Multi-mode operation locking method and system for electronic whiteboard

Abstract

The invention relates to the technical field of electronic whiteboard locking, in particular to a multi-mode operation locking method and system for an electronic whiteboard, which realize the function of automatically matching a locking mode by initializing the electronic whiteboard and loading a driving program and combining user identity verification and neural network identification on the current use scene. According to the method, the locking mode is dynamically adjusted according to the use habit of the user and the specific scene, and unlocking is performed after the unlocking request of the user is captured, so that the operation convenience and safety of the electronic whiteboard in multiple scenes are remarkably improved, and the troubles of misoperation and frequent unlocking are effectively reduced. Meanwhile, according to the scheme, the self-adaptive adjustment of the locking mode is realized in combination with the use habit data of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic whiteboard locking, and in particular to a method and system for locking multi-mode operation of an electronic whiteboard. Background Art

[0002] Electronic whiteboards have gradually become an indispensable tool in scenarios such as corporate meetings and education and training. Electronic whiteboards not only have the writing function of traditional whiteboards, but also support multimedia presentations, remote meetings, interactive teaching and other functions. However, with the widespread application of electronic whiteboards in multiple scenarios, how to ensure the security and ease of operation of electronic whiteboards in different usage scenarios has become an urgent problem to be solved.

[0003] In actual applications, electronic whiteboards often need to be quickly switched between different usage scenarios. For example, in conference mode, users may want to completely lock the electronic whiteboard to avoid misoperation during the meeting; in teaching mode, teachers may need to partially lock the whiteboard function and retain some interactive functions to meet the needs of classroom order. However, traditional electronic whiteboard locking methods usually have a single locking method and lack intelligence and adaptability. In the prior art, although some electronic whiteboards have basic locking functions, most of them are based on simple identity authentication or manual settings, and cannot achieve intelligent, multi-mode operation locking. In addition, the prior art lacks an effective method to capture the user's blocked operations in the locked state and make intelligent adjustments, and cannot dynamically adjust the lock mode according to the user's actual usage scenario. Summary of the invention

[0004] To solve the above problems, the present invention provides an electronic whiteboard multi-mode operation locking method and system, which combines user identity authentication and neural network to identify the current usage scenario, adjusts the locking mode according to the user's usage habits and specific scenarios, and realizes dynamic adjustment of the locking mode.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for locking multi-mode operation of an electronic whiteboard comprises the following steps:

[0007] Initialize the electronic whiteboard and load the lock driver;

[0008] Verify user identity and identify current usage scenarios through neural networks;

[0009] Match the electronic whiteboard lock mode according to the current usage scenario and user usage habit data;

[0010] Capture the user's unlock request and unlock the electronic whiteboard.

[0011] Furthermore, the initialization of the electronic whiteboard includes initializing a processor, a touch sensor, a button interface and a display screen of the electronic whiteboard.

[0012] Furthermore, verifying the user identity includes:

[0013] Providing identity verification methods, including password verification, face recognition verification, fingerprint recognition verification and voice verification;

[0014] Perform comparison and verification through identity verification;

[0015] Load user usage habit data based on the verification results.

[0016] Furthermore, the training step through the neural network includes:

[0017] Get the task image with the highest screen ratio of the electronic whiteboard;

[0018] Annotating the task image with usage scenarios, where the usage scenarios include a conference mode and a teaching mode;

[0019] The task image with completed scene annotation is used as the input of the convolutional neural network, and the usage scenario is output through back-propagation training.

[0020] Furthermore, the user usage habit data includes operations during screen lock and operations after unlock, and the electronic whiteboard lock mode includes button lock, touch lock and full lock, and the full lock is to trigger button lock and touch lock at the same time.

[0021] Furthermore, matching the electronic whiteboard lock mode according to the current usage scenario and the user usage habit data includes:

[0022] If the power state of the electronic whiteboard is not in standby mode, matching of the electronic whiteboard lock mode is performed;

[0023] If the power state of the electronic whiteboard is standby, only the power button unlocking is triggered.

[0024] Furthermore, the matching of the electronic whiteboard lock mode includes:

[0025] Generate operation vectors for operations during screen lock and operations after screen unlock within a preset time period;

[0026] Calculate the cosine similarity between the operation vector during screen lock and the operation vector after screen unlock, and determine the blocked operation based on the cosine similarity;

[0027] According to the current usage scenario, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard, the electronic whiteboard lock mode is matched through the random forest model.

