Classroom sign-in and formability evaluation method and system based on image recognition and seat occupation

Abstract

The invention discloses a classroom sign-in and formability evaluation method and system based on image recognition and seat occupation, and relates to the technical field of educational informationization, and the method comprises the steps: a teacher end presets a graphical virtual seat table and a classroom sign-in rule containing a differential score assignment rule; after the student terminal authorizes login, selecting a target seat on the seating list to trigger sign-in; the system verifies the sign-in request of the student; after the verification is passed, automatically adding corresponding points for the account according to the attribute of the seat successfully occupied by the student; the system also records and aggregates sign-in, integral and classroom behavior data in real time for formative evaluation analysis. According to the method, the problems that a conventional sign-in mode has a sign-in loophole, an actual seat position cannot be verified, and sign-in data and a teaching evaluation link are disjointed are effectively solved, high-credibility identity and position dual verification is realized, and deep fusion of automatic acquisition of classroom participation data and teaching evaluation is promoted.

Description

Technical Field

[0001] This invention relates to the field of educational informatization technology, and in particular to a method and system for classroom attendance and formative assessment based on image recognition and seat reservation. Background Technology

[0002] Class attendance is a fundamental aspect of teaching management, and its technology has evolved from purely manual to digital and intelligent. Traditional verbal roll call is inefficient, consumes valuable class time, and fails to record student behavior data. With technological advancements, electronic attendance systems based on QR codes, GPS / LBS positioning, Bluetooth / Wi-Fi, and other technologies have emerged, achieving a leap from manual to digital recording and significantly improving attendance efficiency.

[0003] However, these existing technological solutions have significant limitations. First, their ability to prevent proxy attendance is weak. For example, QR codes can be shared remotely, and GPS positioning is not accurate enough indoors, making it difficult to prevent impersonation during attendance. Second, the core function of these solutions is only to verify whether a student is "present." The attendance process is completely disconnected from the student's actual physical seating information in the classroom, failing to capture important classroom participation data such as "seat selection." Furthermore, existing attendance functions typically exist as an isolated component, failing to be organically integrated and automatically linked with formative assessment systems such as classroom interaction and performance recording.

[0004] In recent years, biometric technologies such as facial recognition have been introduced into classroom attendance. While this has improved the reliability of identity verification, it has also brought new problems. It requires the deployment of dedicated hardware in classrooms, which is costly and raises privacy concerns. Furthermore, its facial recognition rate is easily affected by factors such as student posture and occlusion. More importantly, this technology can only verify identity; it cannot determine whether a student is sitting in the correct or specific location. Its functionality is rigid and fails to address the issue of deeply integrating attendance with the teaching process.

[0005] While a few solutions attempt to link seating information by deploying Bluetooth beacons on desks and chairs, this further increases hardware costs and deployment complexity, and still cannot effectively verify user identities. Existing classroom interaction tools, although possessing attendance and manual scoring functions, operate these functions separately and lack an automated linkage mechanism. Teachers cannot automatically assign scores based on preset rules such as student seating positions, and there is also a lack of an intuitive and efficient user interface centered on a graphical seating chart to seamlessly record classroom performance.

[0006] Therefore, there is an urgent need in this field for an integrated solution that can simultaneously verify identity and location with low cost and high reliability, and automatically convert this data into formative assessment scores, in order to solve problems such as proxy signing vulnerabilities, fragmented data collection processes, and disconnection from teaching evaluation systems in existing technologies. Summary of the Invention

[0007] To address this, embodiments of the present invention provide a method and system for classroom attendance and formative assessment based on image recognition and seat reservation, which solves the problems of proxy attendance loopholes, inability to verify actual seat positions, and isolation between attendance data and teaching evaluation in the prior art.

[0008] To address the aforementioned technical problems, embodiments of the present invention provide a method for classroom attendance and formative assessment based on image recognition and seat reservation, the method comprising the following steps: The teacher's end is pre-set with a graphical virtual seating chart and classroom attendance rules, including differentiated scoring rules based on seat location; After authorizing login on the student's end, select the target seat on the graphical virtual seating chart to trigger check-in; The system verifies students' sign-in requests, including for students who are not in the first batch of sign-in, it initiates AI cross-verification, comparing their uploaded images with historical uploaded images of students whose seats are adjacent to the target students who have already signed in, and performing a dual comparison of facial features and background environment features. Once the verification is successful, the system will automatically add the corresponding points to the student's account based on the seat attribute that the student ultimately occupies. The system records attendance data, points data, and classroom-related behavior data in real time for formative assessment analysis.

[0009] Preferably, the operation of triggering check-in further includes: the student terminal calling the local image acquisition device to take an image containing their own face and the surrounding environment and uploading it to the system. Preferably, the AI ​​cross-validation specifically includes: Extract the facial feature vector V1 and the background environment feature point set P1 from the image of the student currently checking in; Extract the facial feature vector V2 and the background environment feature point set P2 from the images of students who have checked in at adjacent seats; Calculate the cosine similarity Sf between the face feature vectors V1 and V2, and calculate the number of matching feature points Nb between the background environment feature point sets P1 and P2. When the cosine similarity Sf exceeds the first preset threshold and the number of matched feature points Nb exceeds the second preset threshold, the AI ​​verification is deemed successful.

