A method and system for processing data of non-sensitive borrowing in smart libraries
By designing sensorless borrowing data processing methods and systems in the library, using multi-layer identity verification and risk assessment, the problem of insufficient identity verification and security monitoring in traditional sensorless borrowing technology is solved, and an efficient and secure borrowing process is achieved. Intelligent early warning response improves user experience and management efficiency.
Patent Information
- Application Number
- CN202411112071.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-08-14
AI Technical Summary
Traditional touchless borrowing technology is weak in identity verification and security monitoring, and lacks early warning and processing mechanisms for low-credit users, resulting in the loss or damage of books, which in turn affects its security and efficiency.
Design a smart library's sensorless borrowing data processing method and system, and obtain and classify user appointment data, conduct information verification and multi-layer identity verification, combine historical borrowing data for risk assessment, generate low-credit user warnings, and use intelligent warnings Response and motion sensing monitoring to realize automated borrowing and return process control.
It improves the automation level of the borrowing process, ensures the safety and efficiency of the process, reduces manual intervention, promptly warns and deals with abnormal situations, ensures the safety of library resources, and improves user experience.
Smart Images

Figure CN118656581B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent lending technology, and in particular to a method and system for processing senseless lending data in a smart library. Background Art
[0002] At present, libraries use traditional manual borrowing and returning methods. With the advancement of information technology, barcode and RFID (radio frequency identification) technologies have been introduced to realize semi-automatic borrowing and returning services. Although this method improves efficiency, users still need to manually scan barcodes or RFID tags.
[0003] With the popularization of mobile Internet, smart libraries have begun to use mobile applications and cloud computing technology to realize book borrowing and returning operations by scanning QR codes with mobile phones. At this stage, the contactless borrowing technology has taken shape, but users still need to actively scan the code.
[0004] In order to further improve the user experience, the contactless borrowing technology has gradually developed to a higher level, using sensor networks and big data analysis to achieve true contactless borrowing. In this process, the application of AI (artificial intelligence) and machine learning technology plays a key role. Through smart sensors and cameras, the system can automatically identify users and books, and realize contactless borrowing operations.
[0005] In addition, advances in image recognition and deep learning algorithms have enabled the system to more accurately identify and record users’ borrowing behavior. However, traditional contactless borrowing currently relies on manual operations or semi-automated systems, with slow processing speeds. It is also relatively weak in identity authentication and security monitoring, and lacks early warning and processing mechanisms for low-credit users, which can easily cause books to be lost or damaged, leading to lower security and efficiency of contactless borrowing. Summary of the invention
[0006] Based on this, it is necessary to provide a method and system for processing seamless borrowing data in a smart library to solve at least one of the above technical problems.
[0007] To achieve the above-mentioned object, the present invention discloses a method and system for processing data of a smart library without sensing borrowing, wherein a method for processing data of a smart library without sensing borrowing acts on a channel gate of a smart library, wherein the channel gate of the smart library includes a first gate of an entrance channel, a first gate of an exit channel, a second gate of an exit channel, and a third gate of an exit channel, and the method includes the following steps:
[0008] Step S1: Obtain user library reservation data; classify the user library reservation data by reservation type, generate user borrowing reservation data and user return reservation data; associate information with the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate a smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data;
[0009] Step S2: obtaining user borrowing book information data; controlling the first gate of the exit channel according to the user borrowing book information data, and generating control data of the first gate of the exit channel; performing multi-layer identity authentication of the user of the second gate of the exit channel on the control data of the first gate of the exit channel through the smart library channel gate information verification instruction, and generating borrowing identity authentication result data; performing supervision and control of the borrowing process of the exit channel according to the borrowing identity authentication result data, and generating non-sensing control data of the borrowing process;
[0010] Step S3: historical data collection is performed on user borrowing book information data to obtain historical user borrowing book information data; borrowing risk assessment is performed on the user borrowing book information data according to the historical user borrowing book information data to generate user borrowing risk assessment data; gate warning processing is performed on the borrowing process senseless control data and the return process senseless control data based on the user borrowing risk assessment data to generate low-credit user warning data;
[0011] Step S4: Perform intelligent warning response on the smart library channel gate according to the low-credit user warning data to generate intelligent warning response data; perform motion sensor trigger monitoring on the smart library channel gate according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data to execute the smart library contactless borrowing operation.
[0012] Preferably, step S1 comprises the following steps:
[0013] Step S11: Obtaining user library reservation data;
[0014] Step S12: preprocessing the user library reservation data to generate standard user library reservation data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization;
[0015] Step S13: classify the standard user library reservation data into reservation types, generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction, wherein the smart library channel gate information verification instruction includes the entry information verification instruction and the exit information verification instruction;
[0016] Step S14: perform return information retrieval on the user's return reservation data through the door entry information verification instruction to generate return information retrieval data; perform book return process supervision and control on the first gate of the door entry channel based on the return retrieval data to generate book return process non-contact control data.
[0017] Preferably, step S14 includes the following steps:
[0018] Step S141: performing return information retrieval on the user return reservation data through the door entry information verification instruction to generate return information retrieval data;
[0019] Step S142: Performing a user identity inventory trigger based on the first set of opposite light curtains on the first gate of the entrance passage, and obtaining the trigger inventory data of the first opposite light curtain at the entrance;
[0020] Step S143: Map the inventory data triggered by the first pair of light curtains at the entrance to the display screen on the first gate of the entrance channel to visualize the returned book information, and control the gate of the first gate of the entrance channel according to the second set of pair of light curtains on the first gate of the entrance channel to generate the senseless control data of the book return process.
[0021] Preferably, step S2 comprises the following steps:
[0022] Step S21: Obtaining user borrowed book information data;
[0023] Step S22: Triggering the first gate of the exit passage to count the identity information of the exit user according to the book borrowing information data of the user, and generating the triggering count data of the first light curtain at the exit;
[0024] Step S23: controlling the gate based on the first gate of the exit passage by triggering the inventory data through the first pair of light curtains at the exit, and generating control data of the first gate at the exit;
[0025] Step S24: Perform multi-layer user identity authentication on the second light curtain on the second gate at the exit through the control data of the first gate at the exit, and generate borrowing identity authentication result data;
[0026] Step S25: Verify and judge the borrowing identity verification result data. When the borrowing identity verification result data is unsuccessful, the third gate of the exit channel is supervised and controlled to generate the third gate control data; when the borrowing identity verification result data is successful, the second gate of the exit channel is supervised and controlled to generate the second gate control data; the first exit gate control data, the second exit gate control data and the third exit gate control data are integrated to generate the borrowing process seamless control data.
[0027] Preferably, step S24 includes the following steps:
[0028] Step S241: dividing the identity access mode through the exit information verification instruction to generate the identity access mode, wherein the identity access mode includes the IC card access mode, the ID card access mode and the user face access mode;
[0029] Step S242: when it is confirmed that the identity access mode is the IC card access mode or the ID card access mode, the corresponding data based on the IC card access mode or the ID card access mode is scanned to generate identity information scan data; the identity information scan data is matched with the standard user library reservation data to generate borrowing identity verification result data;
[0030] Step S243: When the identity access mode is confirmed to be the user face access mode, the face recognition camera on the second gate of the entrance channel is used to virtually read the user access face image to obtain the user reservation access face image; the user reservation access face image and the face image in the standard user library reservation data are aligned with the face feature points to generate face access feature alignment data; the matching degree of the face access feature alignment data is calculated to generate the access face matching degree;
[0031] Step S244: Compare the access face matching degree with the preset face matching threshold. When the access face matching degree is greater than or equal to the preset face matching threshold, generate identity information trusted data according to the corresponding user scheduled access face image; when the access face matching degree is less than the preset face matching threshold, mark the corresponding user scheduled access face image as a suspicious face image;
[0032] Step S245: Perform user liveness verification detection on the suspicious facial image to generate suspicious user liveness verification detection data; perform deep information investigation on the suspicious facial image through the suspicious user liveness verification detection data to generate identity information deep verification data; integrate the identity information trusted data and the identity information deep verification data to also generate borrowing identity verification result data.
[0033] Preferably, performing user liveness verification detection on a suspicious face image includes:
[0034] Perform facial texture analysis on suspicious face images to generate user facial texture feature data;
[0035] The user's light reflection is detected through the suspicious face image, and the reflection data is collected through the optical sensor to obtain the facial spectral reflection data; the facial spectral reflection data is analyzed for reflection characteristics to generate the user's facial light reflection characteristic data;
[0036] The facial three-dimensional structure information of the suspicious face image is extracted by the structured light sensor to obtain the facial three-dimensional structure information data; the facial three-dimensional structure information data is analyzed for facial depth features to generate the user's facial depth feature data;
[0037] Perform dynamic facial behavior recognition on the user corresponding to the suspicious face image to generate user dynamic behavior recognition data; perform comprehensive user liveness verification detection based on user facial texture feature data, user facial light reflection feature data, user facial depth feature data and user dynamic behavior recognition data to generate suspicious user liveness verification detection data.