[0028] Further, the matching of the electronic whiteboard lock mode by the random forest algorithm includes:

[0029] The usage scenarios, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard are cleaned and standardized into feature vectors, and then integrated to generate a feature vector set;

[0030] Use the feature vector set as input, train it through the random forest algorithm, determine the optimal model parameters through cross-validation, and build a random forest model;

[0031] The current usage scenario, real-time monitored blocked operation frequency and unlocking frequency data are input into the trained random forest model, and the electronic whiteboard lock mode is automatically matched according to the model output results.

[0032] Furthermore, the formula of the random forest model is as follows:

[0033]

[0034] in, is the final matching electronic whiteboard lock mode; Scene is the usage scenario; Freq Blocked Freq is the frequency of blocked operations; Unlock is the unlocking frequency; T i (Scene,Freq Blocked , Freq Unlock ) is the prediction result of the i-th decision tree on the usage scenario, the frequency of blocked operations and the unlocking frequency; mode is the mode of the prediction results of all decision trees; n is the number of decision trees.

[0035] Furthermore, capturing the user unlocking request and unlocking the electronic whiteboard includes:

[0036] Authentication is performed according to the user's unlocking request. If the authentication is passed, the electronic whiteboard is unlocked; if the authentication is not passed, the electronic whiteboard remains locked.

[0037] An electronic whiteboard multi-mode operation locking system, applied to any of the electronic whiteboard multi-mode operation locking methods described above, comprises a driver loading module, an identity authentication module, a remote lock matching adjustment module and a log recording module;

[0038] The driver loading module is used to initialize the electronic whiteboard and load the locking driver;

[0039] The identity verification module is used to verify the user's identity, capture the user's unlocking request and unlock the electronic whiteboard;

[0040] The remote lock matching adjustment module is used to identify the current usage scenario through a cloud server equipped with a neural network, and match the electronic whiteboard lock mode according to the current usage scenario and the user's usage habit data;

[0041] The log recording module is used to store user usage habit data, blocked operation logs and unlocking logs.

[0042] The beneficial effect of the present invention is that the present invention automatically identifies the current usage scenario (such as conference mode, teaching mode, etc.) of the electronic whiteboard through neural network technology. This process realizes accurate recognition of different scenarios through deep learning analysis of task images. Through scene recognition, the electronic whiteboard can intelligently match the corresponding lock mode to meet the operation requirements in different scenarios. For example, in conference mode, the electronic whiteboard will automatically enter a fully locked state to prevent misoperation; while in teaching mode, the system will only partially lock certain functions according to teaching needs, retaining necessary interactive functions, thereby ensuring classroom order. Intelligent matching of lock modes is performed through a random forest algorithm. Specifically, the system first generates an operation vector and calculates the cosine similarity to identify the blocked operations that the user attempted to perform but failed in the lock screen state. The frequency of these blocked operations and the unlocking frequency data are input into the random forest model, and the model dynamically adjusts the lock mode of the electronic whiteboard according to these data, thereby reducing the occurrence of blocked operations and improving the user's operation fluency. In addition, this solution also captures user usage habit data and combines these data with the current usage scenario to realize intelligent and adaptive adjustment of the lock mode, avoiding the trouble of frequently manually setting the lock mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention is a flowchart of the steps of a method for locking multi-mode operation of an electronic whiteboard.

[0044] Figure 2 It is a flow chart of the steps of matching the electronic whiteboard lock mode in the present invention.

[0045] Figure 3 It is a flowchart of the steps of matching the electronic whiteboard lock mode by using the random forest algorithm in the present invention. DETAILED DESCRIPTION

[0046] See also Figure 1-3 As shown, the present invention relates to a method for locking multi-mode operation of an electronic whiteboard, comprising the following steps:

[0047] Initialize the electronic whiteboard and load the lock driver;

[0048] Verify user identity and identify current usage scenarios through neural networks;

[0049] Match the electronic whiteboard lock mode according to the current usage scenario and user usage habit data;

[0050] Capture the user's unlock request and unlock the electronic whiteboard.

