning and fake check-in, improves the accuracy and robustness of inspection location verification, and reduces the false judgment rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123075366A_ABST
    Figure CN123075366A_ABST
Patent Text Reader

Abstract

The embodiment of the application relates to the technical field of inspection management, and provides an indoor inspection position verification method and device based on code scanning positioning, which comprises the following steps: acquiring environment feature fingerprints of an inspection point collected in advance as standard environment fingerprints; in response to a code scanning operation of an inspection personnel on the inspection point, acquiring an inspection point identifier and real-time environment fingerprints synchronously collected and encrypted and uploaded by an inspection terminal; decrypting the real-time environment fingerprints and calculating the similarity between the real-time environment fingerprints and the standard environment fingerprints, and determining that the inspection position is legal when the similarity exceeds a preset similarity threshold. The application utilizes the uniqueness and difficult-to-reproduce nature of indoor environment fingerprints, and does not depend on a global positioning system to realize accurate verification of an indoor inspection position, effectively prevents remote virtual positioning and fake clock-in, and improves the credibility of inspection records.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of inspection management technology, and more specifically to an indoor inspection location verification method and device based on QR code positioning. Background Technology

[0002] Centralized apartment operators typically manage multiple locations across different areas. They need to establish inspection points around buildings, floors, rooms, model rooms, vacant rooms, and public areas to conduct periodic or ad-hoc inspections of fire safety, hygiene service quality, and overall operational performance. Inspection personnel must carry mobile devices to each inspection point, scan the QR code posted there to access the corresponding inspection task form, fill in the inspection results, and submit the inspection record. The back-end system uses this information to confirm the execution of the inspection task and the presence of the inspection personnel.

[0003] Current inspection methods typically rely on the Global Positioning System (GPS) to determine the location of inspection personnel. After scanning the QR code at the inspection point, the mobile terminal automatically obtains and reports the GPS location information, or the inspection personnel manually select the inspection point location. The backend compares the reported location information with the preset location of the inspection point to determine whether the inspection personnel have arrived at the designated inspection point.

[0004] However, GPS positioning accuracy is low in indoor scenarios such as within buildings, between floors, and within rooms. It is also susceptible to signal obstruction and location drift, making it difficult to effectively distinguish between adjacent rooms and different floors, resulting in inaccurate verification of inspection locations. Furthermore, the QR codes posted at inspection points can be photographed, copied, or moved. It is common for inspection personnel to forge inspection records through remote virtual positioning and proxy attendance, making it difficult to verify the actual presence of inspection personnel based solely on reported location information. In summary, existing inspection methods are insufficient in terms of accuracy in verifying indoor inspection locations and preventing cheating during attendance checks, necessitating a new solution. Summary of the Invention

[0005] This application provides an indoor inspection location verification method and device based on QR code positioning to solve the technical problem of poor inspection location verification accuracy caused by low positioning accuracy and easy drift in existing inspection methods in indoor scenarios.

[0006] In a first aspect, embodiments of this application provide an indoor inspection location verification method based on QR code positioning, the method comprising: Obtain the standard environmental fingerprint of the inspection point corresponding to the inspection task. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that was collected in advance when the inspection task was issued. In response to the scanning operation of the inspection personnel at the inspection point, the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint synchronously collected and encrypted by the inspection terminal are obtained; The real-time environment fingerprint is decrypted, and the similarity between the real-time environment fingerprint and the standard environment fingerprint is calculated. When the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0007] Secondly, embodiments of this application provide an indoor inspection location verification device based on QR code positioning, which has the function of implementing the indoor inspection location verification method based on QR code positioning provided in the first aspect above. The function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above function, and the modules can be software and / or hardware.

[0008] In one embodiment, the indoor patrol location verification device based on QR code positioning includes: The input / output module is configured to acquire the standard environmental fingerprint of the inspection point corresponding to the inspection task, and respond to the inspection personnel's scanning operation of the inspection point, and receive the real-time environmental fingerprint and the inspection point identifier corresponding to the scanning operation synchronously collected and encrypted by the inspection terminal. The processing module is configured to decrypt the real-time environmental fingerprint and calculate the similarity between the real-time environmental fingerprint and the standard environmental fingerprint; when the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0009] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the indoor patrol location verification method based on barcode scanning as described in the first aspect.

[0010] Fourthly, embodiments of this application provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the indoor patrol location verification method based on barcode scanning as described in the first aspect.

[0011] Fifthly, embodiments of this application provide a computer program product containing instructions, which, when run on a computer, causes the computer to execute the indoor inspection location verification method based on barcode scanning provided in the first aspect.

[0012] Compared to existing technologies, in this embodiment, environmental feature fingerprints around the inspection point are pre-collected as standard environmental fingerprints when the inspection task is issued, and real-time environmental fingerprints are collected simultaneously when the inspection personnel scan the code. The legality of the inspection location is determined by the similarity between the real-time environmental fingerprint and the standard environmental fingerprint. This eliminates the reliance on GPS coordinates for inspection location verification, instead relying on the uniqueness and difficulty in replication of the actual electromagnetic and acoustic environment around the inspection point. Even if the QR code at the inspection point is photographed, copied, or moved, the environmental fingerprint of the environment where the remote forger is located will still not match the standard environmental fingerprint, thus effectively preventing remote virtual positioning and forged check-ins. Furthermore, by performing modal attitudinal quantification and weighted fusion on multimodal features, and combining wireless signal fluctuation variance and media access control address cluster detection to identify forged signals, and combining the spatial topology of the inspection area to perform consistency verification on similarity, the accuracy and robustness of location verification are improved, and the false positive rate is reduced. Attached Figure Description

[0013] The objectives, features, and advantages of the embodiments of this application will become readily understood by referring to the accompanying drawings and the detailed description of the embodiments. Wherein: Figure 1 A flowchart illustrating an indoor inspection location verification method based on QR code positioning provided in this application embodiment; Figure 2 A schematic diagram of an indoor inspection location verification device based on QR code positioning provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a computing device according to an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an inspection terminal in an embodiment of this application; Figure 5 This is a schematic diagram of a server structure in one embodiment of this application.

[0014] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0015] In existing technologies, building inspections typically involve inspectors carrying mobile devices to various inspection points. They scan QR codes posted at each point to access an inspection task form, fill in the inspection results, and upload location information, on-site photos, or videos. The backend compares the GPS location information reported by the mobile device with the preset location of the inspection point to determine if the inspector has arrived at the designated point. However, GPS positioning accuracy is low in indoor scenarios such as inside buildings, between floors, and within rooms, and is susceptible to signal obstruction and location drift, making it difficult to distinguish between adjacent rooms or different floors. Furthermore, the QR codes posted at inspection points can be photographed, copied, or moved, and it is common for inspectors to forge inspection records through remote virtual positioning or proxy attendance. Therefore, relying solely on the reported location information is insufficient to confirm the actual presence of inspectors.

[0016] Compared to existing technologies, in this embodiment, environmental feature fingerprints of the area surrounding the inspection point are pre-collected as standard environmental fingerprints when the inspection task is issued. Simultaneously, the inspection terminal collects the real-time environmental fingerprint of the current environment when the inspection personnel scan the code. The similarity between the real-time environmental fingerprint and the standard environmental fingerprint is calculated, and the legitimacy of the inspection location is determined based on whether the similarity exceeds a preset similarity threshold. Because the wireless signal strength distribution, Bluetooth beacon combinations, geomagnetic, air pressure, and acoustic characteristics of each room or area are unique and difficult to replicate, even if the QR code is photographed and copied, the environmental fingerprint of the remote forger's environment cannot match the standard environmental fingerprint of the inspection point. This effectively prevents remote virtual positioning and forged check-ins without relying on GPS.

[0017] In some implementations, the inspection management device and the inspection terminal are deployed separately. For example, the indoor inspection location verification method based on QR code positioning provided in this application embodiment can be implemented based on an indoor inspection location verification system. This indoor inspection location verification system may include a server 01 and an inspection terminal 02.

[0018] Server 01 can be an inspection management device, in which inspection management programs can be deployed. Inspection terminal 02 can be a terminal device used by inspection personnel, in which inspection applications can be deployed.

