Desktop computer information security management system and method

Abstract

The invention discloses a desktop computer information security management system and method, and relates to the technical field of computer information security. The method comprises the following steps: after the STM32F4 main controller is powered on, hardware initialization and software parameter configuration of the face recognition detection module, the fingerprint recognition detection module, the video isolation management module and the network isolation management module are completed, and communication links and control logic among the modules are established; image acquisition and preprocessing are performed on an area in front of a computer regularly through a face recognition detection module, face area detection is performed on a gray level image based on a cascade classifier algorithm, and the presence state of a user is judged in real time. According to the invention, periodic face state monitoring is adopted as a main protection triggering mechanism, and a fingerprint identification rapid recovery function is combined, so that intelligent safety state switching without active intervention of a user is realized. According to the design, the real-time performance and reliability of protection are remarkably improved, and the robustness of the system in various application scenes is ensured through a dual-mode recovery mechanism.

Description

Technical Field

[0001] This invention relates to the field of computer information security technology, and in particular to a desktop computer information security management system and method. Background Technology

[0002] In the information age, desktop computers, as core terminals for processing sensitive data in personal and office settings, directly impact user privacy and the protection of business secrets through their information security capabilities. Current mainstream security mechanisms generally rely on operating system-level software strategies, such as screen lock passwords, firewall rules, or remote access control. While these measures can mitigate conventional network threats to some extent, they are insufficient to address physical-level eavesdropping risks and advanced persistent attacks (APAs). Especially when users temporarily leave without actively locking the screen or shutting down the computer, the running computer continues to output display signals and maintain network connectivity, making it highly vulnerable to nearby individuals spying on the screen content or remote malicious programs exploiting vulnerabilities to steal data, exposing security blind spots.

[0003] Among these technologies, automatic protection based on identity authentication is gaining increasing attention. However, existing solutions are mostly limited to software-triggered logic and lack the ability to control the underlying hardware of video output and network links. For example, some systems activate the screensaver after detecting when a user leaves their seat via a camera, but the monitor still receives signals from the graphics card, posing a risk of capturing the image through external devices. Other solutions attempt to cut off network service processes but fail to achieve a true physical disconnection, allowing attackers to still penetrate the underlying protocol stack. Furthermore, existing automatic recovery mechanisms often rely on periodic facial recognition, resulting in long response delays, low verification efficiency, and negatively impacting user experience. The recognition rate also drops significantly in low light or obstructed environments, making it difficult to balance security and convenience.

[0004] Therefore, there is an urgent need for a desktop computer information security management method that integrates biometric recognition and hardware-level on / off control. This method can automatically and reliably cut off the video signal transmission path and network physical link when the user leaves, blocking the information leakage channel at the source. When the user returns, the usage status can be restored immediately through rapid identity verification, thereby building a physical isolation protection system that combines high security, strong reliability and a good interactive experience. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a desktop computer information security management system and method. It uses periodic face status monitoring as the main protection trigger mechanism and combines it with fingerprint recognition for rapid recovery, thereby achieving intelligent security status switching without user intervention.

[0006] In a first aspect, the present invention provides a desktop computer information security management method, applied to a desktop computer information security management system, the system comprising a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a status indication module; the method comprises: Step 1: System initialization and module preparation. After the STM32F4 main controller is powered on, it completes the hardware initialization and software parameter configuration of the face recognition detection module, fingerprint recognition detection module, video isolation management module and network isolation management module, and establishes the communication link and control logic between the modules. Step 2: Periodic face status monitoring. The face recognition detection module periodically collects and preprocesses images of the area in front of the computer, and performs face region detection on the grayscale images based on the cascaded classifier algorithm to determine the user's presence status in real time. Step 3: Maintaining normal operation. When the presence of a user is detected, the STM32F4 main controller sends a keep-on command to the video isolation management module and the network isolation management module to ensure that the video signal is transmitted normally to the display and the network connection remains uninterrupted. Step 4: User leaving seat determination and protection activation. When no user facial features are detected within a continuous monitoring period, the STM32F4 main controller determines that the user has left the seat and then sends a disconnect command to the video isolation management module and the network isolation management module, thereby turning off the display screen and physically disconnecting the network. Step 5: The quick recovery mechanism is activated. After the user returns, they can authenticate their identity through the fingerprint recognition detection module. Feature matching is completed within 2 seconds, and the main controller is triggered to reconnect the video and network, restoring the computer to normal operation.