[0038] Preferably, step S3 comprises the following steps:
[0039] Step S31: historical data collection of user borrowing book information data to obtain historical user borrowing book information data;
[0040] Step S32: performing a borrowing risk assessment on the user's borrowing book information data according to the historical user's borrowing book information data to generate user borrowing risk assessment data;
[0041] Step S33: classifying the user's borrowed book information data into low-credit users based on the user's borrowing risk assessment data, and generating low-credit user grade data;
[0042] Step S34: Perform gate warning processing on the borrowing process contactless control data and the returning process contactless control data through the low-credit user level data to generate low-credit user warning data.
[0043] Preferably, step S32 includes the following steps:
[0044] Step S321: dividing the historical user borrowing book information data into a data set to obtain a model training set and a model test set; training the model training set by a logistic regression algorithm to generate a borrowing risk assessment pre-model;
[0045] Step S322: using the model test set to iterate the model test on the borrowing risk assessment pre-model, thereby generating a borrowing risk assessment model;
[0046] Step S323: Importing the user's borrowing book information data into the borrowing risk assessment model to perform borrowing risk assessment and generate user borrowing risk assessment data.
[0047] Preferably, step S4 comprises the following steps:
[0048] Step S41: Performing intelligent early warning response on the smart library channel gate according to the low-credit user early warning data to generate intelligent early warning response data;
[0049] Step S42: encapsulating the intelligent warning response data, the book return process senseless control data and the borrowing process senseless control data into functional modules to generate a smart library senseless control module;
[0050] Step S43: deploy motion sensors on the smart library channel gate based on the smart library non-sensing control module to obtain a non-sensing borrowing motion sensor network; monitor the module function triggering of the smart library non-sensing control module through the non-sensing borrowing motion sensor network to execute the smart library non-sensing borrowing operation.
[0051] Furthermore, the present invention also discloses a smart library non-sensing borrowing data processing system, which is used to execute the above-mentioned smart library non-sensing borrowing data processing method. The smart library non-sensing borrowing data processing system includes:
[0052] The library entry and book return module is used to obtain user library reservation data; classify the user library reservation data by reservation type, and generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data;
[0053] The library outbound borrowing module is used to obtain the user's borrowing book information data; control the first gate of the outbound channel according to the user's borrowing book information data, and generate the first gate control data; perform multi-layer identity authentication of the second gate user on the first gate control data through the smart library channel gate information verification instruction, and generate the borrowing identity authentication result data; according to the borrowing identity authentication result data; according to the borrowing identity authentication result data, supervise and control the outbound channel borrowing process, and generate the borrowing process non-sensing control data;
[0054] The risk warning module is used to collect historical data of user borrowing book information data to obtain historical user borrowing book information data; conduct borrowing risk assessment on user borrowing book information data based on historical user borrowing book information data to generate user borrowing risk assessment data; conduct gate warning processing on borrowing process non-sensing control data and return process non-sensing control data based on user borrowing risk assessment data to generate low-credit user warning data;
[0055] The motion trigger module is used to perform intelligent warning response to the smart library channel gate according to the low-credit user warning data and generate intelligent warning response data; according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data, the smart library channel gate is motion-sensor-triggered and monitored to execute the smart library contactless borrowing operation.
[0056] The beneficial effects of the present invention are:
[0057] The present invention classifies the reservation types of user library reservation data into two types: borrowing and returning, which helps to clarify different types of operation requirements and generate information verification instructions for the smart library channel gate machine, which can accurately control user entry and exit, ensure the correctness and safety of the process, generate senseless control data for the book return process, improve the automation level of library operations, and reduce manual intervention.
[0058] Secondly, by controlling the first gate at the exit, the user's borrowing process is ensured to proceed smoothly, improving the user experience. The multi-layer identity verification of the user at the second gate ensures the accuracy of the borrowing identity, effectively preventing identity impersonation and fraud. The generated borrowing process senseless control data makes the borrowing process more efficient and seamless, reducing the complexity in operation. Through historical data collection and risk assessment, scientific analysis of user borrowing behavior helps identify potential risks.
[0059] Furthermore, by generating early warning data for low-credit users based on risk assessment results, it is helpful to take timely measures to prevent potential borrowing default risks. By responding to intelligent early warnings for low-credit users, abnormal situations can be handled in real time to ensure the normal operation of the library. Motion-sensing trigger monitoring of the smart library channel gate improves the automation of borrowing operations, reduces manual intervention, and reduces the complexity of user operations through the execution of non-sensing borrowing operations, improving the overall borrowing experience.
[0060] The present invention significantly improves the management efficiency of the library through the automation and intelligent design of each step, reducing manual intervention and errors. Through accurate identity verification and risk warning, the library can effectively ensure the security of resources and reduce unnecessary losses. The automated borrowing and returning process, intelligent warning response and risk management measures enhance the overall service experience of users.
[0061] Therefore, the present invention improves the efficiency, security and user experience of seamless borrowing through automated process control, multi-layer identity authentication, risk assessment and intelligent early warning. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] Figure 1 A schematic diagram of the steps of a method for processing data of seamless borrowing in a smart library;
[0063] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;
[0064] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.
[0065] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG. DETAILED DESCRIPTION
[0066] The technical method of the present invention is clearly and completely described below in conjunction with the accompanying drawings.
[0067] Example 1, a method for processing data of non-sensitive borrowing in a smart library, see Figures 1 to 4 , the method comprises the following steps:
[0068] Step S1: Obtain user library reservation data; classify the user library reservation data by reservation type, generate user borrowing reservation data and user return reservation data; associate information with the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate a smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data;
[0069] Step S2: obtaining user borrowing book information data; controlling the first gate of the exit channel according to the user borrowing book information data, and generating control data of the first gate of the exit channel; performing multi-layer identity authentication of the user of the second gate of the exit channel on the control data of the first gate of the exit channel through the smart library channel gate information verification instruction, and generating borrowing identity authentication result data; performing supervision and control of the borrowing process of the exit channel according to the borrowing identity authentication result data, and generating non-sensing control data of the borrowing process;
[0070] Step S3: historical data collection is performed on user borrowing book information data to obtain historical user borrowing book information data; borrowing risk assessment is performed on the user borrowing book information data according to the historical user borrowing book information data to generate user borrowing risk assessment data; gate warning processing is performed on the borrowing process senseless control data and the return process senseless control data based on the user borrowing risk assessment data to generate low-credit user warning data;
[0071] Step S4: Perform intelligent warning response on the smart library channel gate according to the low-credit user warning data to generate intelligent warning response data; perform motion sensor trigger monitoring on the smart library channel gate according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data to execute the smart library contactless borrowing operation.
[0072] Through the implementation of the above steps, user reservation data can be obtained and classified. The system can accurately distinguish between borrowing and returning reservations and generate information verification instructions, thereby achieving effective supervision and control of the entrance channel. This improves the management efficiency of the library, ensures seamless control of the return process, reduces manual intervention, and improves the fluency and automation level of the borrowing and returning process.
[0073] After acquiring and processing the user's borrowing book information data, the system can automatically control the first gate of the exit channel and ensure the authenticity of the borrowing identity through multi-layer identity authentication. This improves the security of the borrowing process and reduces the need for users to actively operate, realizing seamless borrowing control and improving user experience.
[0074] By conducting risk assessment on historical borrowing data, potential low-credit users can be predicted and identified, and early warning data can be generated, which helps prevent the loss or damage of books, reduce the operational risks of the library, and improve risk management strategies through timely early warning processing. Based on the early warning data of low-credit users, the system conducts intelligent early warning responses, and combines non-contact control data and motion sensor monitoring to enhance the comprehensive monitoring capabilities of the borrowing and return processes, which ensures the smooth execution of non-contact borrowing operations and enhances the ability to respond to abnormal behaviors in a timely manner, thereby improving overall security and operational efficiency.
[0075] Specifically, Figure 1 The figure is a flowchart of the steps of a method for processing data of a smart library without feeling borrowing in Example 1. In this example, the method for processing data of a smart library without feeling borrowing in Example 1 includes the following steps:
[0076] Step S1: Obtain user library reservation data; classify the user library reservation data by reservation type, generate user borrowing reservation data and user return reservation data; associate information with the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate a smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data;
[0077] In the embodiment of the present invention, the user's reservation data is obtained through the library management system. The reservation data includes the user's borrowing reservation and return reservation information. The reservation data should include the user ID, reservation type (borrowing or return), reservation time, reservation book information, etc. The reservation data is classified according to the reservation type to generate user borrowing reservation data and user return reservation data.
[0078] According to the reservation type field, the data is divided into two categories: borrowing reservation data and return reservation data. According to the user's borrowing reservation data and return reservation data, the corresponding channel gate information verification instruction is generated. The verification instruction should contain the user ID, reservation type, reservation time, reservation book information, and channel information that needs to be verified. When the user enters the library according to the return reservation, the information is verified through the channel gate. If the verification is successful, the user is allowed to enter and put the book in the return place; if the verification fails, the user is prevented from entering and an error message is prompted.