[0019] The inspection terminal 02 can collect environmental feature fingerprints around the inspection point when the inspection task is issued, and can also collect real-time environmental fingerprints of the current environment when the inspection personnel scan the code. The real-time environmental fingerprints are then encrypted and sent to the server 01. The server 01 can obtain the standard environmental fingerprints pre-collected at the inspection point, receive the real-time environmental fingerprints synchronously collected and encrypted by the inspection terminal 02, calculate the similarity between the real-time environmental fingerprints and the standard environmental fingerprints, and determine whether the inspection location is legal based on whether the similarity exceeds a preset similarity threshold, thereby determining whether the corresponding inspection record can be submitted.

[0020] It should be noted that the computing devices involved in the embodiments of this application may be servers and / or terminal devices.

[0021] The server involved in the embodiments of this application can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0022] The terminal devices involved in the embodiments of this application can be devices that provide voice and / or data connectivity to users, handheld devices with wireless connectivity, or other processing devices connected to a wireless modem. Examples include mobile phones (or "cellular" phones) and computers with mobile terminals, such as portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile devices that exchange voice and / or data with a wireless access network. Examples include Personal Communication Service (PCS) phones, cordless phones, Session Initiation Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), and other similar devices.

[0023] Reference Figure 1 , Figure 1 This is a flowchart illustrating an indoor inspection location verification method based on QR code positioning, provided in an embodiment of this application. The method can be executed by an indoor inspection location verification device based on QR code positioning. This device can be integrated into inspection management equipment (e.g., a server) or deployed in an inspection terminal. The method includes steps 101-104: Step 101: Obtain the standard environmental fingerprint of the inspection point corresponding to the inspection task. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that was collected in advance when the inspection task was issued. Step 102: In response to the inspection personnel's scanning operation of the inspection point, obtain the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint synchronously collected and encrypted by the inspection terminal. Step 103: Decrypt the real-time environment fingerprint and calculate the similarity between the real-time environment fingerprint and the standard environment fingerprint; Step 104: When the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0024] Steps 101 to 104 above link the acquisition of standard environmental fingerprints, the real-time environmental fingerprint collection triggered by barcode scanning, fingerprint similarity calculation, and location legality determination into a continuous processing procedure. The optional implementation methods for each step are described below.

[0025] Step 101: Obtain the standard environmental fingerprint of the inspection point corresponding to the inspection task. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that was collected in advance when the inspection task was issued.

[0026] Specifically, in step 101, when the inspection task is issued, environmental feature fingerprints are collected in advance around the inspection point, and the environmental feature fingerprints are associated with the inspection point identification and stored to generate a standard environmental fingerprint. A QR code can be affixed to the inspection point. The QR code carries coded information generated by the inspection management device using preset coding rules, and the coded information includes the inspection point identification.

[0027] As an optional embodiment, when the inspection task is issued, pre-collected environmental feature fingerprints around the inspection point are acquired, and the environmental feature fingerprints are associated with and stored with the inspection point identifier to generate the standard environmental fingerprint. When the environment around the inspection point changes, the environmental feature fingerprints around the inspection point are re-collected, and the standard environmental fingerprint corresponding to the inspection point is updated.

[0028] The environmental fingerprint can include at least one of the following: the Media Access Control (MAC) address and Received Signal Strength Indication (RSSI) sequence of a visible Wireless Fidelity (WiFi) access point, Bluetooth Low Energy (BLE) beacon broadcast data, and the ambient background noise spectrum. Because the spatial distribution of indoor wireless signals, Bluetooth beacons, and background noise varies with location and is difficult to replicate, the environmental fingerprint can serve as an invisible landmark for inspection points.

[0029] Taking inspection point A as an example, the inspection management device generates a QR code for inspection point A using a preset coding rule. The QR code carries the inspection point identifier A-01. When an inspection task is issued, the inspection terminal scans the area around inspection point A to obtain three visible Wi-Fi access points. It collects the Media Access Control (MAC) address and received signal strength (RSS) indication of each access point. For example, the MAC address of access point 1 is AA:BB:CC:DD:EE:01, and the RSS indication is -45dBm; the MAC address of access point 2 is AA:BB:CC:DD:EE:02, and the RSS indication is -52dBm; and the MAC address of access point 3 is AA:BB:CC:DD:EE:03, and the RSS indication is -60dBm. Simultaneously, it receives broadcast data and signal strength from a Bluetooth Low Energy beacon and collects the ambient background noise spectrum in the 20Hz to 20kHz frequency band. The inspection management device uses the media access control addresses and received signal strength indication sequences of the three access points, Bluetooth Low Energy beacon broadcast data, and ambient background noise spectrum as the environmental feature fingerprint of inspection point A, and stores them in association with the inspection point identifier A-01 to generate a standard environmental fingerprint for inspection point A. The aforementioned media access control addresses, received signal strength indication values, number of beacons, and frequency band ranges are merely examples and do not constitute a limitation on the embodiments of this application.

[0030] As an optional embodiment, multimodal environmental feature data around the inspection point is collected. The multimodal environmental feature data includes at least one of the following: the media access control address and received signal strength indication sequence of the wireless access point, Bluetooth low power beacon broadcast data, geomagnetic field triaxial data collected by the geomagnetic sensor, air pressure value collected by the barometer, and room impulse response characteristics of active acoustic detection. Feature extraction is performed on each modality of the multimodal environmental feature data to obtain the feature sub-vectors corresponding to each modality. The feature sub-vectors corresponding to each modality are normalized and then concatenated in a preset order to form the environmental feature fingerprint. The environmental feature fingerprint is associated with the inspection point identifier and stored, and a mapping relationship between the inspection point identifier and the environmental feature fingerprint is established.

[0031] The media access control address and received signal strength indication sequence of the wireless access point can be obtained by scanning the visible access points around the inspection point using the wireless fidelity module of the inspection terminal, including the media access control address and received signal strength indication of each access point; Bluetooth low power beacon broadcast data can be obtained by receiving beacon broadcasts around the inspection point through the Bluetooth module; geomagnetic field triaxial data can be obtained by collecting the components of the geomagnetic field around the inspection point along the three orthogonal axes using a geomagnetic sensor; air pressure values ​​can be collected by a barometric pressure sensor; room impulse response characteristics can be obtained by emitting a sound wave from a speaker, receiving the reflected sound wave through a microphone, and extracting the room's impulse response. Different modalities have different data dimensions and value ranges. Extracting suitable features from each modal data, normalizing them to a unified range, and then concatenating them in a preset order to form an environmental feature fingerprint enables the features of different modalities to be comparable in subsequent similarity calculations. The above preset order and specific features are merely examples and do not constitute a limitation on the embodiments of this application.

[0032] In the above embodiments, the mean and variance features of signal strength are extracted from the received signal strength indication sequence and the Bluetooth Low Energy beacon broadcast data; the triaxial components and magnetic field magnitude features are extracted from the geomagnetic field triaxial data; the mean and fluctuation features of air pressure are extracted from the air pressure value; and the frequency band energy features are extracted from the room impulse response features.

[0033] Taking inspection point A as an example, the received signal strength indication sequence collected by the inspection terminal around inspection point A is {-45dBm, -46dBm, -44dBm}, and the mean signal strength is extracted as -45dBm and the variance as 0.67. Furthermore, the signal strength sequence corresponding to the collected Bluetooth Low Energy beacon broadcast data is {-68dBm, -69dBm, -67dBm}, and the mean signal strength is extracted as -68dBm and the variance as 0.67. The collected geomagnetic field triaxial data is (25.3μT, -12.8μT, -33.6μT), and the triaxial components (25.3, -12.8, -33.6) and magnetic field magnitude as 44.2μT are extracted. The collected air pressure value is 1013.2hPa, and the mean air pressure is extracted as 1013.2hPa and the fluctuation as 0.3hPa. Furthermore, the energy of each frequency band is extracted from the room impulse response features obtained by active acoustic detection, resulting in a frequency band energy feature vector. After normalizing the feature sub-vectors corresponding to each mode, they are concatenated in a preset order of wireless signal, geomagnetism, air pressure, and acoustics to obtain the environmental feature fingerprint vector of inspection point A. This environmental feature fingerprint vector is then associated with and stored with the inspection point identifier A-01, establishing a mapping relationship between the inspection point identifier A-01 and the environmental feature fingerprint vector. The specific values, extraction rules, and concatenation order of the above features are merely examples and do not constitute a limitation on the embodiments of this application.