[0051] In some embodiments, first, the electronic whiteboard is initialized and the lock driver is loaded. When the electronic whiteboard is started, the system automatically initializes its hardware and software environment, including the startup of the processor, the calibration of the touch sensor, the activation of the button interface, and the brightness and resolution adjustment of the display. At this time, the system will load a dedicated lock driver, which is responsible for managing the subsequent locking and unlocking operations to ensure that the electronic whiteboard can be properly locked according to the preset rules. The lock driver contains a variety of lock mode configuration options, which will be automatically selected and applied in subsequent steps according to user needs and usage scenarios. After the initialization is completed, the system will perform a user identity authentication step. User identity authentication is achieved through a variety of biometric technologies and traditional authentication methods, including but not limited to password input, fingerprint recognition, face recognition and voice recognition. The system will call the corresponding identity authentication module to compare the authentication information provided by the user with the data stored in the system. If the comparison is successful, the system will load the personalized usage habit data related to the user. These data usually include the user's operating preferences and lock mode selection history in different scenarios, and the system will perform subsequent lock mode matching based on these data. Next, the current usage scenario of the electronic whiteboard is identified through a neural network. Specifically, the system uses the built-in camera and sensor to capture the task image currently displayed on the electronic whiteboard and inputs it into the pre-trained convolutional neural network (CNN) model for processing. The neural network extracts and analyzes the features in the image and compares the analysis results with the predefined scene classification library to determine the current usage scenario of the electronic whiteboard. Typical usage scenarios include conference mode and teaching mode. The results of scene recognition will directly affect the selection of subsequent lock modes, allowing the electronic whiteboard to intelligently adjust the operation mode according to the current actual usage environment. Finally, when the user issues an unlock request, the system will capture the request and perform the corresponding unlock operation. During this process, the system will verify the user's identity again to ensure the legitimacy of the unlock request. If the identity authentication is passed, the system will release the current lock state and restore all the operating functions of the electronic whiteboard. If the identity authentication is not passed, the system will continue to remain locked to prevent unauthorized users from operating. This multi-level security mechanism ensures the operational security of the electronic whiteboard in different usage scenarios, while also providing a high degree of flexibility to adapt to the personalized needs of users.

[0052] Furthermore, the initialization of the electronic whiteboard includes initializing a processor, a touch sensor, a button interface and a display screen of the electronic whiteboard.

[0053] In some embodiments, the processor initialization of the electronic whiteboard is first performed. The processor is the core control unit of the electronic whiteboard, which is responsible for executing all system instructions and processing user input. When the electronic whiteboard is started, a self-test is first performed to ensure that each functional module is working properly. The self-test process includes detecting the validity of the processor's clock signal, cache status, and memory address space. After the processor self-test is completed, the operating system of the electronic whiteboard will be loaded and the core driver will be started, including the locking driver. These drivers provide underlying support for the normal operation of the electronic whiteboard, ensuring that the system can respond quickly when receiving user instructions. Next, the touch sensor is initialized. The touch sensor is the main interface for the electronic whiteboard to interact with the user, and its performance directly affects the user experience. During the system startup process, the touch sensor will be calibrated to ensure that the touch position and force can be accurately identified when the user operates. The calibration process includes detecting the sensitivity and response speed of the sensor. After the calibration is completed, the touch sensor can accurately capture the user's touch operation and pass these operation data to the processor for further processing. In addition, the initialization of the touch sensor also involves the activation of the multi-touch function to support complex gesture operations, which is particularly important for multi-user interaction in conference or teaching scenarios. Then, the button interface is initialized. The button interface is an auxiliary input device of the electronic whiteboard, which is usually used to achieve quick access to specific functions, such as power switch, volume adjustment or mode switching. During the initialization process, the system detects the hardware connection status of each button to ensure that all buttons are in normal working condition physically and electrically. The system also assigns a specific function mapping to each button to ensure that when the user presses the button, the system can correctly identify and execute the corresponding command. Especially in the lock mode, the initialization of the button interface also involves the function restriction of certain buttons to prevent misoperation from affecting the normal operation of the system. For example, in the full lock mode, except for the power button, other buttons may be disabled or redefined as low-level functions to avoid misoperation. Finally, the initialization of the display. The display is an intuitive window for users to interact with the electronic whiteboard. The initialization of the display includes multiple tasks such as brightness adjustment, resolution setting and color correction. The system will first automatically adjust the brightness of the display according to the ambient light conditions to ensure clear display under different lighting conditions. Next, the system will set the resolution of the display to adapt to different application scenarios (such as high-definition image display or multimedia playback). In addition, the initialization of the display also involves color calibration to ensure color accuracy, especially when it comes to teaching content or multimedia presentations, color accuracy is critical. After initialization, the display can provide users with a high-definition, smooth visual experience, and work with the touch sensor to achieve precise user input and output interaction.

[0054] Furthermore, verifying the user identity includes:

[0055] Providing identity verification methods, including password verification, face recognition verification, fingerprint recognition verification and voice verification;

[0056] Compare and verify through identity verification;

[0057] Load user usage habit data based on the verification results.