[0079] The above steps use technologies such as RFID and QR code scanning to achieve seamless control of the process of users returning books at the door. Users do not need to perform additional operations, and the seamless control data of the book return process is stored in the smart library management system for subsequent analysis and improvement. For example, the system obtains user A's reservation information from the database: the reservation type is "return", the reserved book is "Book Title", and the reservation time is 14:00 on August 1, 2024, and classifies the reservation information as user return reservation data. The system generates a channel gate information verification instruction, including user A's ID, reservation type "return", reservation time, and reserved book information.
[0080] User A arrives at the library at the scheduled time and passes through the gate for verification. The gate scans user A's identity information (such as RFID card or QR code) and matches it with the reservation information in the system. After verification, user A is allowed to enter the library to return the book. The system records user A's book return process, including the time of entry, verification results, etc., and these data are stored in the library management system for subsequent use.
[0081] Step S2: obtaining user borrowing book information data; controlling the first gate of the exit channel according to the user borrowing book information data, and generating control data of the first gate of the exit channel; performing multi-layer identity authentication of the user of the second gate of the exit channel on the control data of the first gate of the exit channel through the smart library channel gate information verification instruction, and generating borrowing identity authentication result data; performing supervision and control of the borrowing process of the exit channel according to the borrowing identity authentication result data, and generating non-sensing control data of the borrowing process;
[0082] In the embodiment of the present invention, the user's borrowing book information data is obtained through the library management system. The borrowing data includes user ID, borrowing book information, borrowing time, etc. The borrowing data should include user ID, book ID, borrowing date, due date, etc. The control instruction of the first gate of the exit channel is generated according to the user's borrowing book information data. The control instruction should include user ID, borrowing book information, borrowing time, and channel information to be controlled.
[0083] After the user borrows a book, the first gate is used for information verification and control to ensure that the user's borrowing information is valid, and the control data of the user passing through the first gate is recorded, including the verification time, verification results, etc. The control data of the first gate at the exit is stored in the smart library management system for subsequent analysis and control. The smart library channel gate information verification instruction is used to perform multi-layer identity authentication on the control data of the first gate at the exit. The first layer of verification uses an RFID card or a QR code to scan the user's identity information, and the second layer of verification uses biometric technologies such as facial recognition or fingerprint recognition for further verification. The borrowing identity authentication result data is generated based on the results of the multi-layer identity authentication.
[0084] The borrowing process at the exit channel is supervised and controlled based on the borrowing identity verification result data. If the verification is passed, the user is allowed to leave the library through the second gate. If the verification fails, the user is prevented from passing and a corresponding error message is prompted. The user's borrowing process is seamlessly controlled through automation technology without the need for additional user operations. The user's borrowing process at the exit channel is recorded, including the time and results of multi-layer identity verification, and the borrowing process seamless control data is stored in the smart library management system for subsequent use.
[0085] Step S3: historical data collection is performed on user borrowing book information data to obtain historical user borrowing book information data; borrowing risk assessment is performed on the user borrowing book information data according to the historical user borrowing book information data to generate user borrowing risk assessment data; gate warning processing is performed on the borrowing process senseless control data and the return process senseless control data based on the user borrowing risk assessment data to generate low-credit user warning data;
[0086] In the embodiment of the present invention, historical user borrowing book information data is collected from the library management system. The historical data should include all users' borrowing records, return records, overdue records, etc. The historical data should contain information such as user ID, book ID, borrowing date, due date, actual return date, overdue days, etc.
[0087] Based on historical user borrowing information data, analyze user borrowing behavior and identify high-risk borrowing behavior. Establish a set of evaluation indicators, including the number of overdue returns, the number of overdue days, the borrowing frequency, etc. Calculate each user's borrowing risk score based on the evaluation indicators and generate user borrowing risk assessment data. The specific steps are as follows: count the number of overdue returns for each user, calculate the total number of overdue days for each user, count the number of borrowings by users in a specific time period, calculate the comprehensive risk score based on the above indicators, and generate user borrowing risk assessment data.
[0088] Set warning rules based on user borrowing risk assessment data. If the risk score exceeds a certain threshold, the user is marked as a low-credit user. When a user borrows or returns a book, the system will issue a warning to the gate based on its risk score. When a user goes through the borrowing process, if he is marked as a low-credit user, the gate system will issue a warning signal to prompt the administrator or prevent him from borrowing. When a user goes through the return process, if he is marked as a low-credit user, the gate system will record his return behavior and issue a warning signal. Record the warning information of low-credit users in the borrowing and return processes, including user ID, warning time, warning reason, etc. Store the low-credit user warning data in the smart library management system for subsequent management and improvement.
[0089] Step S4: Perform intelligent warning response on the smart library channel gate according to the low-credit user warning data to generate intelligent warning response data; perform motion sensor trigger monitoring on the smart library channel gate according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data to execute the smart library contactless borrowing operation.
[0090] In the embodiment of the present invention, the latest warning data including user ID, warning time, warning reason, etc. are obtained from the low-credit user warning database. The corresponding warning response rules are set according to the risk level of the low-credit user. For example, high-risk users will be required to perform additional verification, and medium-risk users will be prompted to pay attention to their borrowing behavior.
[0091] According to the warning response rules, the system generates intelligent warning response data, including the actions that need to be taken (such as additional verification, prompt information, etc.). Deploy motion sensors at the gates of the smart library to monitor the user's entry and exit behavior. Set the trigger conditions of the sensor, for example, start the monitoring program when the user approaches the gate. Use motion sensors to collect user motion data in real time, such as entry and exit time, movement trajectory, etc. The system integrates the intelligent warning response data, the book return process non-sensing control data, and the borrowing process non-sensing control data to form a unified control instruction.
[0092] According to the integrated control instructions, the system intelligently controls the gate, including automatic release, additional verification or denial of passage, etc. RFID, QR code scanning, biometrics and other technologies are used to achieve user-free operation and reduce the need for manual operation.
[0093] Preferably, step S1 comprises the following steps:
[0094] Step S11: Obtaining user library reservation data;
[0095] Step S12: preprocessing the user library reservation data to generate standard user library reservation data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization;
[0096] Step S13: classify the standard user library reservation data into reservation types, generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction, wherein the smart library channel gate information verification instruction includes the entry information verification instruction and the exit information verification instruction;
[0097] Step S14: perform return information retrieval on the user's return reservation data through the door entry information verification instruction to generate return information retrieval data; perform book return process supervision and control on the first gate of the door entry channel based on the return retrieval data to generate book return process non-contact control data.
[0098] In an embodiment of the present invention, by extracting the user's reservation data from the library management system, the data includes reservation ID, user ID, reservation time, reservation type (borrowing or returning), reservation status, etc. The acquired data should be structured, such as a table or database format, to facilitate subsequent processing. Remove duplicate reservation records to ensure data uniqueness, identify and process abnormal data, such as reservation time for a future date, etc. Check missing values in reservation data, such as user ID, reservation time, etc., and use specific strategies to fill in missing values, such as using the user's historical data to fill or using default values. Ensure that all date and time formats are consistent, and that the formats of fields such as user ID and reservation ID are unified. Convert fields such as reservation type into standardized codes or descriptions, such as "borrowing" into "1" and "return" into "2".
[0099] Then, extract the record with reservation type "borrowing" from the standard user library reservation data to generate user borrowing reservation data, and extract the record with reservation type "returning" from the standard user library reservation data to generate user return reservation data. Generated based on user return reservation data, used to verify whether the user has a return reservation and allow the user to enter the library, generated based on user borrowing reservation data, used to verify whether the user has a borrowing reservation and allow the user to leave the library, and integrate the above instructions to generate the smart library channel gate information verification instruction.
[0100] When the user arrives at the library entrance, the user will verify whether he / she has a valid return reservation through the entrance information verification command. After the verification is passed, the return information retrieval data is generated, including user ID, reservation ID, reservation time, etc. According to the return information retrieval data, the first gate of the entrance channel is controlled to allow the user to enter the library to return the book. After the user completes the return, the system automatically records the return information and generates the non-sensing control data of the book return process to ensure that the entire process is seamless and non-sensing.
[0101] The above steps ensure the consistency and integrity of the data by preprocessing the user library reservation data (data cleaning, missing value filling and standardization), and improve the accuracy and reliability of subsequent data processing and analysis. Dividing the standard user library reservation data into two categories: borrowing reservations and return reservations, it helps to accurately manage and track users' borrowing and returning behaviors and improve the utilization efficiency of library resources. By analyzing the borrowing reservation data and return reservation data, the corresponding smart library channel gate information verification instructions are generated, which enhances the intelligent level of library management and makes the user entry and exit verification process more convenient and efficient.