[0034] Step 102: In response to the inspection personnel's scanning operation of the inspection point, obtain the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint synchronously collected and encrypted by the inspection terminal.

[0035] Specifically, in step 102, after arriving at the inspection point, the inspection personnel scan the QR code posted at the inspection point using the inspection terminal. In response to this scanning operation, the inspection terminal simultaneously collects environmental characteristic data of the current environment, forming a real-time environmental fingerprint, which is then encrypted and uploaded to the server. The server parses the QR code to obtain the inspection point identifier.

[0036] As an optional embodiment, the encoded information carried by the QR code corresponding to the scanning operation is parsed to obtain the inspection point identifier. The encoded information is generated by the inspection management device using preset encoding rules, and other software cannot parse and recognize this encoded information. Then, based on the inspection point identifier, the identity information of the inspection personnel and the address information reported by the inspection terminal are verified.

[0037] Optionally, the dynamic refresh parameters or timestamps in the encoded information are parsed to verify whether they are within a preset validity period. If the verification passes, the identity information of the inspector is retrieved from the pre-stored authorized inspector information database to determine whether the inspector is an authorized inspector; the expected location range corresponding to the inspection point identifier is obtained, and it is determined whether the address information reported by the inspection terminal falls within the expected location range. If the inspector is an authorized inspector and the address information falls within the expected location range, the verification passes.

[0038] The QR code carries encoded information generated by the inspection management device using preset encoding rules. Each code is unique and cannot be parsed or recognized by other software, preventing the QR code at the inspection point from being replaced or forged. The encoded information may include dynamic refresh parameters or timestamps to prevent inspection personnel from reusing the QR code after taking a photo; verification only passes when the dynamic refresh parameters or timestamp are within a preset validity period. Identity and address information verification are used to confirm that the person performing the scanning operation is an authorized inspection personnel and that the inspection terminal is within the expected location range corresponding to the inspection point. Together with subsequent environmental fingerprint verification, this constitutes a multi-layered anti-fraud verification process.

[0039] Specifically, a pre-defined encoding rule is used to generate a QR code for each inspection point. The QR code carries encoding information that may include the inspection point identifier, dynamic refresh parameters, and a timestamp. Upon arriving at the inspection point, inspection personnel scan the QR code using their inspection terminal. The terminal parses the encoding information to obtain the inspection point identifier, dynamic refresh parameters, and timestamp. Simultaneously, it collects a real-time environmental fingerprint of the current environment, encrypts the fingerprint, and uploads it along with the parsed encoding information to the server. Upon receiving the uploaded data, the server first parses the dynamic refresh parameters or timestamp in the encoding information to verify if it is within a pre-defined validity period, for example, by determining if the difference between the current time and the timestamp does not exceed a pre-defined duration. Then, based on the inspection point identifier, it queries the pre-stored authorized inspector information database to determine if the inspector is an authorized inspector; and it obtains the expected location range corresponding to the inspection point identifier to determine if the address information reported by the inspection terminal falls within this expected location range. If the inspector is an authorized inspector and the address information falls within the expected location range, the identity and address information verifications pass, and the process proceeds to the subsequent calculation of the similarity between the real-time environmental fingerprint and the standard environmental fingerprint. If any verification fails, further processing is refused, and a prompt message is returned to the inspection terminal. The above-described encoded information structure, validity period determination method, and verification order are merely examples and do not constitute a limitation on the embodiments of this application.

[0040] In one optional implementation, the inspection terminal can encrypt the real-time environmental fingerprint using a symmetric encryption algorithm. For example, a shared session key or key seed is pre-configured between the inspection terminal and the server. The inspection terminal uses the Advanced Encryption Standard (AES) algorithm or the Chinese national cryptographic standard SM4 algorithm to encrypt the real-time environmental fingerprint using the session key, generating ciphertext. Alternatively, the inspection terminal can also use a hybrid encryption method, that is, first negotiating a session key using an asymmetric encryption algorithm (such as the RSA algorithm or the Chinese national cryptographic standard SM2 algorithm), and then using the negotiated session key to encrypt the real-time environmental fingerprint using a symmetric encryption algorithm. Compared with directly using an asymmetric encryption algorithm to encrypt the real-time environmental fingerprint, the hybrid encryption method takes into account the encryption efficiency of symmetric encryption and the key distribution security of asymmetric encryption, and is more suitable for scenarios with large amounts of environmental fingerprint data that need to be uploaded frequently.

[0041] To ensure the integrity and prevent replay of real-time environmental fingerprints, the inspection terminal, while encrypting the real-time environmental fingerprint, can also calculate a digest of the real-time environmental fingerprint using a hash algorithm (such as SHA-256 or the national cryptographic SM3 algorithm), and upload it along with the collection timestamp or random number. Upon receiving the ciphertext, the server decrypts it using the decryption key corresponding to the one used by the inspection terminal to obtain the real-time environmental fingerprint; it then calculates the digest of the decrypted real-time environmental fingerprint using the same hash algorithm and compares the calculated digest with the digest uploaded by the inspection terminal. If they match, it is determined that the real-time environmental fingerprint has not been tampered with during transmission. The server can also determine whether the uploaded data is replay data based on the collection timestamp or random number. For example, if the collection timestamp exceeds a preset valid duration, or the random number has already appeared in the historical record, it is determined to be a replay attack and the data is rejected. By combining encryption, integrity verification, and replay prevention verification, the real-time environmental fingerprint can be prevented from being eavesdropped on, tampered with, or replayed during transmission, ensuring the credibility of the inspection location verification. The above encryption algorithm, hash algorithm, and key negotiation method are only examples and do not constitute a limitation on the embodiments of this application.

[0042] Step 103: Decrypt the real-time environment fingerprint and calculate the similarity between the real-time environment fingerprint and the standard environment fingerprint.

[0043] Specifically, in step 103, the server uses the decryption key corresponding to the encryption key used by the inspection terminal to decrypt the encrypted real-time environmental fingerprint to obtain the decrypted real-time environmental fingerprint; the server performs feature analysis on the decrypted real-time environmental fingerprint to obtain a multimodal real-time feature vector; based on the multimodal real-time feature vector and the multimodal standard feature vector corresponding to the standard environmental fingerprint, the server simulates the current environment and determines the sub-environment labels corresponding to each modality; the server performs a weighted summation of the sub-environment labels corresponding to each modality based on the preset weights of each modality to obtain the comprehensive similarity between the real-time environmental fingerprint and the standard environmental fingerprint.

[0044] As an optional embodiment, for each modality, real-time feature sub-vectors of each modality are extracted from the multimodal real-time feature vector, and standard feature sub-vectors of each modality are extracted from the multimodal standard feature vector; the feature distance or similarity between the real-time feature sub-vectors and the standard feature sub-vectors is calculated, and the feature distance or similarity is used as the initial sub-environment label corresponding to each modality; fluctuation variance analysis is performed on the wireless signal strength sequence in the real-time environment fingerprint, and if the fluctuation variance is less than a preset variance threshold, it is determined that the first modality has a suspected spoofed signal, and the weight of the initial sub-environment label of the first modality is reduced; it is detected whether there is an abnormal media access control address cluster in the real-time environment fingerprint, and if so, it is determined that there is a suspected spoofed signal, and the weight of the initial sub-environment label of the corresponding second modality is reduced; the initial sub-environment labels corresponding to each modality are summarized and determined as the sub-environment labels corresponding to each modality.

[0045] The feature distance or similarity can be calculated using methods such as Euclidean distance and cosine similarity, and is used to characterize the degree of consistency between the current environment of the inspection terminal under the corresponding modality and the environment corresponding to the standard environment fingerprint. Under normal conditions, wireless signal strength naturally fluctuates over time. If the variance of the fluctuation of the wireless signal strength sequence is too small, it indicates that the signal may be generated by a stable replay of a forged signal source, and belongs to the category of suspected forged signals. An abnormal media access control address cluster refers to multiple access points using adjacent or regularly arranged media access control addresses, which are usually generated by the same forged signal source in batches. Through the above fluctuation variance analysis and abnormal media access control address cluster detection, forged signals can be identified, and the weight of the corresponding modality can be reduced to avoid interference from forged signals in the similarity calculation.