[0058] In some embodiments, first, the system provides a variety of identity authentication methods for users to choose from. These authentication methods include password authentication, face recognition authentication, fingerprint recognition authentication, and voice authentication. Specifically, password authentication is a traditional and commonly used identity authentication method. The user needs to enter a preset password, and the system determines the legitimacy of the identity by comparing the password entered by the user with the encrypted password data stored in the database. Face recognition authentication uses the built-in camera and depth sensor of the electronic whiteboard to capture the user's facial features, and analyzes and compares them through a pre-trained deep learning model. The system will judge the user's identity based on the matching degree of facial features. This method is not only convenient but also highly secure. Fingerprint recognition authentication relies on the fingerprint sensor of the electronic whiteboard. The user presses his finger on the sensor, and the system confirms the identity by reading the fingerprint image and comparing it with the stored fingerprint data. Voice authentication uses the user's voice features for identity authentication. The system analyzes the user's voice waveform, frequency and other features through voice recognition technology, and matches them with pre-stored voice samples to achieve identity authentication. After the user selects and authenticates the identity, the system will immediately perform the comparison and verification process. Verification is achieved by accurately comparing the identity information provided by the user with the authentication data stored in the system database. In order to improve the accuracy and security of the comparison, the system will use multi-layer encryption algorithms to protect the data during the comparison process to prevent unauthorized access and data leakage. In face recognition and fingerprint recognition verification, the system uses advanced machine learning algorithms to improve the accuracy of the comparison to ensure that the user's identity can still be accurately identified under diverse environments and lighting conditions. In addition, the system also supports multi-factor verification mode, that is, users can choose two or more authentication methods to use in combination to further enhance the security of the system.

[0059] Furthermore, the training step through the neural network includes:

[0060] Get the task image with the highest screen ratio of the electronic whiteboard;

[0061] Annotating the task image with usage scenarios, where the usage scenarios include a conference mode and a teaching mode;

[0062] The task image with completed scene annotation is used as the input of the convolutional neural network, and the usage scenario is output through back-propagation training.

[0063] In some embodiments, first, the system obtains the task image currently displayed on the electronic whiteboard screen. In order to improve the effectiveness and representativeness of the image, the system selects the task image with the highest screen ratio as a training sample. This means that the system will give priority to obtaining the main image content that currently occupies most of the screen area, because this part of the image usually contains key information of the current usage scenario, such as documents, presentation slides, or interactive content. By analyzing these high-ratio images, the system can more accurately capture the current operating scenario. Next, the system annotates the acquired task image with the usage scenario. The annotation of the usage scenario is performed manually or automatically, depending on the deployment environment and requirements of the system. Common usage scenarios include conference mode and teaching mode. In conference mode, the task image usually includes a slide show, a video conference interface, or a meeting record, while in teaching mode, the image may contain teaching courseware, blackboard content, or real-time interactive feedback. The annotation process is critical because it provides the neural network with the label data required for training, ensuring that the network can learn and recognize the features of different scenarios. After completing the scene annotation, the system uses the annotated task image as input data for the convolutional neural network (CNN). The convolutional neural network is a deep learning model that is particularly suitable for processing and analyzing image data. During the training process, the system inputs the labeled task image into the CNN and gradually extracts the features of the image through multiple layers of convolution, pooling and fully connected layers. The feature extraction process includes identifying the edges, textures, shapes and high-level semantic information in the image, which together constitute the feature representation of different usage scenarios. The CNN is trained using the back propagation algorithm. Back propagation is a common optimization algorithm that calculates the error between the network output and the actual annotation and adjusts the weight parameters in the network layer by layer to gradually reduce the prediction error. Specifically, the system backpropagates the error of each training to each layer of the network and optimizes the weights of each layer using the gradient descent method. After multiple iterations of training, the CNN can accurately classify the input task image into specific usage scenarios, such as conference mode or teaching mode. The convolutional neural network can recognize the usage scenarios of the electronic whiteboard in real time in actual use. The system automatically matches the appropriate lock mode based on the recognition results to ensure the safety and convenience of the electronic whiteboard in different scenarios. For example, when the conference mode is recognized, the system may automatically select the full lock mode to prevent misoperation; when the teaching mode is recognized, the system may select the partial lock mode to retain the necessary interactive functions. This intelligent scene recognition and lock pattern matching mechanism significantly improves the practicality and user experience of the electronic whiteboard.