[0102] By using the door information verification instructions to retrieve the return information and generate the return information retrieval data, the book return process of the first gate of the door channel can be supervised and controlled. This not only simplifies the user's book return process, but also effectively avoids confusion and errors in the book return process. Through intelligent reservation data management and channel gate control, the user experience when borrowing and returning books has been significantly improved, reducing waiting time and operation complexity, making library services more humane and efficient. Accurate reservation data classification and information verification instructions help libraries to reasonably arrange and dispatch resources, reduce the loss and delay of books, and improve the efficiency of book circulation. Through in-depth analysis of user library reservation data, libraries can better understand user behavior and needs, so as to formulate more targeted service strategies and decisions, and improve the overall management level and service quality.
[0103] Specifically, step S14 includes the following steps:
[0104] Step S141: performing return information retrieval on the user return reservation data through the door entry information verification instruction to generate return information retrieval data;
[0105] Step S142: Performing a user identity inventory trigger based on the first set of opposite light curtains on the first gate of the entrance passage, and obtaining the trigger inventory data of the first opposite light curtain at the entrance;
[0106] Step S143: Map the inventory data triggered by the first pair of light curtains at the entrance to the display screen on the first gate of the entrance channel to visualize the returned book information, and control the gate of the first gate of the entrance channel according to the second set of pair of light curtains on the first gate of the entrance channel to generate the senseless control data of the book return process.
[0107] In an embodiment of the present invention, when a user arrives at the library entrance, the system verifies the user's return reservation data through an entry information verification instruction, and the system retrieves the return reservation data, including user ID, reservation ID, reservation time, returned book list, etc. After the verification is passed, the system generates return information retrieval data and transmits it to relevant devices.
[0108] The first set of directional light curtains is installed on the first gate of the entrance channel to detect the user's entry behavior. When the user passes through the first set of directional light curtains, the light curtain triggers and records the user's entry time and identity information. The system generates the first directional light curtain trigger inventory data, including user ID, trigger time, identity verification result, etc. The first directional light curtain trigger inventory data is mapped to the display screen on the first gate of the entrance channel to display the user's returned book information, such as book name, return deadline, etc. The user confirms the returned book information by checking the display screen.
[0109] A second set of photoelectric screens is installed at the exit of the first gate of the entrance channel to further monitor the passage of users. When the user passes through the second set of photoelectric screens, the system controls the gate according to the triggering inventory data of the first set of photoelectric screens at the entrance, ensuring the smooth and safe passage of the user through the gate. The system records the user's passage status and book return status, generates non-sensing control data for the book return process, and realizes non-sensing book return process management.
[0110] For example, user B arrives at the library entrance and verifies his return reservation data through the system. The system retrieves user B's return reservation data: user ID is U002, reservation ID is 2001, reservation time is 2024-08-02 14:00, and the returned book list includes "Basics of Data Science", "Introduction to Artificial Intelligence", etc.
[0111] The system generates return information retrieval data and transmits it to the first gate of the entrance channel. The first set of light curtains is installed on the first gate of the entrance channel. User B passes through the first set of light curtains, which trigger and record the user's entry time and identity information. The system generates the first light curtain trigger inventory data, including user ID U002, trigger time 2024-08-02 14:01, and identity verification result passed.
[0112] The system maps the inventory data triggered by the first reflected light curtain at the entrance to the display screen on the first gate of the entrance channel, displaying the returned book information: "Basics of Data Science" and "Introduction to Artificial Intelligence". User B confirms the returned book information on the display screen, and installs the second set of reflected light curtains at the exit of the first gate of the entrance channel. The system records the passage status and book return status of user B, and generates seamless control data for the book return process to ensure that the entire book return process is seamless and smooth. When user B passes through the second set of reflected light curtains, the system controls the gate according to the inventory data of the first set of reflected light curtains, allowing the user to pass smoothly.
[0113] This step retrieves the return information of the user's return reservation data through the door entry information verification instruction, ensuring the accuracy and timeliness of the return information and reducing the possibility of data omission and misjudgment. The first group of light curtains is used to trigger the user identity inventory, automatically recording the user's door entry information, improving the efficiency and accuracy of user identity authentication and avoiding the tediousness and errors of manual inventory.
[0114] Furthermore, the inventory data triggered by the first pair of light curtains at the entrance is mapped to the gate display screen, which displays the information of the user returning the books in real time, enhancing the interactive experience between the user and the system, so that the user can clearly understand the book return process and status. The first gate of the entrance channel is controlled by the second set of light curtains, realizing the senseless control of the book return process. Users do not need to perform additional operations when returning books, which greatly improves the convenience and smoothness of returning books.
[0115] The touchless control design greatly simplifies the user's book return operation process, reduces user waiting time, improves the overall user experience, and makes the book return process more convenient and humane. Accurate identity inventory and real-time information visualization help library managers to grasp the return status in a timely manner, optimize management processes, and improve management efficiency.
[0116] The entire process of the above steps is automated through intelligent equipment and systems, without excessive human intervention, reducing errors and interference caused by human factors, and ensuring the efficiency and accuracy of the book return process. The automatic retrieval, inventory and visualization of return information have accumulated a large amount of user behavior data for the library, which is helpful for subsequent analysis and management decisions, and improves the scientificity and pertinence of library services.
[0117] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:
[0118] Step S21: Obtaining user borrowed book information data;
[0119] Step S22: Triggering the first gate of the exit passage to count the identity information of the exit user according to the book borrowing information data of the user, and generating the triggering count data of the first light curtain at the exit;
[0120] Step S23: controlling the gate based on the first gate of the exit passage by triggering the inventory data through the first pair of light curtains at the exit, and generating control data of the first gate at the exit;
[0121] Step S24: Perform multi-layer user identity authentication on the second light curtain on the second gate at the exit through the control data of the first gate at the exit, and generate borrowing identity authentication result data;
[0122] Step S25: Verify and judge the borrowing identity verification result data. When the borrowing identity verification result data is unsuccessful, the third gate of the exit channel is supervised and controlled to generate the third gate control data; when the borrowing identity verification result data is successful, the second gate of the exit channel is supervised and controlled to generate the second gate control data; the first exit gate control data, the second exit gate control data and the third exit gate control data are integrated to generate the borrowing process seamless control data.
[0123] In the embodiment of the present invention, the user's borrowing book information data is extracted from the library management system, and the data includes the borrowing ID, user ID, borrowing time, borrowing book list, etc. The acquired data should be structured, such as a table or database format, to facilitate subsequent processing. A first set of opposing light curtains is installed on the first gate of the exit passage to detect the user's leaving behavior. When the user passes through the first set of opposing light curtains, the light curtain triggers and records the user's leaving time and identity information.
[0124] The system generates the first pair of light curtains triggering inventory data, including user ID, trigger time, identity verification results, etc. The system controls the opening and closing of the first gate of the exit channel according to the first pair of light curtains triggering inventory data, ensuring that the user's identity information is recorded completely. The system generates the control data of the first gate of the exit channel, including user ID, trigger time, gate status, etc. A second set of pair of light curtains is installed on the second gate of the exit channel to further verify the user's identity information.
[0125] Through the control data of the first gate at the exit, the second set of light curtains on the second gate at the exit are used to perform multi-layer user identity authentication to ensure the accuracy of the user's identity. The system generates borrowing identity authentication result data, including user ID, authentication result, authentication time, etc. When the borrowing identity authentication result data is unsuccessful, the system supervises and controls the third gate at the exit channel and generates the third gate control data to prevent the user from leaving without authentication. When the borrowing identity authentication result data is successful, the system supervises and controls the second gate at the exit channel and generates the second gate control data to allow the user to leave.
[0126] The control data of the first gate, the second gate, and the third gate are integrated to generate the non-sensing control data of the borrowing process to ensure that the entire process is non-sensing and smooth. For example, the borrowing book information data of user C is extracted: the borrowing ID is 3001, the user ID is U003, the borrowing time is 2024-08-03 15:00, and the borrowing book list includes "Basics of Machine Learning" and "Introduction to Deep Learning". The first set of anti-reflection light curtains is installed on the first gate of the exit channel. User C passes through the first set of anti-reflection light curtains, which trigger and record the user's departure time and identity information.
[0127] The system generates the first exit light curtain trigger inventory data, including user ID U003, trigger time 2024-08-03 15:05, and identity verification result passed. The system controls the opening and closing of the first gate of the exit channel according to the first exit light curtain trigger inventory data. The system generates the first exit gate control data, including user ID U003, trigger time 2024-08-03 15:05, and gate status is open.
[0128] Install a second set of reflected light curtains on the second gate of the exit channel. User C passes through the second set of reflected light curtains, and the system performs multi-layer identity authentication on the user through the control data of the first gate of the exit. The system generates borrowing identity authentication result data, including user ID U003, verification result success, and verification time 2024-08-03 15:06. The system supervises and controls the second gate of the exit channel, generates the control data of the second gate, and allows user C to leave smoothly. If the verification is unsuccessful, the system supervises and controls the third gate of the exit channel, generates the control data of the third gate, and prevents user C from leaving. The system integrates the control data of the first gate, the second gate, and the third gate, and generates the non-sensing control data of the borrowing process to ensure that the entire process is non-sensing and smooth.