[0010] Preferably, if the AI ​​cross-validation fails, the system pushes a verification failure message to the student and provides an entry point for manual review application. After the student submits the application, the review request will be pushed to the preset review end.

[0011] Preferably, the review authority of the review terminal can be configured and granted to student cadre accounts by the teacher terminal, forming a hierarchical review mechanism.

[0012] Preferably, the seat attributes include the number of rows, columns, or area numbers of the seats, and seats with different numbers of rows, columns, or area numbers correspond to different preset score values; the differentiated scoring rules can be customized and modified by the teacher before class.

[0013] Preferably, the method further includes: the teacher can preset a threshold for the number of students signing in for the first batch or a time threshold. For students who meet the threshold conditions, the system processes them according to a preset strategy, which includes "default verification passed" or "default entry into the pending review state".

[0014] Preferably, throughout the teaching process, the teacher continuously displays the graphical virtual seating chart. By clicking on the corresponding student's seat on the seating chart, the teacher can quickly add classroom performance records and adjust additional points.

[0015] Preferably, the system's verification of student attendance requests also includes concurrent conflict control: an optimistic locking mechanism is used to lock the target seat, and if a seat is detected to be occupied simultaneously, a conflict prompt is returned to the subsequent requesting students and the status of the conflicting seats is marked.

[0016] This invention also provides a classroom attendance and formative assessment system based on image recognition and seat reservation. This system implements the aforementioned classroom attendance and formative assessment method based on image recognition and seat reservation, including: Seating Chart Management Module: Used to generate, render, and update graphical virtual seating charts in real time; Sign-in request processing and concurrency control module: used to receive student sign-in requests and handle seat concurrent occupancy conflicts through an optimistic locking mechanism; Image recognition and AI verification module: used to extract facial features and background environment features from uploaded images and execute AI cross-validation algorithms; The access control and review workflow module is used to configure hierarchical review permissions and handle the creation, push, and decision feedback of manual review work orders. Scoring rule engine and data statistics module: used to store and execute differentiated scoring rules, and aggregate attendance, points, and classroom performance data.

[0017] This invention also provides an electronic device, which includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the image recognition and seat reservation classroom attendance and formative assessment method described above.

[0018] This invention also provides a computer storage medium storing a computer software product, the computer software product including several instructions to cause a computer device to execute the above-described image recognition and seat reservation classroom attendance and formative assessment method.

[0019] As can be seen from the above technical solutions, this invention application has the following beneficial effects: (1) Traditional check-in methods (such as QR codes and GPS) are easily subject to remote proxy check-in, while simple facial recognition cannot verify whether a student is actually in the designated seat. This invention uses a "photo + AI cross-verification" mechanism, requiring students to upload an image containing their face and the surrounding environment, and then compare the facial features and background features of adjacent students who have already checked in. This method not only verifies the student's identity but also verifies their actual location through background consistency, effectively preventing cheating behaviors such as "remote proxy check-in" and "pre-uploading photos," and improving the authenticity and credibility of check-in data.

[0020] (2) Traditional attendance tracking and classroom performance recording are often separate processes, requiring teachers to manually record and compile data, which is inefficient. This invention uses a graphical virtual seating chart as the core interactive interface, seamlessly connecting seat selection, attendance verification, point assignment, and classroom performance recording. The system can automatically assign corresponding points to students according to preset differentiated scoring rules, and teachers can also quickly record classroom performance by clicking on seat grids. All data is automatically aggregated, supporting the generation of multi-dimensional visual reports, greatly reducing the statistical burden on teachers and realizing the automation and datafication of formative assessment.

[0021] (3) Existing hardware-dependent solutions (such as face recognition terminals and Bluetooth beacons) are costly and complex to deploy. This invention makes full use of students' own smartphones and cameras, combined with WeChat mini-programs as the front-end carrier, without the need for additional hardware investment. The system adopts a microservice architecture, with the AI ​​verification service deployed independently, supporting horizontal expansion and balancing performance and cost. At the same time, the system supports a hierarchical review mechanism (teachers can authorize class committee members to participate in the review), enhancing the system's flexibility and practicality, and making it suitable for classroom teaching scenarios of different sizes. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Referring to the drawings will make the features and advantages of the present invention clearer. The drawings are illustrative and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart of a classroom attendance and formative assessment method based on image recognition and seat reservation provided by the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention; Figure 3 This is a schematic diagram of the method flow for presetting a graphical virtual seating chart and check-in rules according to the present invention; Figure 4 This is a schematic diagram of the method for selecting a target seat on a graphical virtual seating chart to trigger check-in according to the present invention; Figure 5 This is a schematic diagram of the method for adding corresponding points to a student's account according to the present invention; Figure 6 This is a block diagram of a classroom attendance and formative assessment system based on image recognition and seat reservation provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1: To address the issues of proxy signing loopholes, inability to verify actual seating positions, and the isolation between attendance data and teaching evaluation processes in existing technologies, such as... Figure 1 As shown, this invention proposes a method for classroom attendance and formative assessment based on image recognition and seat reservation. This method includes the following steps: S1: The teacher's end is pre-set with a graphical virtual seating chart and classroom attendance rules, including differentiated scoring rules based on seat location; S2: After authorizing login on the student's end, select the target seat on the graphical virtual seating chart to trigger check-in; S3: The system verifies students' sign-in requests, including for students who are not in the first batch of sign-in, it initiates AI cross-verification, and performs a dual comparison of facial features and background environment features with the historically uploaded images of students whose seats are adjacent to the target students who have already signed in. S4: After verification, the system will automatically add the corresponding points to the student's account based on the seat attribute that the student ultimately occupies. S5: The system records attendance data, points data, and classroom-related behavior data in real time for formative assessment analysis.