[0064] Furthermore, the user usage habit data includes operations during screen lock and operations after unlock, and the electronic whiteboard lock mode includes button lock, touch lock and full lock, and the full lock is to trigger button lock and touch lock at the same time.

[0065] In some embodiments, first, the system generates operation data during the lock screen period by recording all operation behaviors of the user during the lock screen period. These operation data include the operations that the user attempts to perform when the electronic whiteboard is locked, such as trying to press a button, touching the screen for interaction, etc. The operation behavior during the lock screen period reflects the user's needs and behavior patterns in the lock state, and the collection of these data is of great significance for the subsequent optimization of the lock mode. Record the user's operation behavior after unlocking, and generate the unlocked operation data. These data include the operations actually performed by the user after unlocking the electronic whiteboard, such as pressing the volume button to adjust the volume, performing touch operations on the screen, or starting an application. By comparing and analyzing the operations during the lock screen period with the operations after unlocking, the system can identify which operations are blocked in the lock screen state, and these blocked operations are blocked operations. The blocked operation data has a direct guiding significance for the adjustment of the lock mode. Combined with the user's operation behavior data during the lock screen period and after unlocking, the system can establish a comprehensive set of user usage habit data. Through in-depth analysis of these data, the system can identify the user's operation preferences and habit patterns, and optimize the lock mode selection of the electronic whiteboard accordingly. This data-driven analysis method not only improves the intelligence of the system, but also ensures that the lock mode can better adapt to the actual needs of users. In terms of the selection of lock modes, the present invention provides a variety of lock methods, including button lock, touch lock and full lock. Button lock refers to restricting users from operating through physical buttons only, such as volume adjustment, device switch, etc., while the screen touch function is still retained. Touch lock means that the touch operation of the screen is disabled, but the physical button operation is still available. The full lock mode is the most stringent lock mode, which triggers button lock and touch lock at the same time, completely restricting any interactive operations of the user. This mode is generally used in scenarios that require high security or prevent misoperation.

[0066] Furthermore, matching the electronic whiteboard lock mode according to the current usage scenario and the user usage habit data includes:

[0067] If the power state of the electronic whiteboard is not in standby mode, matching of the electronic whiteboard lock mode is performed;

[0068] If the power state of the electronic whiteboard is standby, only the power button unlocking is triggered.

[0069] In some embodiments, first, the system monitors the power state of the electronic whiteboard in real time, and the power state can be divided into two situations: standby and non-standby. When the electronic whiteboard is in non-standby state, it means that the device is in normal operation, the user may be performing interactive operations or the device is in an active state. In this case, the system will match the appropriate lock mode based on the current usage scenario and the user's usage habit data. Specifically, the system will first identify the current usage scenario, such as conference mode, teaching mode, or demonstration mode. The recognition of the usage scenario is usually achieved by analyzing the display content of the electronic whiteboard by a neural network. Combined with the user's operation habit data, the system can judge the user's regular operation behavior in this scenario. For example, in conference mode, users may use the screen display and recording functions more frequently, so the system may choose a partial lock mode to retain the necessary interactive functions and avoid misoperation interfering with the meeting process. In teaching mode, the system may choose to moderately lock some functions, such as disabling some buttons to ensure classroom order while still retaining blackboard writing or interactive functions. On the other hand, when the electronic whiteboard is in standby mode, the system handles it differently. The standby state usually means that the device is in low-power mode, the user has not performed any operation or the device has been idle for a preset time, which is usually a relatively long time such as 3min-30min. At this time, the system will trigger the power button unlock function to ensure the security and low power consumption of the device. In this case, the user can only wake up the device and unlock it by pressing the power button, and other buttons or touch operations will remain disabled to prevent misoperation when the device is not fully started.

[0070] Furthermore, the matching of the electronic whiteboard lock mode includes:

[0071] Generate operation vectors for operations during screen lock and operations after screen unlock within a preset time period;

[0072] Calculate the cosine similarity between the operation vector during screen lock and the operation vector after screen unlock, and determine the blocked operation based on the cosine similarity;

[0073] According to the current usage scenario, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard, the electronic whiteboard lock mode is matched through the random forest model.