[0129] This step ensures the comprehensiveness and accuracy of book borrowing information by obtaining user borrowing information data, and provides basic data support for subsequent exit management. The user identity information inventory is triggered by the reflected light curtain of the first gate of the exit channel, and the user's exit behavior is automatically recorded, which improves the efficiency and accuracy of user identity verification and reduces the error of manual inventory.
[0130] The first pair of light curtains at the exit are used to trigger the inventory data to control the first gate of the exit channel, realizing the intelligent management of the exit process and ensuring the safety and compliance of users' exit. The second pair of light curtains are used to perform multi-layer user identity authentication, further enhancing the security of user identity authentication, preventing unauthorized users from leaving the library, and improving the library's security management level. According to the success or failure of the borrowing identity authentication result data, the second and third gates of the exit channel are supervised and controlled accordingly, ensuring the flexibility and reliability of the exit process and effectively preventing erroneous or unauthorized exit behaviors. Through the intelligent control and data integration of the exit gates, the borrowing process is controlled without feeling, so that users do not need to perform additional operations during the borrowing and exiting process, which greatly improves the user experience and the smoothness of the process.
[0131] The intelligent and touchless control design simplifies the user's borrowing and exiting operation process, reduces user waiting time, improves the overall user experience, and makes the borrowing and exiting process more convenient and user-friendly. Intelligent inventory and multi-layer identity authentication help library managers to timely grasp the borrowing and exiting situation, optimize management processes, improve management efficiency, and reduce manual intervention and management costs. Through the automated management of the borrowing and exiting process, a large amount of user behavior data has been accumulated, which is helpful for subsequent analysis and management decisions, and improves the scientificity and pertinence of library services.
[0132] Preferably, step S24 includes the following steps:
[0133] Step S241: dividing the identity access mode through the exit information verification instruction to generate the identity access mode, wherein the identity access mode includes the IC card access mode, the ID card access mode and the user face access mode;
[0134] Step S242: when it is confirmed that the identity access mode is the IC card access mode or the ID card access mode, the corresponding data based on the IC card access mode or the ID card access mode is scanned to generate identity information scan data; the identity information scan data is matched with the standard user library reservation data to generate borrowing identity verification result data;
[0135] Step S243: When the identity access mode is confirmed to be the user face access mode, the face recognition camera on the second gate of the entrance channel is used to virtually read the user access face image to obtain the user reservation access face image; the user reservation access face image and the face image in the standard user library reservation data are aligned with the face feature points to generate face access feature alignment data; the matching degree of the face access feature alignment data is calculated to generate the access face matching degree;
[0136] Step S244: Compare the access face matching degree with the preset face matching threshold. When the access face matching degree is greater than or equal to the preset face matching threshold, generate identity information trusted data according to the corresponding user scheduled access face image; when the access face matching degree is less than the preset face matching threshold, mark the corresponding user scheduled access face image as a suspicious face image;
[0137] Step S245: Perform user liveness verification detection on the suspicious facial image to generate suspicious user liveness verification detection data; perform deep information investigation on the suspicious facial image through the suspicious user liveness verification detection data to generate identity information deep verification data; integrate the identity information trusted data and the identity information deep verification data to also generate borrowing identity verification result data.
[0138] In the embodiment of the present invention, the system divides the user's identity access mode through the exit information verification instruction. The identity access mode includes the IC card access mode, the ID card access mode and the user face access mode. When the identity access mode is confirmed to be the IC card access mode or the ID card access mode, the following steps are performed: the corresponding data based on the IC card access mode or the ID card access mode is scanned to generate identity information scan data, the identity information scan data is matched with the standard user library reservation data, and the borrowing identity verification result data is generated.
[0139] When it is confirmed that the identity access mode is the user face access mode, the following steps are performed: use the face recognition camera on the second gate of the exit channel to virtually read the user access face image, obtain the user's scheduled access face image, align the user's scheduled access face image with the face image in the standard user library reservation data for facial feature points, generate face access feature alignment data, calculate the matching degree of the face access feature alignment data, and generate the access face matching degree.
[0140] Compare the access face matching degree with the preset face matching threshold: when the access face matching degree is greater than or equal to the preset face matching threshold, generate identity information trusted data based on the corresponding user's scheduled access face image; when the access face matching degree is less than the preset face matching threshold, mark the corresponding user's scheduled access face image as a suspicious face image. Perform user liveness verification detection on the suspicious face image to generate suspicious user liveness verification detection data. Perform deep information screening on the suspicious face image through the suspicious user liveness verification detection data to generate identity information deep verification data. Integrate the identity information trusted data and the identity information deep verification data to generate borrowing identity authentication result data.
[0141] Preferably, performing user liveness verification detection on a suspicious face image includes:
[0142] Perform facial texture analysis on suspicious face images to generate user facial texture feature data;
[0143] The user's light reflection is detected through the suspicious face image, and the reflection data is collected through the optical sensor to obtain the facial spectral reflection data; the facial spectral reflection data is analyzed for reflection characteristics to generate the user's facial light reflection characteristic data;
[0144] The facial three-dimensional structure information of the suspicious face image is extracted by the structured light sensor to obtain the facial three-dimensional structure information data; the facial three-dimensional structure information data is analyzed for facial depth features to generate the user's facial depth feature data;
[0145] Perform dynamic facial behavior recognition on the user corresponding to the suspicious face image to generate user dynamic behavior recognition data; perform comprehensive user liveness verification detection based on user facial texture feature data, user facial light reflection feature data, user facial depth feature data and user dynamic behavior recognition data to generate suspicious user liveness verification detection data.
[0146] In an embodiment of the present invention, by acquiring a facial image marked as suspicious, facial texture analysis is performed on the suspicious facial image, fine facial texture features in the image are identified and extracted, an advanced image processing algorithm is used to analyze the user's facial texture, fine facial features such as pores and wrinkles are identified, and the extracted texture feature data is organized into the user's facial texture feature data.
[0147] The optical sensor is used to detect light reflection of suspicious facial images, collect light reflection data of the user's face under different lighting conditions, organize and record the collected facial spectral reflection data, analyze the facial spectral reflection data, extract the user's light reflection characteristics, and organize the analyzed characteristic data into user facial light reflection characteristic data.
[0148] The structured light sensor is used to extract the facial three-dimensional structural information of the suspicious face image, obtain the facial three-dimensional structural information data, analyze the facial three-dimensional structural information data, extract the user's facial depth features, such as nose bridge height, eye socket depth, etc., and organize the analyzed depth feature data into the user's facial depth feature data. The user's dynamic facial behavior, such as blinking, opening the mouth, etc., is captured through video, and the user's dynamic facial behavior data is recorded. The dynamic facial behavior data is analyzed to identify the user's specific behavior pattern, and the analyzed behavior recognition data is organized into the user's dynamic behavior recognition data. The user's facial texture feature data, facial light reflection feature data, facial depth feature data and dynamic behavior recognition data are comprehensively analyzed, and the user liveness verification is performed based on the comprehensive analysis results to confirm whether the user is a real person, and the results of the comprehensive liveness verification test are used to generate suspicious user liveness verification test data.
[0149] As mentioned above, this step adapts to the identity authentication needs of different users by dividing the identity access mode (IC card access, ID card access and face access), and improves the flexibility and applicability of identity authentication. Corresponding data scanning and user identity matching for IC card and ID card access modes ensures the accuracy and speed of identity authentication and reduces the possibility of incorrect matching. The face recognition camera is used to perform virtual reading and feature point registration of user access face images, which has a high degree of automation and fast verification speed, greatly improving the efficiency of user identity authentication.
[0150] By calculating the access face matching degree and comparing it with the preset threshold, the reliability and accuracy of face recognition is guaranteed, and misjudgment and misuse by unauthorized users are prevented. For suspicious face images with access face matching degrees below the threshold, the user's identity is further confirmed through user liveness verification detection and deep information investigation, enhancing the security and anti-counterfeiting capabilities of the system. In the case of a successful match, credible identity information data is generated to ensure the reliability of the user's identity and provide strong protection for the subsequent borrowing process.
[0151] By conducting in-depth information screening on suspicious users and generating deep identity information verification data, we ensure that all user identities are strictly verified, thus improving the security of library management. By integrating trusted identity data and deep verification data, we can uniformly generate borrowing identity verification result data, thus improving the efficiency and consistency of data processing. Diversified identity verification methods and efficient identification technology eliminate the need for cumbersome operations when users go out, greatly improving the user experience and the smoothness of the process. Through intelligent identity verification and multi-layer verification mechanisms, library managers can more efficiently grasp the user's borrowing and going out situation, optimize management processes, and improve overall management efficiency.