[0046] As an optional embodiment, at least one of the following contextual information—current time period, regional importance, and historical compliance rate—is obtained, and the preset weights corresponding to each modality are dynamically adjusted based on the contextual information to obtain dynamic weights. Based on the dynamic weights corresponding to each modality, the sub-environmental labels corresponding to each modality are weighted and summed to obtain an initial comprehensive similarity. A graph structure of the inspection area is constructed, where nodes represent inspection points and edges represent the spatial distance between inspection points. Based on the real-time environmental fingerprint and the standard environmental fingerprints of adjacent inspection points, the spatial topological consistency between the real-time environmental fingerprint and the adjacent inspection points is verified through the graph structure to obtain a topological consistency result. If the topological consistency result indicates that the verification is consistent, the initial comprehensive similarity is used as the comprehensive similarity. If the topological consistency result indicates that the verification is inconsistent, the initial comprehensive similarity is reduced to obtain the comprehensive similarity.

[0047] Different modalities exhibit varying discrimination capabilities across different inspection scenarios. For instance, in public areas with dense wireless signals, wireless signal characteristics offer stronger discrimination capabilities, while in well-shielded rooms, acoustic or geomagnetic characteristics provide stronger discrimination capabilities. The weights of each modality can be dynamically adjusted based on the current time period (e.g., day or night), regional importance (e.g., high-risk areas), and historical compliance rates (e.g., historical forgery frequency) to adapt the similarity calculation to the current scenario. Spatial topology consistency verification utilizes the spatial relationships between inspection points. For example, if the real-time environmental fingerprint is more similar to the standard environmental fingerprint of an adjacent inspection point than to the standard environmental fingerprint of the current inspection point, it indicates that the inspection personnel may have scanned the code at an adjacent inspection point rather than the current inspection point, resulting in a topology consistency discrepancy. In this case, the overall similarity is reduced to avoid misjudging it as legitimate.

[0048] In an optional implementation, the similarity calculation can also be implemented using a Siamese neural network model trained based on metric learning. The Siamese neural network model includes two feature extraction branches with shared parameters and a metric module. The network structures of the two feature extraction branches are identical and they share weight parameters. Multimodal real-time feature vectors and multimodal standard feature vectors are input into the two feature extraction branches respectively, and mapped by the feature extraction branches to low-dimensional real-time feature representations and standard feature representations. The metric module calculates the metric distance between the real-time feature representations and the standard feature representations, and obtains the similarity based on the metric distance. For example, each feature extraction branch can employ a multi-layer network consisting of multiple fully connected layers or convolutional layers, such as an input layer, two hidden layers, and an output layer. Each hidden layer uses a rectified linear unit activation function, and the output layer outputs a feature representation of a preset dimension. The metric module can calculate the distance between the two feature representations using Euclidean distance or cosine distance.

[0049] During the training phase, environmental feature fingerprints collected at the same inspection point at different times are used as positive sample pairs, and environmental feature fingerprints collected at different inspection points are used as negative sample pairs to construct training samples. A contrastive loss function is used to train the Siamese neural network model, ensuring that the metric distance between positive sample pairs is less than that between negative sample pairs. For example, the contrastive loss function can be expressed as L=y·d²+(1-y)·max(0,md)², where d is the metric distance between two feature representations, y is the label (positive sample pairs are set to 1, negative sample pairs to 0), and m is a preset boundary value. Through this training, the Siamese neural network model can automatically learn to weight and transform the features of each modality, ensuring that environmental feature fingerprints from the same inspection point are close to each other in the feature space, while environmental feature fingerprints from different inspection points are far apart.

[0050] Compared to directly using a fixed distance metric, the Siamese neural network model in this embodiment performs the following improvements for indoor inspection location verification scenarios: First, the two feature extraction branches share parameters, ensuring that real-time environmental fingerprints and standard environmental fingerprints are mapped to the same feature space; second, positive and negative sample pairs are constructed using fingerprints from the same point at different times and from different points, making the feature representation learned by the model sensitive to the positional differences between inspection points, but insensitive to the environmental drift of the same inspection point over time; third, the trained model can automatically adjust the weights of each modality feature according to the environmental characteristics of each inspection point, without the need to manually set distance metric parameters for each inspection point, thus adapting to the environmental differences of different inspection points. The above network structure, number of layers, dimensions, activation function, loss function, and boundary values ​​are only examples and do not constitute a limitation on the embodiments of this application.

[0051] For example, spatial topology consistency verification can also be implemented using a spatial topology verification model based on a graph neural network. The spatial topology verification model constructs a graph structure using inspection points as graph nodes and the spatial distance between adjacent inspection points as edge weights. The standard environmental fingerprints of each inspection point are input into the graph neural network as node features of the corresponding graph nodes. The graph neural network aggregates the node features of adjacent graph nodes through multi-layer message passing, updating the node representations of each graph node to obtain the topological environment representation of each inspection point. During verification, the real-time environmental fingerprints are mapped to real-time feature representations, and the real-time feature representations are matched with the topological environment representations of the corresponding inspection point and its adjacent inspection points. Topological consistency is determined based on the matching results. For example, the graph neural network can employ a two-layer graph convolutional network. Each layer updates the node representation by weighted aggregation of the node representations of the graph nodes and their adjacent graph nodes, followed by a nonlinear transformation. The edge weights are used to weight the aggregation contribution of adjacent graph nodes.

[0052] During the training phase, the standard environmental fingerprints of each inspection point can be used as node features, and the spatial topological relationships between inspection points (such as adjacency relationships or spatial distance thresholds) can be used as supervision signals to train the spatial topology verification model. This will make the topological environment representations of topologically adjacent inspection points similar to each other, while the topological environment representations of topologically non-adjacent inspection points will be distinguishable from each other. Alternatively, the correspondence between the standard environmental fingerprints and the corresponding inspection point identifiers can be used as supervision signals to train the model to learn the mapping from environmental feature fingerprints to inspection point locations.

[0053] Compared to directly calculating the distance between real-time and standard environmental fingerprints, the spatial topology verification model in this application embodiment performs the following improvements for indoor inspection location verification scenarios: First, it introduces the spatial topological relationship between inspection points into similarity verification, making location verification not only dependent on fingerprint matching of a single inspection point but also constrained by the fingerprints of adjacent inspection points; Second, it aggregates the environmental features of adjacent inspection points through a message passing mechanism, enabling it to identify situations where the real-time environmental fingerprint is more similar to that of adjacent inspection points than to the current inspection point, i.e., identifying cheating behavior where inspection personnel impersonate the current inspection point after scanning the code at an adjacent inspection point; Third, the graph neural network can automatically learn the topological association weights between different inspection points without the need for manually setting topological consistency judgment rules. The above graph structure construction method, number of graph neural network layers, aggregation method, and training signal are only examples and do not constitute a limitation on the embodiments of this application.

[0054] Step 104: When the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0055] As an optional embodiment, when the similarity does not exceed the preset similarity threshold, the inspection record corresponding to the inspection point is marked as a suspicious inspection record, or the submission of the inspection record is rejected, and a prompt message for re-collection or relocation is returned to the inspection terminal. When the similarity exceeds the preset similarity threshold, the submission of the inspection record corresponding to the inspection point is allowed.

[0056] Specifically, after calculating the comprehensive similarity between the real-time environmental fingerprint and the standard environmental fingerprint, the server compares the comprehensive similarity with a preset similarity threshold. If the comprehensive similarity does not exceed the preset similarity threshold, it indicates that the environment in which the inspection terminal is currently located does not match the standard environment of the inspection point, which may be due to remote virtual positioning, proxy attendance, or being in the wrong location. The server marks the inspection record corresponding to the inspection point as a suspicious inspection record, or directly rejects the submission of the inspection record, and returns a prompt message to the inspection terminal to re-collect the real-time environmental fingerprint or relocate, so as to prompt the inspection personnel to re-scan the code at the designated inspection point. If the comprehensive similarity exceeds the preset similarity threshold, it indicates that the environment in which the inspection terminal is currently located matches the standard environment of the inspection point. The server determines that the inspection location is legal and allows the submission of the inspection record corresponding to the inspection point. The above threshold comparison method and processing actions are only examples and do not constitute a limitation on the embodiments of this application.