[0025] As can be seen from the above technical solution, this invention proposes a classroom attendance and formative assessment method based on image recognition and seat reservation. First, the teacher's end presets a graphical virtual seating chart and attendance rules including differentiated scoring, realizing the visualization and flexible customization of teaching incentive strategies. Next, students log in with authorization and select a target seat on the seating chart to trigger attendance, transforming the abstract attendance into an intuitive "seat reservation" behavior. The operation is natural and introduces anti-cheating data from the source. The system then verifies the attendance request, especially for students not in the first batch, by initiating AI cross-validation, comparing their uploaded on-site images with adjacent signed-in students. The system performs dual facial and background feature comparison on students' historical images, achieving highly reliable dual verification of identity and spatial location without the need for specialized hardware, effectively preventing proxy signing. After successful verification, the system automatically adds corresponding points to the student's account based on the seat attribute they successfully occupied, completing the automated association from "seat selection" to "incentive feedback" and accurately guiding classroom behavior. Finally, the system records and integrates attendance, points, and classroom behavior data in real time, providing a multi-dimensional and traceable data foundation for formative assessment and realizing a complete closed loop of teaching management from data collection, analysis, to feedback.

[0026] This invention's image recognition and seat reservation-based classroom attendance and formative assessment system adopts a front-end / back-end separation and microservice hybrid architecture. Through modular design, it achieves high cohesion and low coupling, balancing reliability, flexibility, and low cost. Its overall architecture corresponds to the attached... Figure 2 The specific technology selection and implementation plan are as follows: 1. Front-end layer (teacher / student end) The technology chosen is WeChat Mini Program, and in terms of specific implementation, it adopts... <view>The tag-based CSS Flex / Grid layout dynamically renders graphical seating charts, while integrating WeChat's identity system for one-click authorization login. Its core advantages include: no additional installation required, cross-platform compatibility with iOS / Android, and low barriers to entry; natural association with user identity, simplifying the login process; and the ability to adapt to mobile image capture and upload operations via WeChat Mini Program APIs (such as wx.chooseMedia and wx.uploadFile), making it more convenient to use.

[0027] 2. Backend business layer The technology chosen is Java Spring Boot + WeChat Cloud development. The specific implementation is divided into two parts: the core business (user management, classroom management, and attendance process) is built on Spring Boot using a RESTful API; the lightweight business (image storage and simple query) is implemented using Node.js cloud functions. The core advantages of this solution are: Spring Boot supports high concurrency processing, which can meet the needs of simultaneous attendance in multiple classrooms; cloud functions simplify the deployment and maintenance process and reduce server costs; and the hybrid architecture can balance system performance and development efficiency.

[0028] 3. AI Algorithm Layer The chosen technology is a Python Flask microservice, specifically an independently deployed image recognition service that communicates with the main business layer via HTTP API, focusing on the task of comparing face and background features. Its core advantages are: decoupling computationally intensive tasks from core business, avoiding single points of failure; facilitating algorithm iteration and upgrades (such as replacing feature extraction models) without affecting the overall system; and supporting horizontal scaling to meet the concurrent verification needs of large-scale classrooms.

[0029] 4. Data Persistence Layer The technology selection is MySQL + Redis + cloud storage. Specifically, MySQL is used to store structured data (user table, class table, seat table, points table, etc.); Redis caches real-time seat status and hot data, and supports Pub / Sub message push; Tencent Cloud OSS / WeChat Cloud Storage is used to save attendance images. The core advantages of this solution are: MySQL ensures data consistency and meets transaction requirements such as seat locking; Redis caching reduces database pressure and improves seat status synchronization speed; cloud storage provides highly available image storage capabilities and supports the expansion of massive amounts of data.

[0030] 5. Communication Protocol The technology chosen is HTTPS + WebSocket. Specifically, regular requests (attendance applications, rule queries) are transmitted using HTTPS encryption; real-time data (seat status changes, approval work orders) are pushed via WebSocket. Its core advantages are: HTTPS ensures the secure transmission of user identity information and image data, avoiding the risk of privacy leaks; WebSocket enables proactive server-side pushing, ensuring that the status of teachers and students can be synchronized in real time (such as immediate notifications after a seat is claimed).