[0152] By analyzing users in multiple dimensions such as facial texture, light reflection, three-dimensional structure and dynamic behavior, the comprehensiveness and accuracy of liveness verification are improved, and misjudgment caused by single feature verification is avoided. Facial texture feature data can accurately capture the subtle features and texture information of the user's skin, increase the level of detail of identity authentication, and improve the accuracy of verification. Through light reflection detection and facial spectral reflection feature analysis, real people and forged images such as photos can be effectively distinguished, which improves the reliability of liveness verification. Using structured light sensors to extract facial three-dimensional structural information, through deep feature analysis, the three-dimensional sense and authenticity of facial recognition are ensured, and deception by two-dimensional forged images is prevented.
[0153] By performing dynamic facial behavior recognition on users and capturing the dynamic changes and behavioral characteristics of their faces, it helps to further confirm the authenticity of the user's identity and prevent forgery of static photos or videos. Comprehensive analysis of facial texture, light reflection, three-dimensional structure and dynamic behavioral characteristics forms a comprehensive user liveness verification test to ensure high accuracy and high credibility of the verification results. Multi-dimensional and multi-level liveness verification tests effectively prevent forgery and fraud, enhance the security of the system, and ensure the authenticity and uniqueness of the user's identity.
[0154] Through the efficient liveness verification detection process, users can complete identity authentication without being aware of the complex verification process, which improves the smoothness and convenience of the user experience. Automated liveness verification detection reduces the need for manual intervention, improves the management efficiency of the system, and allows library managers to focus more on other important tasks. Through multi-dimensional liveness verification detection, identity theft and fraud are effectively prevented, and user privacy and security are protected.
[0155] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:
[0156] Step S31: historical data collection of user borrowing book information data to obtain historical user borrowing book information data;
[0157] Step S32: performing a borrowing risk assessment on the user's borrowing book information data according to the historical user's borrowing book information data to generate user borrowing risk assessment data;
[0158] Step S33: classifying the user's borrowed book information data into low-credit users based on the user's borrowing risk assessment data, and generating low-credit user grade data;
[0159] Step S34: Perform gate warning processing on the borrowing process contactless control data and the returning process contactless control data through the low-credit user level data to generate low-credit user warning data.
[0160] In the embodiment of the present invention, the user's current borrowing book information data is obtained from the database. All borrowing records of the user are extracted from the historical record database to generate historical user borrowing book information data. The data includes borrowing time, return time, borrowing book type, overdue record, etc., and the user's borrowing behavior is analyzed, including borrowing frequency, overdue number, return timeliness, etc. The borrowing risk assessment model is used to calculate the historical data to generate user borrowing risk assessment data.
[0161] The assessment model can be based on machine learning algorithms and combined with the user's historical behavior characteristics to perform risk scoring. Based on the model calculation results, the user's borrowing risk assessment data is generated, including the user's risk level and risk score. Credit grading rules are formulated to grade users based on risk assessment data. The rules may include high credit, medium credit, low credit and other levels. According to the user's risk score, the user is classified into different credit levels, the grading results are recorded, and the low credit user level data is generated.
[0162] By formulating early warning rules, the gate is warned according to the low-credit user level data. When the user's credit level is low, the system triggers the early warning mechanism. The system records and generates low-credit user early warning data as the basis for the gate warning processing. More specifically, the information of the book currently borrowed by the user is obtained from the library management system, including the book title, borrowing date, etc. The user's historical borrowing records are extracted from the database, including borrowing time, return time, overdue records, etc. For example, all the borrowing records of the user in the past year, and the user's borrowing frequency, overdue times, and timely return rate are counted.
[0163] Further use machine learning models (such as random forests, logistic regression, etc.) to analyze the user's historical behavior data. The input data includes borrowing frequency, number of overdue times, timeliness of return, etc., and the user's risk score is output. According to the risk score calculated by the model, the user's borrowing risk assessment data is generated. Set credit grading standards, for example: high credit: risk score ≥ 80; medium credit: 60 ≤ risk score <80; low credit: risk score <60. According to the user's risk score, the user is classified into the corresponding credit level, and the grading results are recorded in the system to generate low-credit user level data. Formulate warning rules, for example, when a low-credit user borrows or returns a book, a warning is triggered. When the system detects a low-credit user, the warning mechanism of the gate is triggered, and the system generates low-credit user warning data, prompting staff to conduct manual verification or take other measures.
[0164] As mentioned above, this step accumulates rich user behavior data by collecting historical data of user borrowing information data, which helps to fully understand the user's borrowing habits and borrowing records. Borrowing risk assessment based on historical user borrowing information data can effectively identify potential high-risk users and abnormal borrowing behaviors, and improve the library's ability to predict user borrowing behaviors.
[0165] By using the borrowing risk assessment data to classify users with low credit, the library can implement targeted management and control of low-credit users and reduce the risk of book losses and borrowing defaults. By using the low-credit user grade data to perform gate warning processing on the borrowing and return process, real-time monitoring and warning can be carried out when users enter and exit the library, preventing improper behavior of low-credit users and improving the management security of the library. Combined with the low-credit user warning data, the borrowing and return process can be controlled without feeling, so that low-credit users can be effectively monitored and managed throughout the process without affecting the borrowing experience of normal users.
[0166] The automated risk assessment and early warning processing of the above steps reduces the need for manual intervention, improves library management efficiency, and enables managers to focus more on other core tasks. Through accurate risk assessment and early warning mechanisms, normal users can enjoy convenient services without any sense of control, while abnormal behaviors of low-credit users can be discovered and handled in a timely manner, optimizing the overall user experience. Through the classification and early warning processing of low-credit users, book losses and borrowing defaults are effectively prevented, and the library's security management level is improved.
[0167] By using historical user borrowing data and risk assessment data, libraries can make scientific decisions based on data, improving the accuracy and reliability of management decisions. Through risk assessment and early warning mechanisms, libraries can identify and respond to potential risks in advance, achieve preventive management, and reduce the cost and complexity of post-event remediation.
[0168] Preferably, step S32 includes the following steps:
[0169] Step S321: dividing the historical user borrowing book information data into a data set to obtain a model training set and a model test set; training the model training set by a logistic regression algorithm to generate a borrowing risk assessment pre-model;
[0170] Step S322: using the model test set to iterate the model test on the borrowing risk assessment pre-model, thereby generating a borrowing risk assessment model;
[0171] Step S323: Importing the user's borrowing book information data into the borrowing risk assessment model to perform borrowing risk assessment and generate user borrowing risk assessment data.
[0172] In the embodiment of the present invention, features are extracted from historical user borrowing information data, including borrowing frequency, overdue records, timeliness of return, etc., to perform data preprocessing, handle missing values, outliers, and perform standardization. The cleaned data set is divided into a training set and a test set. The common ratio is 80% for training and 20% for testing. The logistic regression algorithm is selected as the basis for model training.
[0173] Logistic regression is suitable for binary classification problems, such as high and low borrowing risk classification. Use the training set to train the model and train the logistic regression model. Feature selection includes borrowing frequency, overdue times, borrowed book types, etc., optimize model parameters, fit training data, and generate a borrowing risk assessment pre-model. Input the test set data into the trained borrowing risk assessment pre-model for prediction. Use common indicators such as precision, recall rate, F1 score, etc. to evaluate the performance of the model on the test set.
[0174] Adjust the model parameters according to the test results and optimize the logistic regression model. Perform multiple iterations until the model performance reaches the expected standard. After multiple tests and optimizations, generate the final borrowing risk assessment model. Import the user's current borrowing book information data into the generated borrowing risk assessment model. The data includes the user's borrowing frequency, borrowing type, overdue record, etc. Use the borrowing risk assessment model to predict the imported data, calculate the user's risk score, and generate the user's borrowing risk assessment data, including the risk level (such as high, medium, low) and the specific risk score.
[0175] As mentioned above, this step divides the data set of historical user borrowing information data into model training sets and model test sets, which is helpful for model training and evaluation and ensures the generalization ability and accuracy of the model. Using the logistic regression algorithm to train the model training set can effectively identify and capture the characteristics of user borrowing behavior, generate a borrowing risk assessment pre-model, and provide a basis for subsequent risk assessment.
[0176] The borrowing risk assessment pre-model is tested and iterated through the model test set, and the model performance and accuracy are continuously optimized to ensure that the borrowing risk assessment model generated in the end has efficient and reliable risk assessment capabilities. Importing user borrowing book information data into the optimized borrowing risk assessment model for risk assessment can accurately identify potential high-risk users and abnormal borrowing behaviors, and generate user borrowing risk assessment data.
[0177] Through logistic regression algorithms and model testing iterations, the borrowing risk assessment model can achieve high-accuracy and high-reliability risk assessment results, reducing misjudgments and missed judgments. Based on model training and risk assessment of historical user borrowing data, the library can make scientific management decisions based on the assessment results, improving the accuracy and effectiveness of management.
[0178] The borrowing risk assessment model can realize real-time user risk assessment, automatically identify high-risk users, and improve the efficiency and security of library management. Through the risk assessment of user borrowing behavior, the library can identify and prevent potential borrowing defaults and book losses in advance, realize preventive management, and reduce the cost of subsequent remediation.