[0057] Optionally, after obtaining the comprehensive similarity between the real-time environmental fingerprint and the standard environmental fingerprint, fluctuation variance analysis can be performed on the wireless signal strength sequence in the real-time environmental fingerprint. If the fluctuation variance is less than a preset variance threshold, a suspected forged signal is identified, and the comprehensive similarity is reduced. The system also detects whether an abnormal media access control address cluster exists in the real-time environmental fingerprint; if so, a suspected forged signal is identified, and the comprehensive similarity is reduced. The preset similarity threshold is dynamically adjusted based on the current time period, regional importance, historical compliance rate, or time decay model. Specifically, the preset similarity threshold is increased for high-risk areas or nighttime periods. By dynamically adjusting the threshold, location verification in high-risk areas and nighttime periods becomes more stringent, further improving the reliability of anti-cheating measures.

[0058] Specifically, fluctuation variance analysis can calculate the variance of the wireless signal strength sequence in the real-time environmental fingerprint within a preset time window. Under normal conditions, wireless signal strength fluctuates naturally due to factors such as personnel movement and environmental changes, and its variance is usually greater than the preset variance threshold. However, the signal strength sequence stably replayed by a forged signal source has minimal fluctuation. If the calculated variance is less than the preset variance threshold, it is determined that a suspected forged signal exists, and the overall similarity is reduced accordingly. Abnormal media access control address cluster detection can statistically analyze the media access control addresses of each access point in the real-time environmental fingerprint. If multiple access points are found to have adjacent media access control addresses or exhibit a regular arrangement (e.g., only the last digit increases sequentially), it indicates that these access points may have been batch-simulated by the same forged signal source, and a suspected forged signal is determined, and the overall similarity is reduced accordingly. Dynamic threshold adjustment can dynamically adjust the preset similarity threshold based on the current time period (e.g., increasing the threshold during nighttime), regional importance (e.g., increasing the threshold in high-risk areas), historical compliance rate (e.g., increasing the threshold in areas with high historical forgery frequency), or time decay model, so that the strictness of location verification is adapted to the risk level of the inspection scenario. The above-mentioned variance threshold, time window, address pattern determination method, and threshold adjustment method are only examples and do not constitute a limitation on the embodiments of this application.

[0059] Optionally, fluctuation variance analysis can be achieved by: acquiring multiple sampled values ​​of the wireless signal strength sequence in the real-time environmental fingerprint within a preset time window, such as the received signal strength indication value obtained by the inspection terminal continuously sampling the same wireless fidelity access point multiple times within a preset time window after scanning; calculating the statistical variance of these multiple sampled values ​​as the fluctuation variance. Because wireless signal strength under normal conditions is affected by factors such as personnel movement, multipath propagation, and environmental changes, the received signal strength indication value at the same inspection point will naturally fluctuate over time; while the received signal strength indication value of a signal stably replayed by a spoofed signal source remains essentially unchanged. For example, under normal conditions, if five consecutive data samples are collected from the same access point within one second after scanning, the received signal strength indication values ​​may be -45dBm, -46dBm, -44dBm, -47dBm, and -45dBm, with a relatively large variance. However, when a spoofed signal source is stably replayed, these five collected values ​​may be -45dBm, -45dBm, -45dBm, -45dBm, and -45dBm, with a variance approaching zero. Therefore, if the calculated fluctuation variance is less than a preset variance threshold, the signal corresponding to the wireless signal strength sequence is determined to be a suspected spoofed signal, and the overall similarity is reduced accordingly. The preset variance threshold can be pre-set based on the fluctuation variance distribution of historically collected normal inspection point wireless signal strength sequences, for example, taking the lower limit of the normal fluctuation variance distribution. The above sampling number, time window, variance calculation method, and preset variance threshold are only examples and do not constitute a limitation on the embodiments of this application.

[0060] Anomaly media access control address cluster detection is achieved by extracting the media access control addresses of each access point from the real-time environmental fingerprint. The media access control address is typically represented by a 48-bit binary number, where the first 24 bits are the Organizationally Unique Identifier (OUI), used to identify the access point's manufacturer, and the last 24 bits are the network interface card (NIC) address assigned by that manufacturer. In a real inspection environment, access points typically come from different manufacturers or different production batches, and their OUI prefix and last 24 bits usually exhibit a random, irregular distribution. However, when multiple access points are generated in batches by the same forged signal source, the forged signal source typically assigns these access points media access control addresses with the same OUI prefix and consecutive or regularly arranged last 24 bits. Therefore, the media access control addresses (MUCs) of each access point in the real-time environmental fingerprint can be statistically analyzed to detect whether there are MUC clusters with the same OUI prefix and adjacent or regularly arranged last 24 bits. For example, multiple MUCs may have the same OUI prefix and their last 24 bits may increase sequentially. If a MUC cluster satisfying the above pattern is detected, and the number of MUCs in the cluster exceeds a preset threshold, then an abnormal MUC cluster is identified, leading to a suspected forged signal, and the overall similarity is reduced accordingly. For example, the MUCs of access points simulated by a forged signal source may be AA:BB:CC:00:00:01, AA:BB:CC:00:00:02, and AA:BB:CC:00:00:03, all with the OUI prefix AA:BB:CC and continuously increasing last 24 bits. However, the MUCs of actual access points usually come from different OUI prefixes, and their distribution is irregular. The above-mentioned MUC structure, pattern determination method, and threshold are only examples and do not constitute a limitation on the embodiments of this application.

[0061] Furthermore, a basic similarity threshold is pre-set, and the threshold adjustment amount is determined based on at least one of the following: current time period, regional importance, historical compliance rate, or time decay model. The threshold adjustment amount is then added to the basic similarity threshold to obtain the adjusted preset similarity threshold. Specifically, the current time period can be divided into daytime and nighttime periods based on the time of inspection task execution. Due to relatively weaker supervision and higher forgery risk during nighttime periods, the threshold adjustment amount can be increased during nighttime periods. Regional importance can be classified according to the importance of the area where the inspection point is located (e.g., whether it involves key areas such as fire-fighting facilities and power distribution rooms), and the threshold adjustment amount is increased for high-risk areas with higher importance. Historical compliance rate can be determined based on the proportion of forged or abnormal records in the historical inspection records of the inspection point or area, and the threshold adjustment amount is increased for inspection points or areas with low historical compliance rates (high frequency of historical forgery). The time decay model assigns decay weights to historical records based on their time to the present; the more recent the historical record, the greater its impact on the threshold adjustment, and the more distant the historical record, the smaller its impact on the threshold adjustment. For example, a basic similarity threshold can be set to 90%. When the inspection task is performed during nighttime, the threshold is increased by 5 percentage points. When the inspection point is located in a high-risk area, the threshold is increased by another 5 percentage points. When the historical compliance rate of the inspection point is lower than the preset compliance rate threshold, the threshold is increased by another 3 percentage points, resulting in the adjusted preset similarity threshold. By dynamically adjusting the threshold, the location verification in high-risk areas and during nighttime is made more stringent, further improving the reliability of anti-cheating measures. The aforementioned basic threshold, adjustment range, time period division method, regional importance division method, historical compliance rate calculation method, and time decay model are merely examples and do not constitute a limitation on the embodiments of this application.

[0062] As an optional embodiment, after determining that the inspection personnel's inspection location is legitimate, an image acquisition command can be sent to the inspection terminal. This command instructs the inspection personnel to capture images of the inspection point. The system receives the images uploaded by the inspection terminal and uses moiré pattern recognition algorithms and edge spectral analysis to determine whether the images were taken from a photograph of an electronic screen. If a photograph is found to be taken from a screen, the corresponding inspection record is marked as invalid, and a prompt for re-inspection is returned to the inspection terminal. If a photograph is not taken from a screen, the corresponding inspection record is marked as valid and submitted. Images taken from an electronic screen typically retain moiré patterns, and their edge spectral characteristics differ from directly captured images. By detecting moiré patterns in the image using a moiré pattern recognition algorithm and detecting high-frequency artifacts introduced by photographing from a screen using edge spectral analysis, it is possible to identify whether an image was taken from a screen, further preventing inspection personnel from using photographed historical images to impersonate current images.