[0031] This invention uses "pre-class planning - in-class attendance verification - in-class performance recording - post-class data statistics" as its core process, combined with... Figure 1 The technical solutions and advantages of each step are as follows: In step S1, the teacher's end presets a graphical virtual seating chart and attendance rules, including differentiated scoring rules based on seat location, such as... Figure 3 As shown, it specifically includes: S11: After logging into the mini-program, teachers can enter the "Classroom Management" module and choose any of the following methods to create a classroom: Method 1: Manually create a class: Teachers click "Create Classroom" and manually fill in basic information such as course name, class time, and attendance window (e.g., 15 minutes before class to 5 minutes after class). Method 2: Import courses from the academic affairs system or timetable: Teachers can click "Import from Timetable" or "Synchronize Academic Data," and the system will provide one or a combination of the following two import methods: File import: Supports uploading standard format timetable files (such as Excel and CSV). After the system parses the file, it will automatically fill in fields such as course name, class time, and classroom. Interface synchronization: If the school has opened the academic affairs system interface, teachers can log in to the academic affairs system through authorization. The system will automatically pull the list of courses taught by the teacher this semester. After the teacher selects the corresponding course, the system will automatically fill in the course information.

[0032] After selecting the import method, the system automatically fills in basic information such as course name and class time. Teachers can then further set personalized parameters such as attendance time window, seating chart, and grading rules.

[0033] S12: Enter the "Seating Table Settings" interface, and configure the number of rows (e.g., 5 rows) and columns (e.g., 6 columns) of seats by dragging the slider. The system will automatically generate a grid-like graphical virtual seating table, and each seat will be assigned a unique identifier (e.g., "A-3-2" means seat 2 in row 3 of section A). S13: Enter the "Assignment Rules Settings" interface. Teachers can enter the corresponding points by clicking the row number / column number (e.g., first row +3 points, second row +2 points, edge area +1 point). The rules are stored in JSON format (example below). { "pointRules": [ {"target": "seat", "condition": {"row": [1]}, "action": {"type":"add", "points": 3}}, {"target": "seat", "condition": {"row": [2]}, "action": {"type":"add", "points": 2}}, {"target": "behavior", "condition": {"type": "quick_response"}, "action": {"type": "add", "points": 1}} ] } S14: Advanced Settings: Teachers can set a threshold for the number of students signing in for the first time (e.g., the first 5 students) or a time threshold (e.g., the first 3 minutes) and select the default processing strategy ("Default Pass" or "Default Pending Review").

[0034] This step has the following advantages: 1. The graphical seating chart intuitively reproduces the physical layout of the classroom, solving the problem of lack of spatial awareness in existing list-based sign-in technology; 2. The differentiated scoring rules can be customized to accurately match teachers' teaching incentive needs, making it more flexible than the fixed score model; 3. A pre-defined strategy for handling the first batch of students is developed, balancing attendance efficiency (rapid passage for the first batch) with verification rigor (subsequent AI cross-verification), and balancing the teaching pace with the need to prevent proxy attendance.

[0035] In step S2, after authorizing login, the student selects a target seat on the graphical virtual seating chart to trigger check-in, such as... Figure 4 As shown, it specifically includes: S21: Students log in to the mini-program via WeChat authorization, view the list of created classes, click to enter the check-in interface of the corresponding class, and the interface displays a graphical seating chart. Empty seats are marked as clickable, and occupied / conflicted seats are marked in blue / red respectively. S22: When a student clicks on a target empty seat, the mini-program automatically calls the wx.chooseMedia API to activate the phone's camera, prompting the student to take a photo of themselves and the surrounding environment (such as the neighboring desks and chairs, a corner of the blackboard). After the student confirms, the photo, class ID, seat ID, and student ID are uploaded to the server via wx.uploadFile. S23: If a student clicks on a seat that has already been taken, the front end will immediately display a "seat already taken" message, guiding the student to select a new seat.

[0036] This step has the following advantages: 1. The integrated "seat reservation + photo taking" operation transforms the abstract sign-in process into a concrete seat selection behavior, which aligns with students' classroom habits and has a low barrier to entry. 2. Forced collection of images containing human faces and the environment provides dual data support for subsequent AI verification, avoiding vulnerabilities such as remote sharing of QR codes and proxy signing of mobile phones from the source; 3. Seat status is visualized in real time, allowing students to intuitively view the remaining seats, reducing conflicts caused by blindly grabbing seats and improving the check-in experience.

[0037] In step S3, the system verifies the student's attendance request, specifically including: S31: Concurrency Conflict Control (Optimistic Locking Mechanism) After receiving the check-in request, the server executes the following logic within a database transaction (pseudocode below): @Transactional public SignInResult handleSignInRequest(SignInRequest request) { / / 1. Use a lock to query the current seat status to prevent false positives caused by concurrent queries. Seat seat = seatMapper.selectForUpdate(request.getSeatId()); / / 2. Check if the seats are available if (seat.getStatus() != SeatStatus.AVAILABLE) { return SignInResult.conflict("The seat has been taken"); } / / 3. Update the seat status to "Occupied - Pending Approval", and simultaneously update the optimistic locking version number. seat.setStatus(SeatStatus.OCCUPIED_PENDING); seat.setStudentId(request.getStudentId()); int updatedRows = seatMapper.updateWithVersion(seat); / / WHERE condition includes the old version number / / 4. If the version number does not match, it means the seat has been taken by someone else, and a conflict message will be returned. if (updatedRows == 0) { return SignInResult.conflict("Seat claim failed, please try again"); } / / 5. Conflict resolution complete, proceed to verification process return proceedToVerification(request, seat); } Each seat object contains a lockVersion property (initially 0), which increments with each update. The update is successful only if the version number in the request matches the version number in the database.