[0007] Preferably, step 1 includes: The STM32F4 main controller includes flash memory and random access memory; The STM32F4 main controller sends a reset signal to the face recognition detection module and waits for a ready response to confirm that the power supply voltage of the face recognition detection module is stable and the data interface is in an idle state. Then, the fingerprint recognition detection module is configured with registers, the acquisition mode is set to continuous trigger, the template storage area address is mapped to random access memory, and the preset matching threshold parameters are loaded. The STM32F4 main controller initializes the channel selection of the two video switching chips in the video isolation management module, sets the default output channel to the on state, and verifies the switching control pin level logic. For the network isolation management module, the STM32F4 main controller detects the driving capability of the electromagnetic relay drive circuit to ensure that the contacts can be reliably engaged or disengaged under subsequent control commands. In addition, the STM32F4 main controller also establishes an interrupt service routine entry point, which is bound to the face recognition timer interrupt, fingerprint recognition event interrupt, and status indicator update task, respectively.

[0008] Preferably, step 2 includes the following steps: The face recognition detection module includes an image sensor and an OV7670 camera, and transmits image data with the STM32F4 main controller through a 16-bit parallel data interface; During the data acquisition phase, the STM32F4 main controller sends a start acquisition command to the face recognition detection module through the general input / output interface. The OV7670 camera then activates the light source and begins scanning line by line to generate the raw image data stream. The STM32F4 main controller triggers image preprocessing immediately after receiving the raw image data stream. The image preprocessing includes automatic white balance correction, gamma correction, and grayscale conversion, which are used to restore color, enhance details, and simplify data in the raw image data stream to generate a grayscale image. Then, a cascaded classifier algorithm based on Haar features is called to detect face regions in the grayscale image. The cascaded classifier algorithm is composed of multiple strong classifiers connected in series. Each level is a weighted combination of weak classifiers trained by AdaBoost. It only enters the next level of detection when the current level determines that the face is a candidate region. During the detection process, the cascaded classifier algorithm traverses the sliding window positions in the grayscale image. The window size starts at 24×24 pixels and is gradually enlarged by a scaling factor of 1.1 until it covers the entire image. If a face region that conforms to the Haar feature distribution is detected in any window, the center coordinates and scale parameters of the face region are recorded and marked as a valid face feature. The entire facial feature analysis process is carried out with the support of a single-user facial feature template library, which stores only the frontal face feature vector of an authorized user. If a valid facial feature is detected within a single cycle, the STM32F4 main controller marks the cycle as the user is present; otherwise, it marks the cycle as the user has left the seat.

[0009] Preferably, step 3 includes the following steps: The video isolation management module includes a CV10211E1 chip, a CV10211E2 chip, a DP port, an HDMI interface, and a VGA interface. The input terminals of the CV10211E1 chip and the CV10211E2 chip are respectively connected to the corresponding video output interfaces of the desktop computer host, and the output terminals are connected to an external monitor. Under normal use, the STM32F4 main controller transmits low-level signals to the input terminals of the CV10211E1 chip and the CV10211E2 chip through general-purpose input / output pins, keeping their internal analog switches on and allowing the video signal to pass through without loss. Meanwhile, the network isolation management module uses an electromagnetic relay to control the physical layer network connection. In the disconnected state, the network interface resistance is greater than 100 megohms. The normally open contact of the electromagnetic relay is connected in series in the Ethernet physical layer line of the desktop computer host. The STM32F4 main controller controls the relay coil current through a driving transistor. During the user's continuous presence, the STM32F4 main controller maintains the coil energized, the contacts are closed, and the network signal is transmitted normally. In addition, when the user is present, the STM32F4 main controller synchronously updates the status indicator module, driving the green LED to remain constantly lit, providing the user with intuitive feedback that the system is in a safe operating state.