[0063] Specifically, a moiré pattern recognition algorithm can perform frequency domain analysis on the scene image to detect the presence of periodic stripes, i.e., moiré patterns, formed by the interference between the pixel grid of the electronic screen and the pixel grid of the shooting sensor. If obvious moiré patterns are detected, it indicates that the scene image may have been copied from the electronic screen. Edge spectral analysis can extract the edge and spectral features of the scene image. Because the copied image has been displayed on the electronic screen and photographed twice, additional high-frequency artifacts, color shifts, or brightness distortions are usually introduced at the edges, resulting in differences in edge spectral features compared to the directly photographed scene image. By comparing the edge spectral features of the scene image with preset normal shooting features, it is determined whether the scene image was copied. If the moiré pattern recognition algorithm or edge spectral analysis determines that the scene image was copied, the corresponding inspection record is marked as invalid, and a prompt for re-inspection is returned to the inspection terminal. If it is determined that it was not copied, the corresponding inspection record is marked as valid and submitted. The above moiré pattern recognition method, edge spectral analysis method, and determination threshold are only examples and do not constitute a limitation on the embodiments of this application.

[0064] In one alternative implementation, the moiré pattern recognition algorithm can be implemented through frequency domain analysis. Specifically, the scene image is first preprocessed, for example, by converting the scene image into a grayscale image and performing noise reduction; then, a two-dimensional Fourier transform is performed on the preprocessed grayscale image to obtain a spectrum. Due to the spatial frequency difference between the pixel grid of the electronic screen and the pixel grid of the imaging sensor, the image obtained by copying will exhibit periodic peaks formed by pixel grid interference in the spectrum, in addition to the central low-frequency components, at positions off-center, which are the spectral peaks corresponding to moiré patterns. Therefore, it is possible to detect whether there are periodic peaks in the spectrum outside the central low-frequency region whose intensity exceeds a preset peak threshold. For example, the number of peak points in the spectrum whose peak intensity exceeds the preset peak threshold can be counted. If the number of peak points exceeds a preset number threshold, or the average intensity of the peak points exceeds a preset intensity threshold, it is determined that the scene image has moiré patterns, indicating that the scene image may have been copied from an electronic screen. For example, the spectrum can be high-pass filtered first to remove the central low-frequency components before detecting peaks to improve the accuracy of detection. The grayscale conversion, denoising, Fourier transform, peak detection methods and thresholds described above are merely examples and do not constitute a limitation on the embodiments of this application.

[0065] In one alternative implementation, edge spectral analysis can be achieved by combining edge feature extraction and spectral feature analysis. For edge feature extraction, edge detection operators (such as the Sobel or Canny operators) can be used to detect edges in the scene image and extract its edge map. Because the reproduced image has been displayed on an electronic screen and photographed a second time, its edges are often blurred or jagged due to the screen pixel grid and secondary sampling, resulting in lower edge sharpness. Furthermore, false edges such as screen borders and reflections may be mixed in at the edges. Features such as edge density, edge sharpness, or edge width of the edge map can be calculated, and the calculated edge features can be compared with the edge features corresponding to the directly captured normal image to determine whether the edge features of the scene image deviate from the normal captured features. In terms of spectral feature analysis, the on-site image can be frequency-domain transformed, dividing the spectrum into multiple frequency bands, and the energy of each frequency band can be calculated to obtain the frequency band energy distribution. Re-enhanced images typically show an abnormal increase in energy in the high-frequency band (e.g., high-frequency artifacts introduced by the screen pixel grid). The ratio between the high-frequency band energy and the low-frequency band energy can be calculated. If this ratio exceeds a preset ratio threshold, or if the difference between the frequency band energy distribution and the preset normal shooting frequency band energy distribution exceeds a preset difference threshold, then the on-site image is determined to have high-frequency artifacts introduced by re-enhancing. For example, edge features and spectral features can be combined for a comprehensive judgment. For instance, when the edge sharpness is lower than a preset sharpness threshold and the high-frequency band energy ratio exceeds a preset ratio threshold, the on-site image is determined to be a re-enhanced image. The above-mentioned edge detection operator, frequency band division method, feature extraction method, and threshold are merely examples and do not constitute a limitation on the embodiments of this application.

[0066] In this embodiment, standard environmental fingerprints are pre-collected when the inspection task is issued, and real-time environmental fingerprints are simultaneously collected during barcode scanning for similarity comparison. This allows inspection location verification to be based on the uniqueness and difficulty in replication of the environmental features surrounding the inspection point, effectively preventing remote virtual positioning and forged check-ins. Furthermore, the accuracy and robustness of location verification are improved through multimodal feature segmentation, weighted fusion, forged signal detection, and spatial topology verification.

[0067] The above describes an indoor inspection location verification method based on QR code positioning in the embodiments of this application. The following describes the indoor inspection location verification device based on QR code positioning that performs the above method.

[0068] See Figure 2 , Figure 2This is a schematic diagram of the structure of the indoor inspection location verification device based on QR code positioning provided in this application embodiment. The indoor inspection location verification device 60 based on QR code positioning can be integrated into an inspection management device, an edge computing device, or a server, or it can be deployed in an inspection terminal, or it can be implemented jointly by an inspection terminal and an inspection management device. The device 60 is used to execute the above-described indoor inspection location verification method based on QR code positioning.

[0069] In one embodiment, the device 60 includes an input / output module 601 and a processing module 602. The input / output module 601 and the processing module 602 can be software modules, hardware modules, or a combination of both. The two modules can be deployed in the same device or separately in the inspection terminal and the inspection management device. The input / output module 601 is used to acquire and output data such as standard environmental fingerprints, real-time environmental fingerprints, inspection point identifiers, and inspection records. The processing module 602 is used to decrypt the acquired data, calculate similarity, and determine location legitimacy.

[0070] The input / output module 601 is configured to acquire the standard environmental fingerprint of the inspection point corresponding to the inspection task, and in response to the inspection personnel's scanning operation of the inspection point, receive the real-time environmental fingerprint and the inspection point identifier corresponding to the scanning operation synchronously collected and encrypted by the inspection terminal.

[0071] The processing module 602 is configured to decrypt the real-time environmental fingerprint and calculate the similarity between the real-time environmental fingerprint and the standard environmental fingerprint; when the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0072] In some embodiments, the processing module 602 may also perform the processing corresponding to the above embodiments, such as extracting features from multimodal environmental feature data and vectorizing it to generate environmental feature fingerprints, parsing real-time environmental fingerprints to obtain multimodal real-time feature vectors, calculating feature distances or similarities for each modality as sub-environment labels and performing fluctuation variance analysis and abnormal media access control address cluster detection, dynamically adjusting the weights of each modality according to context information and performing weighted summation, and constructing a graph structure to perform spatial topology consistency verification, etc.

[0073] In some embodiments, the input / output module 601 is further configured to acquire pre-collected environmental feature fingerprints around the inspection point when the inspection task is issued, and to re-collect the environmental feature fingerprints around the inspection point when the environment around the inspection point changes; the processing module 602 is further configured to associate and store the environmental feature fingerprints with the inspection point identifier to generate the standard environmental fingerprint, and to update the standard environmental fingerprint corresponding to the inspection point when the environment around the inspection point changes.

[0074] In some embodiments, the input / output module 601 is further configured to collect multimodal environmental feature data around the inspection point. The multimodal environmental feature data includes at least one of the following: media access control address and received signal strength indication sequence of a wireless access point, Bluetooth low power beacon broadcast data, geomagnetic field triaxial data collected by a geomagnetic sensor, air pressure value collected by a barometer, and room impulse response characteristics of active acoustic detection. The processing module 602 is further configured to extract features from each modality of the multimodal environmental feature data to obtain feature sub-vectors corresponding to each modality, normalize the feature sub-vectors corresponding to each modality, and concatenate them in a preset order to form the environmental feature fingerprint. The environmental feature fingerprint is then associated with and stored with the inspection point identifier to establish a mapping relationship between the inspection point identifier and the environmental feature fingerprint.

[0075] In some implementations, the processing module 602 is further configured to parse the encoded information carried by the QR code corresponding to the scanning operation to obtain the inspection point identifier; and to verify the identity information of the inspection personnel and the address information reported by the inspection terminal based on the inspection point identifier.

[0076] In some implementations, the processing module 602 is further configured to parse the dynamic refresh parameters or timestamps in the encoded information, verify whether the dynamic refresh parameters or timestamps are within a preset validity period; if the verification is successful, query the identity information of the inspector from the pre-stored authorized inspector information database to determine whether the inspector is an authorized inspector; obtain the expected location range corresponding to the inspection point identifier, and determine whether the address information reported by the inspection terminal falls within the expected location range.