[0038] This step has the following advantages: 1. Compared to pessimistic locking, optimistic locking does not require long-term locking of database resources, supports high-concurrency scenarios (such as multiple students vying for the same seat at the same time), and has a fast response speed; 2. In-transaction atomic operations (query-update) ensure data consistency and avoid problems such as "one person occupying multiple seats" and "seat status confusion"; 3. The front end provides real-time feedback on conflict results, allowing students to quickly reselect without affecting overall attendance efficiency.

[0039] S32: AI Cross-Validation (Dual Comparison of Face and Background) For students not in the first batch of sign-in, AI cross-validation is initiated, comparing their uploaded images with historically uploaded images of students whose seats are adjacent to the target students who have already signed in, using both facial features and background environmental features. Specifically, this includes: S321: The system first determines whether the student is among the "first batch of sign-ins" (meets the quantity / time threshold): if yes, it processes the student according to the preset strategy (default verification passes or default entry into the pending review state); if no, it initiates AI cross-validation. S322: Feature Extraction Facial features: The AI ​​service calls the MTCNN model to detect and align the facial regions in the photo, and then uses the Facenet model to convert the facial image into a 128-dimensional feature vector (current student's facial feature vector V1, adjacent checked-in student's facial feature vector V2). Background features: The SIFT / ORB algorithm is used to extract scale-invariant feature points and descriptors of the environment in the photo (the current student's background environment feature point set P1 and the adjacent student's background environment feature point set P2). S323: Cross-alignment: Face matching: Calculate the cosine similarity Sf between face feature vectors V1 and V2, set a threshold T1=0.65, and if Sf>T1, the face is considered to match (confirmed to be a real student in the classroom). Background comparison: The feature descriptors of the background environment feature point sets P1 and P2 are matched using the FLANN matcher. False matches are removed using the RANSAC algorithm. The number of stable matching points Nb is counted. The threshold T2=15 is set. If Nb>T2, the background is considered to be consistent (confirmed to be the same classroom scene). S324: Comprehensive decision: AI verification passes only if Sf>T1 and Nb>T2; otherwise, verification fails.

[0040] This step has the following advantages: 1. Innovative "face + background" dual verification mode. Compared with simple face recognition (which only verifies identity), it adds spatial consistency verification, completely eliminating vulnerabilities such as "remote photo signing" and "pre-taken photos for uploading". 2. Employing mature algorithms such as MTCNN, Facenet, and SIFT, balancing recognition accuracy and computational efficiency, and considering the limited computing power of students' mobile devices, centralized processing is achieved through microservices on the server side, balancing cost and performance. 3. The threshold can be dynamically adjusted (e.g., T1 and T2 can be optimized according to classroom size and lighting conditions) to adapt to different teaching scenarios.

[0041] S33: Tiered Review Workflow In response to the AI ​​verification failure in step S32, this invention employs a tiered review mechanism for manual review of student requests, specifically including: S331: If AI verification fails, the student's end will display a "Verification failed" page and provide a "Apply for manual review" button. After the student clicks, the system will create a review ticket (stored in the approval_tickets table, with fields including class ID, student ID, seat ID, image URL, creation time, etc.). S332: Permission Configuration: Teachers can promote some students to "Class Committee (monitor)" in "Class Settings". The role field of the user table is marked as teacher / monitor / student. The backend approval interface only allows teachers or monitors to access the system. S333: Review process: Teachers / class committee members view work orders in the "Pending Review" module of the management terminal, click to view the photos uploaded by students, and make a "pass" or "reject" decision. The decision result is pushed to the student's terminal in real time via WebSocket. When rejecting, the reason must be filled in (such as "unclear face" or "inconsistent background").

[0042] This step has the following advantages: 1. Decentralize authority at different levels, assign some review work to class committees, reduce the burden on teachers of handling a large number of work orders per class period, and improve review efficiency; 2. Human review serves as a supplement to AI verification, handling special scenarios (such as students looking down or their faces being obscured, leading to AI misjudgments) to ensure the fairness of the verification process; 3. Implement a work order-based management and review process to record the review process, facilitating subsequent traceability and avoiding disputes.

[0043] In step S4, after successful verification, the system automatically adds corresponding points to the student's account based on the seat attributes that the student ultimately occupied. Figure 3 As shown, it specifically includes: S41: After the system confirms that the verification is successful (either AI verification or manual review), it reads the scoring rules corresponding to the seat from the database (e.g., 2 points for seat 2 in row 3). S42: The scoring rule engine calls the points_record table to insert score records, and associates them with student ID, class ID, score value, and scoring reason (e.g., "seat A-3-2 scoring"). S43: The student's end will display "Sign-in successful! Seat: A-3-2 Points: +2" page in real time, and the corresponding seat on the teacher's end will be updated to "Sign-in + Points" status.