[0010] Preferably, step 4 includes the following steps: The user leaving the seat determination condition is that no valid facial feature is detected for two consecutive monitoring cycles. The STM32F4 main controller is configured with a counter to record the number of consecutive user leaving the seat cycles. Whenever a monitoring cycle ends and no valid facial feature is detected, the counter is incremented by 1. If a valid facial feature is detected, the counter is reset to zero. When the counter value reaches 2, the STM32F4 main controller immediately triggers the user leaving the seat determination. Upon detecting that the user has left their seat, the STM32F4 main controller first outputs a high-level signal to the input of the video isolation management module. The internal analog switches of the CV10211E1 and CV10211E2 chips quickly disconnect, cutting off the video signal path, and the display immediately enters a black screen state with no signal. At the same time, the STM32F4 main controller shuts off the drive current of the electromagnetic relay, and the relay contacts spring open under the action of spring force, physically disconnecting the Ethernet line, so that the external network cannot communicate with the host through this interface. In the disconnected state, the network interface resistance is greater than 100 megohms, cutting off the underlying network attack path. After protection is activated, the red LED of the STM32F4 main controller drive status indicator module will remain lit, alerting the user that the system has entered protection mode.

[0011] Preferably, step 5 includes the following steps: The fingerprint recognition detection module uses an optical fingerprint sensor. When a user's finger touches the sensing area of ​​the fingerprint recognition detection module, the optical sensor immediately captures the light and dark image formed by the fingerprint ridges and valleys, generating a fingerprint grayscale image. A minutiae matching algorithm is used to identify the endpoints and bifurcation points in the fingerprint grayscale image and extract their coordinates, orientation angles, and type information to form a feature point set. The feature point set is compared with the internally stored registered fingerprint templates to calculate the similarity score between the two. If the score exceeds the matching threshold, the user is deemed a legitimate user. After successful verification, the STM32F4 main controller immediately sends a connection command to the video isolation management module and the network isolation management module to restore the video and network connection and drive the status indicator module to switch to a solid green light.

[0012] Preferably, the desktop computer is equipped with a dual-mode recovery mechanism, which retains the periodic automatic recovery function for face recognition in addition to the rapid recovery of fingerprint recognition; the STM32F4 main controller will still periodically wake up the face recognition detection module to perform a quick scan in the protection state. If the facial features of an authorized user are detected, the recovery process will also be triggered to ensure that normal use can still be restored through face recognition when the fingerprint module fails.

[0013] Secondly, the present invention also provides a desktop computer information security management system, which applies a desktop computer information security management method as described above, wherein the system includes a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a status indication module; The face recognition detection module is used to acquire and preprocess images of the area in front of the computer, and to detect face regions in grayscale images based on a cascaded classifier algorithm to determine the user's presence in real time. The fingerprint recognition detection module is used to capture the user's fingerprint image through an optical fingerprint sensor and to perform feature extraction and matching using a minutiae matching algorithm to verify the user's identity. The video isolation management module is used to connect or disconnect the video signal path under the control of the STM32F4 main controller, so as to realize the normal transmission of video signal of the display or the black screen state. The network isolation management module is used to connect or disconnect the physical network connection under the control of the STM32F4 main controller to ensure that the network connection remains unobstructed or physically disconnected. The STM32F4 main controller is used for system initialization and configuration of parameters of each module, establishing communication links and control logic, periodically triggering face status monitoring and controlling video and network isolation modules according to the results, and processing fingerprint verification to quickly restore the system to normal state. The status indicator module is used to adjust the LED indicator light according to the number of users present.

[0014] Compared with the prior art, the present invention has the following advantages and beneficial effects: Compared to traditional security mechanisms that rely on software strategies, this invention automatically triggers the physical disconnection of video signals and the hardware disconnection of network connections when a user temporarily leaves the site, fundamentally eliminating the risk of information leakage due to exposed screen content or active network interfaces. This invention employs periodic facial recognition as the primary protection trigger mechanism, combined with fingerprint recognition for rapid recovery, achieving intelligent security state switching without user intervention. This design not only significantly improves the real-time performance and reliability of protection but also ensures the system's robustness across various application scenarios through a dual-mode recovery mechanism. Furthermore, hardware-level on / off control makes the protection measures difficult to bypass by malicious software attacks, greatly enhancing the system's anti-penetration capabilities.

[0015] Overall, this invention ensures a high level of security while also taking into account user convenience and system energy efficiency management. It effectively resolves the contradiction between security blind spots and user experience in existing technologies, and provides an efficient and reliable solution for computer terminal management in sensitive environments. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the first process of a desktop computer information security management method.

[0017] Figure 2 This is a schematic diagram of the second process of a desktop computer information security management method.