[0074] In some embodiments, the user's operation behavior is first recorded within a preset time window, and operation vectors for operations during the lock screen and after unlocking are generated respectively. The generation of the operation vector during the lock screen involves recording all interaction attempts of the user when the electronic whiteboard is locked, which may include attempts to click buttons, touch the screen, or make other inputs. Although these operations are not executed, they reflect the user's potential needs in the lock screen state. Similarly, the operation vector after unlocking records the operation behavior performed by the user immediately after unlocking the electronic whiteboard. These operations are usually operations that the user failed to complete during the lock screen period, reflecting the user's actual needs and priorities. Next, the system determines the similarity between the operation vector during the lock screen and the operation vector after unlocking by calculating the cosine similarity between the two. Cosine similarity is an effective method to measure the similarity between two vectors, and its value varies between -1 and 1. When the cosine similarity is close to 1, it indicates that the operation during the lock screen is highly similar to the operation after unlocking, which means that these operations are successfully completed after unlocking. This similarity is defined by the system as a "blocked operation", that is, the user's needs in the lock screen state are not met until they are unlocked. By analyzing this similarity, the system can identify which operations are key operations that are blocked by the user in the lock screen state. On the basis of determining the blocked operation, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard in the current usage scenario are further counted. These data are important bases for the system to match the lock mode. For example, if the frequency of blocked operations and the unlocking frequency are both high in a certain usage scenario, it means that the current lock mode may not be suitable for the scenario, and the system should consider adjusting the lock mode to improve the user experience. On the contrary, if the blocked operation frequency is low and the unlocking frequency is moderate, it means that the current lock mode is more suitable, and the system may maintain the existing settings. In order to achieve intelligent matching of the lock mode, the present invention uses a random forest model for decision making. Random forest is an integrated learning method that improves the prediction accuracy by constructing multiple decision trees and voting on their prediction results. In this embodiment, the system uses the usage scenario, the frequency of blocked operations and the unlocking frequency as input features, and inputs them into the random forest model for training and prediction. The model outputs an optimal lock mode suggestion through a comprehensive analysis of these features, and the system automatically adjusts the lock mode of the electronic whiteboard according to the suggestion. For example, in a scenario where blocked operations occur frequently, the system may recommend a partial lock mode to reduce the number of times the user unlocks; in a scenario where high security is required, the system may recommend a full lock mode.

[0075] Further, the matching of the electronic whiteboard lock mode by the random forest algorithm includes:

[0076] The usage scenarios, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard are cleaned and standardized into feature vectors, and then integrated to generate a feature vector set;

[0077] Use the feature vector set as input, train it through the random forest algorithm, determine the optimal model parameters through cross-validation, and build a random forest model;

[0078] The current usage scenario, real-time monitored blocked operation frequency and unlocking frequency data are input into the trained random forest model, and the electronic whiteboard lock mode is automatically matched according to the model output results.

[0079] In some embodiments, the collected raw data is first cleaned and standardized. Specifically, the system extracts data related to usage scenarios, frequency of blocked operations, and unlocking frequency from historical operation records. These data may be incomplete or inaccurate due to the presence of noise, missing values, or outliers, so they need to be cleaned. The data cleaning process includes removing abnormal data points, filling missing values, and eliminating irrelevant features to ensure the quality and consistency of the data. The cleaned data will be further standardized to eliminate the impact of dimensions between different features so that all feature values ​​are in the same range (usually between 0 and 1). The standardized feature vector ensures that the contributions of different features in the model training process are treated equally and improves the prediction accuracy of the model. After completing data cleaning and standardization, the system integrates these processed data to generate a feature vector set. The feature vector set includes key indicators such as the current usage scenario, frequency of blocked operations, and unlocking frequency. These feature vectors are important inputs for the system to train the random forest model, reflecting the user's operating mode and needs in different scenarios. By integrating these features, the system can capture the complexity and diversity of user behavior and provide a solid data foundation for subsequent lock pattern matching. Next, the system uses the generated feature vector set as input to train the random forest model. The random forest algorithm is an ensemble learning method based on multiple decision trees that can handle high-dimensional data and has strong resistance to overfitting. During the training process, the system constructs multiple decision trees by randomly sampling the feature vector set multiple times, and each decision tree makes independent predictions on the input features. The system determines the optimal model parameters (such as the number of decision trees, maximum depth, etc.) through the cross-validation method to ensure the generalization ability and prediction accuracy of the model. Cross-validation divides the data set into a training set and a validation set, repeatedly trains and verifies the performance of the model, and thus finds the parameter configuration that best adapts to the data. After the model training is completed and optimized, the system inputs the current usage scenario, the frequency of blocked operations and the unlocking frequency data monitored in real time into the trained random forest model. The model outputs a prediction result of the lock mode by analyzing these real-time data. Specifically, the model will comprehensively consider the current operating environment and user behavior and automatically select the most appropriate lock mode. For example, in the case of high frequency of blocked operations and unlocking operations, the model may recommend enabling partial lock mode to reduce the user's unlocking operation frequency; in more sensitive scenarios, the model may recommend enabling full lock mode to ensure the security of the system.