[0044] This step has the following advantages: 1. Seat selection is automatically linked to points, eliminating the need for manual operation by teachers and solving the problems of existing technology that separates attendance and point accumulation, resulting in cumbersome processes; 2. Real-time feedback of points enhances the incentive effect, guides students to actively choose high-quality seats such as the front row, and improves classroom participation; 3. The points record is traceable, providing objective data support for subsequent formative assessment.

[0045] In step S5, the system records attendance data, points data, and classroom-related behavioral data in real time for formative assessment analysis, specifically including: S51: Quick Record of Classroom Performance During Lesson The system records attendance data, points data, and classroom-related behavior data in real time, specifically including: S511: Throughout the teaching process, the teacher continuously displays a graphical seating chart, with each seat marked with the student's name and current score; S512: When teachers observe positive behaviors (such as actively answering questions and quickly completing exercises) or negative behaviors (such as sleeping or playing on their phones) in students, they can click on the corresponding seat cell to bring up a quick operation menu (such as "+1 point (answer first)," "-0.5 points (distracted)," "Record performance: actively participate"). After the teacher selects a position, the system will update the score and performance record immediately. S513: Performance records are automatically associated with seat and student identity and stored in the class_performance table, supporting filtering and querying by time and behavior type.

[0046] This step has the following advantages: 1. The seating chart serves as a unified operating interface, eliminating the need for teachers to switch between lists to find students. The operation takes less than 1 second and does not interrupt the teaching rhythm. 2. Performance records are linked with sign-in and points data to form a three-dimensional data chain of "attendance-seat-performance," solving the problem of scattered and difficult-to-integrate data in existing technologies; 3. Supports customizable performance tags to adapt to the teaching needs of different subjects (such as "clear problem-solving ideas" for science subjects and "novel viewpoints" for humanities subjects), offering high flexibility.

[0047] S52: Post-Class Multi-Dimensional Data Statistics and Report Generation The process of statistically analyzing and generating visualizations from the recorded multi-dimensional data includes, specifically: S521: After class, the teacher clicks the "Data Analysis" module, and the system automatically executes a predefined SQL query to aggregate the following data: Basic data: attendance rate (actual number of attendees / expected number of attendees), lateness rate, and seat selection distribution (percentage of attendees in the front row, back row, and middle area). Points data: individual student points ranking, points trend for a single class, and average points for each seating area; Performance data: frequency of various classroom behaviors (answering questions quickly, daydreaming, asking for leave, etc.), and student performance heatmap; S522: The system encapsulates the data into ECharts chart configuration items, and the front end renders a visual report, including line charts (personal points trend), heatmaps (seat participation), and tables (performance details). S523: Teachers can click the "Export" button, and the system backend will generate an Excel file from the data, which will be provided to the teacher through the mini-program download interface. It can be imported into the teaching affairs system as a basis for formative assessment.

[0048] This step has the following advantages: 1. Data is automatically aggregated, eliminating the need for teachers to manually compile statistics and significantly reducing the burden of teaching management; 2. Multi-dimensional visualization presentation, which is more intuitive than pure text reports, helps teachers quickly grasp the classroom status (such as whether the participation of front-row seats is higher than that of back-row seats). 3. The data supports export and traceability, meeting the archiving requirements of the academic affairs department for formative assessment, and providing data support for teachers to adjust teaching strategies (such as optimizing seating arrangements and enhancing interaction among students in the back row).

[0049] Example 2: Figure 6 As shown, this invention provides a classroom attendance and formative assessment system based on image recognition and seat reservation. This system is used to implement the classroom attendance and formative assessment method based on image recognition and seat reservation described in Embodiment 1 above, specifically including: Seating Chart Management Module: Used to generate, render, and update graphical virtual seating charts in real time; Sign-in request processing and concurrency control module: used to receive student sign-in requests and handle seat concurrent occupancy conflicts through an optimistic locking mechanism; Image recognition and AI verification module: used to extract facial features and background environment features from uploaded images and execute AI cross-validation algorithms; The access control and review workflow module is used to configure hierarchical review permissions and handle the creation, push, and decision feedback of manual review work orders. Scoring rule engine and data statistics module: used to store and execute differentiated scoring rules, and aggregate attendance, points, and classroom performance data.

[0050] In this embodiment, the seating chart management module is used to generate, render, and update the graphical virtual seating chart in real time, specifically including: 1. Generation and Rendering: The front-end dynamically creates an N-row, M-column array using JavaScript. <view>Each seat corresponds to an object (with the following properties), and the status is visualized by binding the status property in CSS (free: white, occupied: blue, pending: yellow, conflict: red). { seatId: "A-3-2", / / Unique identifier status: 0, / / 0-Idle, 1-Occupied, 2-Pending review, 3-Conflict studentId: null, / / Reserving a student ID lockVersion: 0 / / Optimistic locking version number } 2. Data Synchronization: When the seat status changes (e.g., occupied, approved), the server publishes a status update message through Redis Pub / Sub, and WebSocket pushes the message to all online teachers and students. The front end receives the message and updates the local seat table rendering. 3. Template Management: Supports teachers to save commonly used seating chart templates (such as 5×6, 6×8), which can be reused directly when creating classes later without repeated configuration.