[0018] Figure 3 This is a schematic diagram of the structure of a desktop computer information security management system. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0020] Example 1 Please see Figures 1-3This embodiment provides a desktop computer information security management method, applied to a desktop computer information security management system. The system includes a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a power management module. The method includes: Step 1: System initialization and module preparation. After the STM32F4 main controller is powered on, it completes the hardware initialization and software parameter configuration of the face recognition detection module, fingerprint recognition detection module, video isolation management module and network isolation management module, and establishes the communication link and control logic between the modules. Step 2: Periodic face status monitoring. The face recognition detection module periodically collects and preprocesses images of the area in front of the computer, and performs face region detection on the grayscale images based on the cascaded classifier algorithm, thereby determining the user's presence status in real time. Step 3: Maintaining normal operation. When the presence of a user is detected, the STM32F4 main controller sends a keep-on command to the video isolation management module and the network isolation management module to ensure that the video signal is transmitted normally to the display and the network connection remains uninterrupted. Step 4: User leaving seat determination and protection activation. When no user facial features are detected within a continuous monitoring period, the STM32F4 main controller determines that the user has left the seat and then sends a disconnect command to the video isolation management module and the network isolation management module, thereby turning off the display screen and physically disconnecting the network. Step 5: The quick recovery mechanism is activated. After the user returns, they can authenticate their identity through the fingerprint recognition detection module. Feature matching is completed within 2 seconds, and the main controller is triggered to reconnect the video and network, restoring the computer to normal operation.

[0021] In the above-mentioned desktop computer information security management method, step 1 includes: The STM32F4 main controller is a 32-bit microcontroller based on the ARM Cortex-M4 architecture, with a working frequency of 168MHz. It is equipped with 512 kilobytes of flash memory and 192 kilobytes of random access memory, and establishes physical connections with various functional modules through general-purpose input / output interfaces.

[0022] The STM32F4 main controller first performs an internal clock calibration and power management module self-test. Then, it sequentially sends a reset signal to the face recognition detection module and waits for its ready response, confirming that the image sensor power supply voltage of the face recognition detection module is stable and the data interface is idle. Next, it configures the registers of the fingerprint recognition detection module, sets the acquisition mode to continuous trigger, maps the template storage area address to random access memory, and loads the preset matching threshold parameters. Simultaneously, the STM32F4 main controller initializes the channel selection of the two video switching chips in the video isolation management module, sets the default output channel to the ON state, and verifies the switching control pin level logic. For the network isolation management module, the STM32F4 main controller tests the driving capability of the electromagnetic relay drive circuit to ensure reliable contact engagement or release under subsequent control commands. After all modules are initialized, the STM32F4 main controller establishes interrupt service routine entry points, binding face recognition timer interrupts, fingerprint recognition event interrupts, and status indicator update tasks, forming a complete multi-task scheduling framework.

[0023] In addition, the STM32F4 main controller loads a single-user facial feature template library and a set of registered fingerprint templates from its internal random access memory, completing the loading of the local biometric database and providing data support for the subsequent identity verification process.

[0024] In the above-mentioned desktop computer information security management method, step 2 includes: The face recognition detection module includes an image sensor and an OV7670 camera, supporting VGA resolution image acquisition at 30 frames per second. Image data is transmitted to the STM32F4 main controller via a 16-bit parallel data interface. At the start of each monitoring cycle, the STM32F4 main controller sends a start acquisition command to the face recognition detection module through a general-purpose input / output interface. The OV7670 camera then activates its light source and begins line-by-line scanning, generating a raw image data stream with a resolution of 640×480 pixels. This raw image data stream is transmitted to the STM32F4 main controller's random access memory via a 16-bit parallel bus. Upon receiving the raw image data stream, the STM32F4 main controller immediately triggers image preprocessing. Image preprocessing includes automatic white balance correction, gamma correction, and grayscale conversion, used to generate a more stable and efficient grayscale image from the raw image data stream through color restoration, detail enhancement, and data simplification.