[0080] Furthermore, the formula of the random forest model is as follows:

[0081]

[0082] in, is the final matching electronic whiteboard lock mode; Scene is the usage scenario; Freq Blocked Freq is the frequency of blocked operations; Unlock is the unlocking frequency; Ti(Scene, Freq Blocked , Freq Unlock ) is the prediction result of the i-th decision tree on the usage scenario, the frequency of blocked operations and the unlocking frequency; mode is the mode of the prediction results of all decision trees; n is the number of decision trees.

[0083] Specifically, each decision tree is first used to independently analyze the input feature vector set to obtain its own prediction results. Then, the system uses the mode function to count the prediction results of all decision trees and selects the prediction value with the highest frequency as the final electronic whiteboard lock mode. This method improves the accuracy and robustness of the prediction by integrating the advantages of multiple decision trees, ensuring that the system can adapt to complex and changing usage scenarios and user behaviors. Finally, the system will automatically adjust the lock mode of the electronic whiteboard according to the prediction results to optimize the user experience and operational safety.

[0084] Furthermore, capturing the user unlocking request and unlocking the electronic whiteboard includes:

[0085] Authentication is performed according to the user's unlocking request. If the authentication is passed, the electronic whiteboard is unlocked; if the authentication is not passed, the electronic whiteboard remains locked.

[0086] In some embodiments, first, the system monitors and captures the user's unlock request in real time. The unlock request is usually issued by the user through physical button operation, touching the screen or entering a specific unlock command. The system accurately captures these unlock signals through built-in sensors and interface modules, and triggers the start of the unlock process. Next, the system authenticates the user's unlock request. Identity authentication is a core step to ensure the legitimacy and security of the unlock operation. The system provides a variety of identity authentication methods, including but not limited to password input, face recognition, fingerprint recognition, and voice recognition. After the user issues an unlock request, the system will require the user to provide corresponding identity information according to the preset verification method. For example, if the user selects fingerprint recognition, the system will activate the fingerprint sensor, capture the user's fingerprint image, and compare it with the fingerprint template stored in the database. If the identity authentication is passed, the system will unlock the electronic whiteboard. Specifically, the system will restore the disabled functions one by one according to the current lock mode (such as button lock, touch lock or full lock), allowing the user to operate the electronic whiteboard normally. For example, in the full lock mode, the system will restore the functions of the button and touch screen at the same time, allowing the user to fully access and operate the electronic whiteboard. After successful unlocking, the system will also record the detailed information of the unlocking operation, including the unlocking time, unlocking method and verification results, for future auditing and security management. However, if the identity authentication fails, the system will keep the electronic whiteboard locked and reject the user's unlocking request. In this case, the system may prompt the user to re-authenticate, or trigger additional security measures after multiple failures, such as sending an alarm message or temporarily disabling the unlocking function to prevent brute force cracking. In this locked state, users will not be able to perform any unauthorized operations, effectively protecting the security of the system and the integrity of the data.

[0087] The present invention also includes an electronic whiteboard multi-mode operation locking system, including a driver loading module, an identity authentication module, a remote locking matching adjustment module and a log recording module;

[0088] The driver loading module is used to initialize the electronic whiteboard and load the locking driver;

[0089] The identity verification module is used to verify the user's identity, capture the user's unlocking request and unlock the electronic whiteboard;

[0090] The remote lock matching adjustment module is used to identify the current usage scenario through a cloud server equipped with a neural network, and match the electronic whiteboard lock mode according to the current usage scenario and the user's usage habit data;

[0091] The log recording module is used to store user usage habit data, blocked operation logs and unlocking logs.

[0092] Among them, the remote lock matching adjustment module is used to carry a neural network through the cloud server to identify the current usage scenario, match the electronic whiteboard lock mode according to the current usage scenario and the user's usage habit data, and effectively reduce the dependence on local computing power. The introduction of the remote lock matching adjustment module enables the system to continue to learn and optimize, and maintain efficient and intelligent operation lock management. The logging module is responsible for recording and storing user usage habit data, blocked operation logs, and unlocking logs. By recording these log data in detail, the system can provide basic data support for subsequent analysis and optimization. For example, the blocked operation log can help the system identify the user's needs in the locked screen state, and the unlocking log helps to audit the user's access behavior. The logging module also supports uploading this data to the cloud for remote monitoring and management.