[0051] This module has the following advantages: 1. Dynamic rendering and real-time synchronization ensure that all users see the same seat status, avoiding conflicts caused by information asymmetry; 2. Template-based management improves teacher allocation efficiency and adapts to different classroom sizes (e.g., small classes with 30 students and large classes with 60 students). 3. Based on the WeChat Mini Program front-end implementation, it is compatible with mobile devices such as mobile phones and tablets, allowing teachers to operate it while moving around the podium and classroom, providing high flexibility.

[0052] In this embodiment, the attendance request processing and concurrency control module is used to receive student attendance requests and handle seat concurrent occupancy conflicts through an optimistic locking mechanism, specifically including: 1. Request Reception: Spring Boot's @RestController is used to receive the sign-in requests uploaded by students, and image files are received via MultipartFile, stored in cloud storage, and then the URL is returned. 2. Concurrency control: As described in step S31, seat preemption conflicts are handled through an optimistic locking mechanism, and database transactions ensure the atomicity of operations; 3. Timeout Handling: If a student uploads a photo but fails to complete the verification (e.g., AI verification timeout), the system will automatically reset the seat status to "vacant" to release the seat and prevent it from being locked for a long time.

[0053] This module has the following advantages: 1. High concurrency support: Spring Boot's default thread pool configuration can handle 500 people signing in at the same time in a single class. With Redis caching, the response latency is less than 500ms. 2. Automatic release mechanism upon timeout to avoid wasting seat resources in abnormal scenarios; 3. In conjunction with cloud storage, image files are stored first and then verified, reducing database storage pressure and improving request processing speed.

[0054] In this embodiment, the image recognition and AI verification module is used to extract facial features and background environment features from the uploaded image and execute an AI cross-validation algorithm, specifically including: 1. Microservice deployment: Build an independent API service (POST / api / verify) based on Python Flask to receive a list of image URLs (current student image + adjacent student images) sent by the main business layer. 2. Algorithm optimization: Face feature extraction: The MTCNN model optimizes the face detection box to adapt to different angles (such as looking down or looking sideways), and the Facenet model outputs a 128-dimensional vector to ensure feature discrimination. Background feature extraction: The SIFT algorithm is robust to changes in lighting and scale, making it suitable for complex classroom environments (such as changes in lighting and the arrangement of desks and chairs at different times). 3. Result Feedback: AI verification results (pass / fail) and key data (cosine similarity Sf, number of matching points Nb) are synchronized to the main business layer for recording and review.

[0055] This module has the following advantages: 1. Microservice architecture facilitates independent scaling. When multiple classrooms are conducting AI verification simultaneously, service instances can be added horizontally to avoid performance bottlenecks. 2. The algorithm is robust and adaptable to complex classroom scenarios, reducing the false judgment rate (actual testing shows an accuracy rate of >95%). 3. No additional dedicated hardware (such as facial recognition terminals) is required. By utilizing students' mobile phone cameras and server computing power, hardware costs are significantly reduced.

[0056] In this embodiment, the permission management and review workflow module is used to configure hierarchical review permissions, handle the creation, push, and decision feedback of manual review work orders, specifically including: 1. Access Control: Based on the RBAC (Role-Based Access Control) model, the role field in the user table distinguishes between teacher / monitor / student, and the backend verifies the access permissions of the interface through interceptors (e.g., the audit interface is only allowed to be called by teacher / monitor). 2. Work order processing: The status of the work order is divided into "Pending review (0)", "Approved (1)" and "Rejected (2)". Teachers can filter the work orders by status and support batch processing (such as batch approval of uncontroversial work orders). 3. Message push: When a new work order is created, a reminder (such as "There is 1 new review request") is pushed to the teacher / class committee via WebSocket to ensure timely processing.

[0057] This module has the following advantages: 1. Refined access control ensures data security (students cannot access the review function) while improving review efficiency (class committee members share the workload). 2. Work order status is visualized, allowing teachers to monitor the review progress in real time and avoid missed reviews; 3. The push notification mechanism reduces manual refresh operations and improves the review response speed (average review time < 30 seconds).

[0058] In this embodiment, the scoring rule engine and data statistics module are used to store and execute differentiated scoring rules, and aggregate attendance, points, and classroom performance data, specifically including: 1. Scoring rule engine: Rules are stored in JSON format and support multiple condition combinations (such as "front row + early arrival" for additional points). The rule engine executes the scoring logic by parsing JSON and supports dynamic modification (teachers can adjust the point values ​​in class and they will take effect immediately). 2. Data Statistics: Individual points trend: Aggregated by week using the query SELECT student_id, WEEK(create_time), SUM(points) FROM points_record GROUP BY student_id, WEEK(create_time); Seating heatmap: The seating chart is divided into regions, the average integral of each region is calculated, and mapped to grayscale values ​​(the higher the integral, the darker the color), and then rendered using ECharts; 3. Report Export: Generates Excel files based on Apache POI, including three main worksheets: attendance details, points details, and performance records, and supports custom export fields.