[0077] In some embodiments, the processing module 602 is further configured to decrypt the encrypted real-time environmental fingerprint using a decryption key corresponding to the encryption key used by the inspection terminal, to obtain a decrypted real-time environmental fingerprint; perform feature parsing on the decrypted real-time environmental fingerprint to obtain a multimodal real-time feature vector, wherein the multimodal real-time feature vector includes at least two of the following: wireless signal feature sub-vectors, geomagnetic feature sub-vectors, barometric pressure feature sub-vectors, and acoustic feature sub-vectors; simulate the current environment and determine the sub-environment labels corresponding to each modality based on the multimodal real-time feature vectors and the multimodal standard feature vectors corresponding to the standard environmental fingerprint; and calculate the comprehensive similarity based on the preset weights of each modality for the sub-environment labels corresponding to each modality.

[0078] In some embodiments, the processing module 602 is further configured to: extract real-time feature sub-vectors of each modality from the multimodal real-time feature vectors, and extract standard feature sub-vectors of each modality from the multimodal standard feature vectors; calculate the feature distance or similarity between the real-time feature sub-vectors and the standard feature sub-vectors, and use the feature distance or similarity as the initial sub-environment label corresponding to each modality; perform fluctuation variance analysis on the wireless signal strength sequence in the real-time environment fingerprint, and if the fluctuation variance is less than a preset variance threshold, determine that the first modality has a suspected spoofed signal and reduce the weight of the initial sub-environment label of the first modality; detect whether there is an abnormal media access control address cluster in the real-time environment fingerprint, and if so, determine that there is a suspected spoofed signal and reduce the weight of the initial sub-environment label of the corresponding second modality; summarize and determine the initial sub-environment label corresponding to each modality as the sub-environment label corresponding to each modality.

[0079] In some implementations, the processing module 602 is further configured to acquire at least one of the following contextual information: current time period, regional importance, and historical compliance rate; dynamically adjust the preset weights corresponding to each modality based on the contextual information to obtain dynamic weights; perform weighted summation on the sub-environment labels corresponding to each modality based on the dynamic weights corresponding to each modality to obtain an initial comprehensive similarity; construct a graph structure for the inspection area, where nodes represent inspection points and edges represent the spatial distance between inspection points; verify the spatial topology consistency through the graph structure based on the real-time environmental fingerprint and the standard environmental fingerprints of adjacent inspection points to obtain a topology consistency result; and determine the comprehensive similarity based on the topology consistency result.

[0080] In some embodiments, the processing module 602 is further configured to mark the inspection record corresponding to the inspection point as a suspicious inspection record or refuse the submission of the inspection record when the similarity does not exceed the preset similarity threshold; and to allow the submission of the inspection record corresponding to the inspection point when the similarity exceeds the preset similarity threshold; the input / output module 601 is further configured to return a prompt message for re-collection or relocation to the inspection terminal.

[0081] In some implementations, the input / output module 601 is further configured to send an image acquisition command to the inspection terminal after determining that the inspection location of the inspection personnel is legal. The image acquisition command is used to instruct the inspection personnel to take a picture of the inspection point and receive the on-site image uploaded by the inspection terminal. The processing module 602 is further configured to use a moiré pattern recognition algorithm and edge spectral analysis to determine whether the on-site image is a copy of an electronic screen. If it is determined to be a copy, the corresponding inspection record is marked as invalid and a prompt message for re-inspection is returned to the inspection terminal. If it is determined not to be a copy, the corresponding inspection record is marked as valid and submitted.

[0082] The above-described processing procedure of the device 60 can be implemented with reference to steps 101 to 104 and their optional embodiments in the foregoing method embodiments.

[0083] In this embodiment, the input / output module 601 acquires data such as standard environmental fingerprints, real-time environmental fingerprints, and inspection point identifiers, while the processing module 602 performs decryption, similarity calculation, and location legitimacy determination. Through data transfer between the input / output module 601 and the processing module 602, the barcode scanning location and environmental fingerprint verification are correlated, providing support for accurate verification of indoor inspection locations and anti-cheating measures.

[0084] The indoor inspection location verification device 60 based on QR code positioning in this application embodiment has been described above from the perspective of modular functional entities. The indoor inspection location verification device based on QR code positioning in this application embodiment will be described below from the perspective of hardware processing.

[0085] It should be noted that, Figure 2 The physical device corresponding to the input / output module 601 shown can be a transceiver, radio frequency circuit, communication module, and input / output (I / O) interface, etc., and the physical device corresponding to the processing module 602 can be a processor.

[0086] Figure 2 The devices shown can all have the following characteristics: Figure 3 The structure shown, when Figure 2 The indoor inspection location verification device 60 based on QR code positioning shown has the following features: Figure 3When the structure shown is used, Figure 3 The processor and transceiver in the device can perform the same or similar functions as the processing module 602 and input / output module 601 provided in the aforementioned device embodiments. Figure 3 The memory stores the computer programs that the processor needs to call when executing the above-mentioned indoor inspection location verification method based on barcode scanning.

[0087] See Figure 4 , Figure 4 This is a schematic diagram of a patrol terminal in one embodiment of this application. The patrol terminal may include a processor 1080, a radio frequency circuit 1010, a memory 1020, an input unit 1030, a display unit 1040, a sensor 1050, an audio circuit 1060, a wireless fidelity module 1070, a Bluetooth module 1081, and a power supply 1090. The input unit 1030 can receive input from patrol personnel regarding patrol tasks and inspection results. The display unit 1040 can display the patrol task, target patrol point, and location verification status. The wireless fidelity module 1070 can scan visible wireless access points around the inspection point to obtain the media access control address and received signal strength indication sequence of each wireless access point; the Bluetooth module 1081 can receive Bluetooth Low Energy beacon broadcast data around the inspection point; the sensor 1050 may include a geomagnetic sensor, a barometric pressure sensor, and a camera component, wherein the geomagnetic sensor is used to collect triaxial data of the geomagnetic field around the inspection point, the barometric pressure sensor is used to collect the barometric pressure value around the inspection point, and the camera component is used to scan the QR code posted at the inspection point and capture images of the inspection point; the audio circuit 1060 may include a speaker and a microphone, the speaker is used to emit detection sound waves, and the microphone is used to receive reflected sound waves to obtain room impulse response characteristics. The radio frequency circuit 1010 or the wireless fidelity module 1070 can send the real-time environmental fingerprint and the inspection point identifier corresponding to the scanning operation, which are synchronously collected and encrypted and uploaded by the inspection terminal, to the server.

[0088] This application also provides a server; please refer to [link / reference]. Figure 5 , Figure 5This is a schematic diagram of a server structure provided in an embodiment of this application. The server 1100 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 1122 (e.g., one or more processors) and memory 1132, and one or more storage media 1130 (e.g., one or more mass storage devices) for storing application programs 1142 or data 1144. The memory 1132 and storage media 1130 can be temporary or persistent storage. The program stored in the storage media 1130 may include one or more modules (not shown in the figure), each module may include a series of instruction operations on the server. Furthermore, the CPU 1122 may be configured to communicate with the storage media 1130 and execute the series of instruction operations in the storage media 1130 on the server 1100. Server 1100 may also include one or more power supplies 1126, one or more wired or wireless network interfaces 1150, one or more input / output interfaces 1158, and / or one or more operating systems 1141, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.

[0089] The steps performed by the server in the above embodiments can be based on this Figure 5 The structure of server 1100 is shown. For example, in the above embodiment, it consists of... Figure 2 The steps performed by the indoor inspection location verification device 60 based on QR code positioning shown can be based on this Figure 5 The server structure is shown. For example, the central processing unit 1122 performs the following operations by calling instructions from memory 1132: The central processing unit 1122 can acquire the standard environmental fingerprint of the inspection point corresponding to the inspection task through the input / output interface 1158. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that is pre-collected when the inspection task is issued. In response to the inspection personnel's scanning operation of the inspection point, the central processing unit 1122 can acquire the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint that is synchronously collected and encrypted and uploaded by the inspection terminal. The central processing unit 1122 can also decrypt the real-time environmental fingerprint and calculate the similarity between the real-time environmental fingerprint and the standard environmental fingerprint; when the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

[0090] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0091] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0092] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or modules, and may be electrical, mechanical, or other forms.

[0093] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0095] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0096] The computer program product includes one or more computer instructions. When the computer program is loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state disk (SSD)).