[0179] The precise risk assessment and management measures in this step enable normal users to enjoy convenient borrowing services, while high-risk users can be identified and handled in a timely manner, optimizing the overall user experience. Model testing iterations and risk assessments can be dynamically adjusted and optimized based on new data to ensure that the model can continuously adapt to and respond to changing user behaviors and risk characteristics.
[0180] As an example of the present invention, refer to Figure 4 As shown, in this example, step S4 includes:
[0181] Step S41: Performing intelligent early warning response on the smart library channel gate according to the low-credit user early warning data to generate intelligent early warning response data;
[0182] Step S42: encapsulating the intelligent warning response data, the book return process senseless control data and the borrowing process senseless control data into functional modules to generate a smart library senseless control module;
[0183] Step S43: deploy motion sensors on the smart library channel gate based on the smart library non-sensing control module to obtain a non-sensing borrowing motion sensor network; monitor the module function triggering of the smart library non-sensing control module through the non-sensing borrowing motion sensor network to execute the smart library non-sensing borrowing operation.
[0184] In the embodiment of the present invention, key information such as user risk level and warning type is extracted from the warning data of low-credit users. The warning data is integrated with the non-sensitive control data of the book return process and the non-sensitive control data of the borrowing process. Intelligent warning response rules are defined, and the intensity and type of the warning are determined according to the level of the warning data (such as high, medium, and low). According to the defined rules, intelligent warning response data are generated, which include control instructions for the channel gate, warning information, and processing measures.
[0185] By designing the smart library non-sensing control module, the functions of intelligent warning response data, non-sensing control data of the book return process and non-sensing control data of the borrowing process are integrated. The above functional modules are encapsulated into an integrated control module. Ensure that the module can run independently and interact effectively with the smart library channel gate. Perform functional tests on the smart library non-sensing control module to ensure that it can correctly respond to intelligent warnings and control the book return and borrowing processes.
[0186] Ensure stability and reliability by adjusting the module's functions according to the test results. Select motion sensors for real-time monitoring and control of the sensorless borrowing operation. Arrange motion sensors on the smart library channel gates to form a sensorless borrowing motion sensor network. Ensure that the sensors cover all key areas. Connect the smart library sensorless control module to the motion sensor network. Configure the sensor's operating parameters, such as sensitivity and trigger threshold. Monitor the movement in the library in real time through the sensorless borrowing motion sensor network. Trigger the functions of the smart library sensorless control module based on the sensor data, including the execution of the sensorless borrowing operation.
[0187] By processing the data collected by the sensors, the smooth operation of the non-sensitive borrowing operation is ensured. The processing includes data cleaning, feature extraction and real-time response. More specifically, for example, the user ID and risk level in the low-credit user warning data are extracted, and the control data of the return and borrowing process are integrated. It is defined that high-risk users need to be processed immediately, medium-risk users need to be monitored, and low-risk users do not need special processing. The corresponding control instructions and warning information are generated to ensure that the gate can respond accordingly.
[0188] In the above steps, the module architecture is designed to include early warning response processing, borrowing and returning book control, ensuring that the module functions can be correctly called and executed. Run the control module in the test environment to verify whether its functions meet expectations, make adjustments, optimize module performance, and ensure the stable operation of the system. Motion sensors are placed at entrances and exits and key areas to ensure that the sensors can cover all key areas of the library. Configure sensor trigger conditions, set the corresponding sensitivity, monitor sensor data, and trigger the relevant functions of the smart library's non-sensing control module based on real-time data to ensure the smooth execution of non-sensing borrowing operations.
[0189] In this step, by responding to the intelligent warning of the smart library channel gate according to the low-credit user warning data, the behavior of low-credit users can be identified and responded to in a timely and effective manner, thereby improving the security management level of the library. The intelligent warning response data, the book return process senseless control data, and the borrowing process senseless control data are encapsulated in functional modules to generate the smart library senseless control module, which realizes data integration and functional modularization and simplifies system management and maintenance.
[0190] Through the smart library's non-sensing control module, motion sensors are deployed on the channel gate to form a non-sensing borrowing motion sensor network, which can monitor users' borrowing and returning behaviors in real time, improving the intelligence level of the system. Through the non-sensing borrowing motion sensor network, the smart library's non-sensing control module is monitored for module function triggering, ensuring that each functional module can respond and execute in time, and complete borrowing and returning operations without perception, improving user experience and the automation of the system. The integrated design of intelligent early warning response and non-sensing control modules enables the library management system to operate efficiently, reduces manual intervention, and improves management efficiency.
[0191] The effective combination of intelligent early warning response data and low-credit user early warning data enhances the security of the library, can prevent and handle abnormal behavior in a timely manner, and ensure the security of library resources. The deployment and monitoring of the non-sensing borrowing motion sensor network realizes non-sensing borrowing and returning operations. Users can complete borrowing and returning books without additional operations, which greatly improves the user experience. The design of functional module encapsulation and sensor deployment makes the system highly flexible and scalable, and functions can be added and adjusted as needed to adapt to different scenarios and needs.
[0192] The motion sensor network for non-sensing borrowing can monitor user behavior in real time, provide timely data feedback and processing capabilities, and enable the library to achieve real-time management and response. Through intelligent early warning response and module function trigger monitoring, the system can automatically perform borrowing and returning operations, reducing human intervention and operational errors, and improving the accuracy and efficiency of operations.
[0193] Embodiment 2: This embodiment provides a smart library non-sensing borrowing data processing system, which is used to execute a smart library non-sensing borrowing data processing method of embodiment 1. The smart library non-sensing borrowing data processing system includes:
[0194] The library entry and book return module is used to obtain user library reservation data; classify the user library reservation data by reservation type, and generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data;
[0195] The library outbound borrowing module is used to obtain the user's borrowing book information data; control the first gate of the outbound channel according to the user's borrowing book information data, and generate the first gate control data; perform multi-layer identity authentication of the second gate user on the first gate control data through the smart library channel gate information verification instruction, and generate the borrowing identity authentication result data; according to the borrowing identity authentication result data; according to the borrowing identity authentication result data, supervise and control the outbound channel borrowing process, and generate the borrowing process non-sensing control data;
[0196] The risk warning module is used to collect historical data of user borrowing book information data to obtain historical user borrowing book information data; conduct borrowing risk assessment on user borrowing book information data based on historical user borrowing book information data to generate user borrowing risk assessment data; conduct gate warning processing on borrowing process non-sensing control data and return process non-sensing control data based on user borrowing risk assessment data to generate low-credit user warning data;
[0197] The motion trigger module is used to perform intelligent warning response to the smart library channel gate according to the low-credit user warning data and generate intelligent warning response data; according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data, the smart library channel gate is motion-sensor-triggered and monitored to execute the smart library contactless borrowing operation.
[0198] In summary, the present invention classifies the reservation types of user library reservation data into two types: borrowing and returning, which helps to clarify different types of operation requirements and generate information verification instructions for the smart library channel gate, which can accurately control user entry and exit, ensure the correctness and security of the process, generate senseless control data for the book return process, improve the automation level of library operations, and reduce manual intervention.
[0199] By controlling the first gate at the exit, the user borrowing process is ensured to proceed smoothly, improving the user experience. The multi-layer identity verification of the user at the second gate ensures the accuracy of the borrowing identity, effectively preventing identity impersonation and fraud. The generated borrowing process senseless control data makes the borrowing process more efficient and seamless, reducing the complexity in operation.
[0200] Through historical data collection and risk assessment, scientific analysis of users' borrowing behavior helps identify potential risks. Generating low-credit user warning data based on risk assessment results helps to take timely measures to prevent potential borrowing default risks. Through intelligent early warning responses to low-credit users, abnormal situations can be handled in real time to ensure the normal operation of the library. Motion sensor trigger monitoring of smart library channel gates improves the automation of borrowing operations and reduces manual intervention. The execution of non-sensing borrowing operations reduces the complexity of user operations and improves the overall borrowing experience.
[0201] The automation and intelligent design of each step in the present invention significantly improves the management efficiency of the library and reduces manual intervention and errors. Through accurate identity verification and risk warning, the library can effectively ensure the security of resources and reduce unnecessary losses. The automated borrowing and returning process, intelligent warning response and risk management measures enhance the overall service experience of users.
[0202] Therefore, the present invention improves the efficiency, security and user experience of seamless borrowing through automated process control, multi-layer identity authentication, risk assessment and intelligent early warning.