[0093] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for locking multi-mode operation of an electronic whiteboard, characterized in that: The following steps are involved: Initialize the electronic whiteboard and load the lock driver; Verify user identity and identify current usage scenarios through neural networks; Match the electronic whiteboard lock mode according to the current usage scenario and user usage habit data; Capture the user's unlock request and unlock the electronic whiteboard.

2. The method for locking multi-mode operation of an electronic whiteboard according to claim 1, characterized in that: The initialization of the electronic whiteboard includes initializing a processor, a touch sensor, a button interface and a display screen of the electronic whiteboard.

3. The method for locking multi-mode operation of an electronic whiteboard according to claim 1, characterized in that: The verification of user identity includes: Providing identity verification methods, including password verification, face recognition verification, fingerprint recognition verification and voice verification; Perform comparison and verification through identity verification; Load user usage habit data based on the verification results.

4. The method for locking multi-mode operation of an electronic whiteboard according to claim 1, characterized in that: The training step of the neural network includes: Get the task image with the highest screen ratio of the electronic whiteboard; Annotating the task image with usage scenarios, where the usage scenarios include a conference mode and a teaching mode; The task image with completed scene annotation is used as the input of the convolutional neural network, and the usage scenario is output through back-propagation training.

5. The method for locking multi-mode operation of an electronic whiteboard according to claim 1, characterized in that: The user usage habit data includes operations during screen lock and operations after unlock. The electronic whiteboard lock modes include button lock, touch lock and full lock. The full lock is to trigger button lock and touch lock at the same time.

6. The method for locking multi-mode operation of an electronic whiteboard according to claim 5, characterized in that: The electronic whiteboard locking mode is matched according to the current usage scenario and the user's usage habit data, including: If the power state of the electronic whiteboard is not in standby mode, matching of the electronic whiteboard lock mode is performed; If the power state of the electronic whiteboard is standby, only the power button unlocking is triggered.

7. The method for locking multi-mode operation of an electronic whiteboard according to claim 6, characterized in that: The matching of the electronic whiteboard locking mode includes: Generate operation vectors for operations during screen lock and operations after screen unlock within a preset time period; Calculate the cosine similarity between the operation vector during screen lock and the operation vector after screen unlock, and determine the blocked operation based on the cosine similarity; According to the current usage scenario, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard, the electronic whiteboard lock mode is matched through the random forest model.

8. The method for locking multi-mode operation of an electronic whiteboard according to claim 7, characterized in that: The electronic whiteboard locking mode matching by random forest algorithm includes: The usage scenarios, the frequency of blocked operations and the unlocking frequency of the electronic whiteboard are cleaned and standardized into feature vectors, and then integrated to generate a feature vector set; Use the feature vector set as input, train it through the random forest algorithm, determine the optimal model parameters through cross-validation, and build a random forest model; The current usage scenario, real-time monitored blocked operation frequency and unlocking frequency data are input into the trained random forest model, and the electronic whiteboard lock mode is automatically matched according to the model output results.

9. The method for locking multi-mode operation of an electronic whiteboard according to claim 8, characterized in that: The formula of the random forest model is as follows: in, is the final matching electronic whiteboard lock mode; Scene is the usage scenario; Freq Blocked Freq is the frequency of blocked operations; Unlock is the unlocking frequency; Ti(Scene, Freq Blocked , Freq Unlock ) is the prediction result of the i-th decision tree on the usage scenario, the frequency of blocked operations and the unlocking frequency; mode is the mode of the prediction results of all decision trees; n is the number of decision trees.

10. The method for locking multi-mode operation of an electronic whiteboard according to claim 1, characterized in that: The capturing of the user unlocking request and unlocking the electronic whiteboard comprises: Authentication is performed according to the user's unlocking request. If the authentication is passed, the electronic whiteboard is unlocked; if the authentication is not passed, the electronic whiteboard remains locked.

11. An electronic whiteboard multi-mode operation locking system, applied to an electronic whiteboard multi-mode operation locking method according to any one of claims 1 to 10, characterized in that: It includes a driver loading module, an authentication module, a remote lock matching adjustment module and a logging module; The driver loading module is used to initialize the electronic whiteboard and load the locking driver; The identity verification module is used to verify the user's identity, capture the user's unlocking request and unlock the electronic whiteboard; The remote lock matching adjustment module is used to identify the current usage scenario through a cloud server equipped with a neural network, and match the electronic whiteboard lock mode according to the current usage scenario and the user's usage habit data; The log recording module is used to store user usage habit data, blocked operation logs and unlocking logs.

Images