[0097] The technical solutions provided in the embodiments of this application have been described in detail above. Specific examples have been used in the embodiments of this application to illustrate the principles and implementation methods of the embodiments of this application. The description of the above embodiments is only for the purpose of helping to understand the methods and core ideas of the embodiments of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments of this application. Therefore, the content of this specification should not be construed as a limitation on the embodiments of this application.

Claims

1. A method for verifying indoor inspection location based on QR code scanning, characterized in that, The method includes: Obtain the standard environmental fingerprint of the inspection point corresponding to the inspection task. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that was collected in advance when the inspection task was issued. In response to the scanning operation of the inspection personnel at the inspection point, the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint synchronously collected and encrypted by the inspection terminal are obtained; The real-time environment fingerprint is decrypted, and the similarity between the real-time environment fingerprint and the standard environment fingerprint is calculated. When the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.

2. The method according to claim 1, characterized in that, The acquisition of the standard environmental fingerprint of the inspection point corresponding to the inspection task includes: When the inspection task is issued, the environmental feature fingerprints around the inspection point are acquired in advance, and the environmental feature fingerprints are associated with the inspection point identifier and stored to generate the standard environmental fingerprint. When the environment around the inspection point changes, the environmental feature fingerprints around the inspection point are re-collected, and the standard environmental fingerprints corresponding to the inspection point are updated.

3. The method according to claim 2, characterized in that, The step of acquiring pre-collected environmental feature fingerprints around the inspection point, associating and storing the environmental feature fingerprints with the inspection point identifier, and generating the standard environmental fingerprint includes: Collect multimodal environmental feature data around the inspection point. The multimodal environmental feature data includes at least one of the following: media access control address and received signal strength indication sequence of wireless access point, Bluetooth low power beacon broadcast data, geomagnetic field triaxial data collected by geomagnetic sensor, air pressure value collected by barometric pressure sensor, and room impulse response characteristics of active acoustic detection. Feature extraction is performed on each modal data in the multimodal environmental feature data to obtain the feature sub-vectors corresponding to each mode; wherein, the mean and variance features of signal strength are extracted from the received signal strength indication sequence and the Bluetooth low power beacon broadcast data, the triaxial components and magnetic field magnitude features are extracted from the geomagnetic field triaxial data, the mean and fluctuation features of air pressure are extracted from the air pressure value, and the frequency band energy features are extracted from the room impulse response features; The feature subvectors corresponding to each modality are normalized and then concatenated in a preset order to form the environmental feature fingerprint. The environmental feature fingerprint is associated with the inspection point identifier and stored, and a mapping relationship between the inspection point identifier and the environmental feature fingerprint is established.

4. The method according to claim 1, characterized in that, The acquisition of the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint synchronously collected and encrypted by the inspection terminal includes: The inspection point identifier is obtained by parsing the encoded information carried in the QR code corresponding to the scanning operation; wherein, the encoded information is generated by the inspection management device using preset encoding rules, and other software cannot parse and recognize the encoded information; The identity information of the inspection personnel and the address information reported by the inspection terminal are verified based on the inspection point identification.

5. The method according to claim 4, characterized in that, The step of verifying the identity information of the inspection personnel and the address information reported by the inspection terminal based on the inspection point identifier includes: Parse the dynamic refresh parameters or timestamps in the encoded information to verify whether the dynamic refresh parameters or timestamps are within the preset validity period; If the verification is successful, the identity information of the inspector will be retrieved from the pre-stored authorized inspector information database to determine whether the inspector is an authorized inspector. Obtain the expected location range corresponding to the inspection point identifier, and determine whether the address information reported by the inspection terminal falls within the expected location range; If the inspection personnel are authorized and the address information falls within the expected location range, the verification passes.

6. The method according to claim 1, characterized in that, The step of decrypting the real-time environment fingerprint and calculating the similarity between the real-time environment fingerprint and the standard environment fingerprint includes: The encrypted real-time environmental fingerprint is decrypted using the decryption key corresponding to the encryption of the inspection terminal to obtain the decrypted real-time environmental fingerprint. The decrypted real-time environmental fingerprint is subjected to feature parsing to obtain a multimodal real-time feature vector, wherein the multimodal real-time feature vector includes at least two of the following: wireless signal feature sub-vector, geomagnetic feature sub-vector, barometric pressure feature sub-vector, and acoustic feature sub-vector. Based on the multimodal real-time feature vector and the multimodal standard feature vector corresponding to the standard environment fingerprint, the current environment is simulated and the sub-environment label corresponding to each modality is determined; The sub-environment labels corresponding to each modality are calculated based on the preset weights of each modality to obtain the comprehensive similarity between the real-time environment fingerprint and the standard environment fingerprint.

7. The method according to claim 6, characterized in that, The step of simulating the current environment and determining the sub-environment labels corresponding to each modality based on the multimodal real-time feature vector and the multimodal standard feature vector corresponding to the standard environment fingerprint includes: For each modality, real-time feature sub-vectors of each modality are extracted from the multimodal real-time feature vectors, and standard feature sub-vectors of each modality are extracted from the multimodal standard feature vectors. Calculate the feature distance or similarity between the real-time feature sub-vector and the standard feature sub-vector, and use the feature distance or similarity as the initial sub-environment label corresponding to each modality; A fluctuation variance analysis is performed on the wireless signal strength sequence in the real-time environmental fingerprint. If the fluctuation variance is less than a preset variance threshold, it is determined that there is a suspected forged signal in the first mode, and the weight of the initial sub-environment label of the first mode is reduced. Detect whether there is an abnormal media access control address cluster in the real-time environment fingerprint; if so, determine that there is a suspected forgery signal and reduce the weight of the corresponding second modality's initial sub-environment label; The initial sub-environment labels corresponding to each modality are summarized and determined as the sub-environment labels corresponding to each modality.

8. The method according to claim 7, characterized in that, The calculation of the sub-environment labels corresponding to each modality based on preset weights for each modality yields the comprehensive similarity between the real-time environment fingerprint and the standard environment fingerprint, including: Obtain at least one piece of contextual information from the current time period, regional importance, and historical compliance rate, and dynamically adjust the preset weights corresponding to each modality based on the contextual information to obtain dynamic weights; Based on the dynamic weights corresponding to each modality, the sub-environment labels corresponding to each modality are weighted and summed to obtain the initial comprehensive similarity. Construct a graph structure for the inspection area, where nodes represent inspection points and edges represent spatial distances between inspection points; based on the real-time environmental fingerprint and the standard environmental fingerprints of adjacent inspection points, verify the spatial topological consistency between the real-time environmental fingerprint and the adjacent inspection points through the graph structure to obtain the topological consistency result. If the topology consistency result indicates that the verification is consistent, then the initial comprehensive similarity is used as the comprehensive similarity; if the topology consistency result indicates that the verification is inconsistent, then the initial comprehensive similarity is reduced to obtain the comprehensive similarity.

9. The method according to claim 1, characterized in that, After determining that the inspection personnel's inspection location is legal, the process also includes: Send an image acquisition command to the inspection terminal, the image acquisition command being used to instruct the inspection personnel to take images of the inspection point; The system receives on-site images uploaded by the inspection terminal and uses moiré pattern recognition algorithm and edge spectral analysis to determine whether the on-site images are obtained by photographing an electronic screen. If it is determined that the photo was copied, the corresponding inspection record will be marked as invalid, and a prompt message for re-inspection will be returned to the inspection terminal. If the record is determined not to be a copy, the corresponding inspection record will be marked as valid and submitted.

10. An indoor inspection location verification device based on QR code positioning, characterized in that, The device includes: The input / output module is configured to acquire the standard environmental fingerprint of the inspection point corresponding to the inspection task. The standard environmental fingerprint is the environmental feature fingerprint of the area around the inspection point that is pre-collected when the inspection task is issued. In response to the inspection personnel's scanning operation of the inspection point, the module acquires the inspection point identifier corresponding to the scanning operation and the real-time environmental fingerprint that is synchronously collected and encrypted by the inspection terminal. The processing module is configured to decrypt the real-time environmental fingerprint and calculate the similarity between the real-time environmental fingerprint and the standard environmental fingerprint; when the similarity exceeds a preset similarity threshold, the inspection location of the inspection personnel is determined to be legal; when the similarity does not exceed the preset similarity threshold, the inspection location of the inspection personnel is determined to be illegal.