[0203] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A method for processing data of seamless borrowing in a smart library, characterized in that: The following steps are involved: Step S1: Obtain user library reservation data; Classify the user's library reservation data by reservation type, and generate user borrowing reservation data and user return reservation data; According to the user's borrowing reservation data and the user's return reservation data, the information of the smart library channel gate is associated, and the smart library channel gate information verification instruction is generated; the user's return reservation data is supervised and controlled through the door information verification instruction, and the book return process is generated. The step S1 specifically includes the following steps: Step S11: Obtaining user library reservation data; Step S12: preprocessing the user library reservation data to generate standard user library reservation data, wherein the data preprocessing includes data cleaning, data missing value filling and data standardization; Step S13: classify the standard user library reservation data into reservation types, generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction, wherein the smart library channel gate information verification instruction includes the entry information verification instruction and the exit information verification instruction; Step S14: perform return information retrieval on the user's return reservation data through the door entry information verification instruction to generate return information retrieval data; perform book return process supervision and control on the first gate of the door entry channel based on the return retrieval data to generate book return process non-sensing control data; Step S14 specifically includes the following steps: Step S141: performing return information retrieval on the user return reservation data through the door entry information verification instruction to generate return information retrieval data; Step S142: Performing a user identity inventory trigger based on the first set of opposite light curtains on the first gate of the entrance passage, and obtaining the trigger inventory data of the first opposite light curtain at the entrance; Step S143: Mapping the inventory data triggered by the first pair of light curtains at the entrance to the display screen on the first gate of the entrance passage to visualize the returned book information, and controlling the gate of the first gate of the entrance passage according to the second pair of light curtains on the first gate of the entrance passage to generate the senseless control data of the book return process; Step S2: obtaining user borrowing book information data; controlling the first gate of the exit channel according to the user borrowing book information data, and generating control data of the first gate of the exit channel; performing multi-layer identity authentication of the user of the second gate of the exit channel on the control data of the first gate of the exit channel through the smart library channel gate information verification instruction, and generating borrowing identity authentication result data; performing supervision and control of the borrowing process of the exit channel according to the borrowing identity authentication result data, and generating non-sensing control data of the borrowing process; Step S2 specifically includes the following steps: Step S21: Obtaining user borrowed book information data; Step S22: Triggering the first gate of the exit passage to count the identity information of the exit user according to the book borrowing information data of the user, and generating the triggering count data of the first light curtain at the exit; Step S23: controlling the gate based on the first gate of the exit passage by triggering the inventory data through the first pair of light curtains at the exit, and generating control data of the first gate at the exit; Step S24: Perform multi-layer user identity authentication on the second light curtain on the second gate at the exit through the control data of the first gate at the exit, and generate borrowing identity authentication result data; Step S24 specifically includes the following steps: Step S241: dividing the identity access mode through the exit information verification instruction to generate the identity access mode, wherein the identity access mode includes the IC card access mode, the ID card access mode and the user face access mode; Step S242: when it is confirmed that the identity access mode is the IC card access mode or the ID card access mode, the corresponding data based on the IC card access mode or the ID card access mode is scanned to generate identity information scan data; the identity information scan data is matched with the standard user library reservation data to generate borrowing identity verification result data; Step S243: When the identity access mode is confirmed to be the user face access mode, the face recognition camera on the second gate of the entrance channel is used to virtually read the user access face image to obtain the user reservation access face image; the user reservation access face image and the face image in the standard user library reservation data are aligned with the face feature points to generate face access feature alignment data; the matching degree of the face access feature alignment data is calculated to generate the access face matching degree; Step S244: Compare the access face matching degree with the preset face matching threshold. When the access face matching degree is greater than or equal to the preset face matching threshold, generate identity information trusted data according to the corresponding user scheduled access face image; when the access face matching degree is less than the preset face matching threshold, mark the corresponding user scheduled access face image as a suspicious face image; Step S245: Perform user liveness verification detection on the suspicious face image to generate suspicious user liveness verification detection data; perform deep information screening on the suspicious face image through the suspicious user liveness verification detection data to generate identity information deep verification data; integrate the identity information trusted data and the identity information deep verification data to also generate borrowing identity verification result data; Step S25: Verify and judge the borrowing identity verification result data. When the borrowing identity verification result data is unsuccessful, the third gate of the exit channel is supervised and controlled to generate the third gate control data; when the borrowing identity verification result data is successful, the second gate of the exit channel is supervised and controlled to generate the second gate control data; integrate the first gate control data, the second gate control data and the third gate control data to generate the borrowing process senseless control data; Step S3: historical data collection is performed on user borrowing book information data to obtain historical user borrowing book information data; borrowing risk assessment is performed on user borrowing book information data according to historical user borrowing book information data to generate user borrowing risk assessment data; gate warning processing is performed on borrowing process non-sensing control data and return process non-sensing control data based on user borrowing risk assessment data to generate low-credit user warning data; Step S3 specifically includes the following steps: Step S31: historical data collection of user borrowing book information data to obtain historical user borrowing book information data; Step S32: Perform a borrowing risk assessment on the user borrowing book information data according to the historical user borrowing book information data to generate user borrowing risk assessment data; Step S32 specifically includes the following steps: Step S321: dividing the historical user borrowing book information data into a data set to obtain a model training set and a model test set; training the model training set using a logistic regression algorithm to generate a borrowing risk assessment pre-model; Step S322: using the model test set to iterate the model test on the borrowing risk assessment pre-model, thereby generating a borrowing risk assessment model; Step S323: importing the user's borrowing book information data into the borrowing risk assessment model to perform borrowing risk assessment and generate user borrowing risk assessment data; Step S33: classifying the user's borrowed book information data into low-credit users based on the user's borrowing risk assessment data, and generating low-credit user grade data; Step S34: performing gate warning processing on the borrowing process non-sensitive control data and the returning process non-sensitive control data through the low-credit user level data to generate low-credit user warning data; Step S4: Perform intelligent warning response on the smart library channel gate according to the low-credit user warning data to generate intelligent warning response data; perform motion sensor trigger monitoring on the smart library channel gate according to the intelligent warning response data, the book return process senseless control data and the borrowing process senseless control data to perform the smart library senseless borrowing operation; Step S4 specifically includes the following steps: Step S41: Performing intelligent early warning response on the smart library channel gate according to the low-credit user early warning data to generate intelligent early warning response data; Step S42: encapsulating the intelligent warning response data, the book return process senseless control data and the borrowing process senseless control data into functional modules to generate a smart library senseless control module; Step S43: deploy motion sensors on the smart library channel gate based on the smart library non-sensing control module to obtain a non-sensing borrowing motion sensor network; monitor the module function triggering of the smart library non-sensing control module through the non-sensing borrowing motion sensor network to perform the smart library non-sensing borrowing operation; The step S245 of performing user liveness verification detection on the suspicious face image includes: Perform facial texture analysis on suspicious face images to generate user facial texture feature data; The user's light reflection is detected through the suspicious face image, and the reflection data is collected through the optical sensor to obtain the facial spectral reflection data; the facial spectral reflection data is analyzed for reflection characteristics to generate the user's facial light reflection characteristic data; The facial three-dimensional structure information of the suspicious face image is extracted by the structured light sensor to obtain the facial three-dimensional structure information data; the facial three-dimensional structure information data is analyzed for facial depth features to generate the user's facial depth feature data; Perform dynamic facial behavior recognition on the user corresponding to the suspicious face image to generate user dynamic behavior recognition data; perform comprehensive user liveness verification detection based on user facial texture feature data, user facial light reflection feature data, user facial depth feature data and user dynamic behavior recognition data to generate suspicious user liveness verification detection data.
2. A smart library seamless borrowing data processing system, characterized in that: Used to execute the method for processing data of seamless borrowing in a smart library as claimed in claim 1, the system for processing data of seamless borrowing in a smart library comprises: The library entry and book return module is used to obtain user library reservation data; classify the user library reservation data by reservation type, and generate user borrowing reservation data and user return reservation data; associate the information of the smart library channel gate according to the user borrowing reservation data and the user return reservation data, and generate the smart library channel gate information verification instruction; supervise and control the entry channel book return process of the user return reservation data through the entry information verification instruction, and generate the book return process non-sensing control data; The library outbound borrowing module is used to obtain the user's borrowing book information data; control the first gate of the outbound channel according to the user's borrowing book information data, and generate the first gate control data; perform multi-layer identity authentication of the second gate user on the first gate control data through the smart library channel gate information verification instruction, and generate the borrowing identity authentication result data; according to the borrowing identity authentication result data; according to the borrowing identity authentication result data, supervise and control the outbound channel borrowing process, and generate the borrowing process non-sensing control data; The risk warning module is used to collect historical data of user borrowing book information data to obtain historical user borrowing book information data; conduct borrowing risk assessment on user borrowing book information data based on historical user borrowing book information data to generate user borrowing risk assessment data; conduct gate warning processing on borrowing process non-sensing control data and return process non-sensing control data based on user borrowing risk assessment data to generate low-credit user warning data; The motion trigger module is used to perform intelligent warning response to the smart library channel gate according to the low-credit user warning data and generate intelligent warning response data; according to the intelligent warning response data, the book return process contactless control data and the borrowing process contactless control data, the smart library channel gate is motion-sensor-triggered and monitored to execute the smart library contactless borrowing operation.