[0059] This module has the following advantages: 1. The rule engine flexibly adapts to different teaching needs, and the scoring strategy can be adjusted without modifying the code; 2. Statistical SQL pre-compilation optimization results in fast query speed (query time for millions of data points is less than 1 second); 3. It supports both visual reports and Excel export, meeting the dual needs of teachers for daily viewing and academic affairs archiving.

[0060] This embodiment provides a classroom attendance and formative assessment system based on image recognition and seat reservation, used to implement the aforementioned classroom attendance and formative assessment method based on image recognition and seat reservation. Therefore, the specific implementation of the classroom attendance and formative assessment system based on image recognition and seat reservation can be found in the previous section on the implementation of the classroom attendance and formative assessment method based on image recognition and seat reservation. To avoid redundancy, it will not be repeated here.

[0061] Example 3: This embodiment of the invention provides an electronic device, which includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to realize the above-mentioned image recognition and seat reservation classroom attendance and formative assessment method.

[0062] Example 4: This embodiment of the invention provides a computer storage medium storing a computer software product. The computer software product includes several instructions to cause a computer device to execute the above-described image recognition and seat reservation classroom attendance and formative assessment method.

[0063] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0064] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0065] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0066] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.< / view> < / view>

Claims

1. A method for classroom attendance and formative assessment based on image recognition and seat reservation, characterized in that, Includes the following steps: The teacher's end is pre-set with a graphical virtual seating chart and classroom attendance rules, including differentiated scoring rules based on seat location; After authorizing login on the student's end, select the target seat on the graphical virtual seating chart to trigger check-in; The system verifies students' sign-in requests, including for students who are not in the first batch of sign-in, it initiates AI cross-verification, comparing their uploaded images with historical uploaded images of students whose seats are adjacent to the target students who have already signed in, and performing a dual comparison of facial features and background environment features. Once the verification is successful, the system will automatically add the corresponding points to the student's account based on the seat attribute that the student ultimately occupies. The system records attendance data, points data, and classroom-related behavior data in real time for formative assessment analysis.

2. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, The operation to trigger check-in also includes: the student's terminal calling the local image acquisition device to take an image containing their own face and the surrounding environment and uploading it to the system.

3. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, The AI ​​cross-validation specifically includes: Extract the facial feature vector V1 and the background environment feature point set P1 from the image of the student currently checking in; Extract the facial feature vector V2 and the background environment feature point set P2 from the images of students who have checked in at adjacent seats; Calculate the cosine similarity Sf between the face feature vectors V1 and V2, and calculate the number of matching feature points Nb between the background environment feature point sets P1 and P2. When the cosine similarity Sf exceeds the first preset threshold and the number of matched feature points Nb exceeds the second preset threshold, the AI ​​verification is deemed successful.

4. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, If the AI ​​cross-validation fails, the system will send a verification failure message to the student and provide an entry point for manual review application. After the student submits the application, the review request will be sent to the preset review end.

5. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 4, characterized in that, The review permissions of the review terminal can be configured by the teacher and granted to student cadre accounts, forming a hierarchical review mechanism.

6. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, The seat attributes include the number of rows, columns, or area numbers of the seats. Seats with different numbers of rows, columns, or area numbers correspond to different preset score values. The differentiated scoring rules can be customized and modified by the teacher before class.

7. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, The method further includes: the teacher can preset a threshold for the number of students who sign in for the first time or a time threshold. For students who meet the threshold conditions, the system processes them according to a preset strategy, which includes "default verification passed" or "default entry into the pending review state".

8. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, Throughout the teaching process, the teacher's end continuously displays the graphical virtual seating chart. By clicking on the corresponding student's seat on the seating chart, the teacher can quickly add classroom performance records and adjust additional points.

9. The image recognition and seat reservation method for classroom attendance and formative assessment according to claim 1, characterized in that, The system's verification of student attendance requests also includes concurrent conflict control: an optimistic locking mechanism is used to lock the target seat, and if a seat is detected to be occupied simultaneously, a conflict prompt is returned to the subsequent requesting students and the status of the conflicting seats is marked.

10. A classroom attendance and formative assessment system based on image recognition and seat reservation, characterized in that, The system is used to implement the image recognition and seat reservation classroom attendance and formative assessment method according to any one of claims 1 to 9, including: Seating Chart Management Module: Used to generate, render, and update graphical virtual seating charts in real time; Sign-in request processing and concurrency control module: used to receive student sign-in requests and handle seat concurrent occupancy conflicts through an optimistic locking mechanism; Image recognition and AI verification module: used to extract facial features and background environment features from uploaded images and execute AI cross-validation algorithms; The access control and review workflow module is used to configure hierarchical review permissions and handle the creation, push, and decision feedback of manual review work orders. Scoring rule engine and data statistics module: used to store and execute differentiated scoring rules, and aggregate attendance, points, and classroom performance data.

11. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a bus system. The processor and the memory are connected through the bus system. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory to implement the image recognition and seat reservation classroom attendance and formative assessment method according to any one of claims 1 to 9.

12. A computer storage medium, characterized in that, The computer storage medium stores a computer software product, which includes several instructions to cause a computer device to execute the image recognition and seat reservation classroom attendance and formative assessment method according to any one of claims 1 to 9.

Images