[0025] Subsequently, the STM32F4 main controller invokes a cascaded classifier algorithm based on Haar features to detect face regions in the grayscale image. This cascaded classifier algorithm consists of multiple strong classifiers connected in series, each level being a weighted combination of weak classifiers trained with AdaBoost. It only proceeds to the next level of detection if the current level determines a face candidate region, thus significantly reducing computational overhead. During detection, the cascaded classifier algorithm traverses all possible sliding window positions in the grayscale image, starting with a window size of 24×24 pixels and scaling up by a factor of 1.1 until the entire image is covered. If a face region matching the Haar feature distribution is detected within any window, the center coordinates and scale parameters of that region are recorded and marked as a valid face feature. The entire facial feature analysis process is supported by a single-user facial feature template library, which stores only the frontal face feature vector of one authorized user. This vector is generated from multiple sample images collected during initial registration after normalization, alignment, and feature extraction. The processing latency is less than 100ms, ensuring sufficient time for multiple samplings and result determination within a 50s cycle. If a valid facial feature is detected in a single cycle, the STM32F4 main controller marks that cycle as the user's presence; otherwise, it marks it as the user leaving their seat.

[0026] In the above-mentioned desktop computer information security management method, step 3 includes: The video isolation management module includes a CV10211E1 chip, a CV10211E2 chip, a DP port, an HDMI interface, and a VGA interface. The input terminals of the CV10211E1 and CV10211E2 chips are connected to the corresponding video output interfaces of the desktop computer host, while their output terminals are connected to an external monitor. Under normal operating conditions, the STM32F4 main controller transmits low-level signals to the input terminals of the CV10211E1 and CV10211E2 chips via general-purpose input / output pins, keeping their internal analog switches conducting and allowing lossless video signal transmission. Simultaneously, the network isolation management module uses electromagnetic relays to control physical layer network connections, supporting 100Mbps and Gigabit Ethernet standards. In the disconnected state, the network interface resistance is greater than 100 megohms. The normally open contacts of the electromagnetic relays are connected in series in the Ethernet physical layer line of the desktop computer host, and the STM32F4 main controller controls the relay coil current through a driving transistor. During continuous user presence, the STM32F4 main controller maintains the coil energized, the contacts are closed, and network signals are transmitted normally. At the end of each monitoring cycle, if the main controller determines that the user is present, it refreshes the hold commands of the video isolation management module and the network isolation management module to prevent accidental disconnection due to signal jitter. In addition, when the user is present, the STM32F4 main controller synchronously updates the status indicator module, driving a green LED to remain constantly lit, providing the user with direct feedback that the system is in a safe operating state.

[0027] In the above-mentioned desktop computer information security management method, step 4 includes: The user leaving the seat determination condition is that no valid facial feature is detected for two consecutive monitoring cycles. After the determination, the STM32F4 main controller outputs a high-level signal to the video isolation management module and the network isolation management module. The STM32F4 main controller is equipped with a counter to record the number of consecutive user leaving the seat cycles; each time a monitoring cycle ends and no valid facial feature is detected, the counter is incremented by 1; if a valid facial feature is detected, the counter is reset to zero. When the counter value reaches 2, the STM32F4 main controller immediately triggers the user leaving the seat determination. After the user leaves the seat, the STM32F4 main controller first outputs a high-level signal to the input terminal of the video isolation management module. The analog switches inside the CV10211E1 chip and the CV10211E2 chip are quickly disconnected, cutting off the video signal path, and the display immediately enters a no-signal black screen state. Simultaneously, the STM32F4 main controller shuts off the drive current of the electromagnetic relay. The relay contacts spring open under spring force, physically disconnecting the Ethernet line. This prevents external networks from communicating with the host through this interface. In the disconnected state, the network interface resistance is greater than 100 megohms, effectively blocking the underlying network attack path. The entire disconnection operation is completed within 5ms, ensuring the immediacy of the protection measures. After protection is activated, the red LED of the STM32F4 main controller's drive status indicator module remains constantly lit, alerting the user that the system has entered protection mode. Furthermore, the STM32F4 main controller suspends the periodic data acquisition task of the face recognition detection module, entering a low-power monitoring mode, retaining only the fingerprint recognition detection module's detection to save system power.

[0028] In the above-mentioned desktop computer information security management method, step 5 includes: The fingerprint recognition detection module uses an optical fingerprint sensor, supporting the storage of up to 100 fingerprint templates. When a user's finger touches the sensing area of ​​the fingerprint recognition detection module, the optical sensor immediately captures the light and dark image formed by the fingerprint ridges and valleys, generating a 512×512 pixel grayscale fingerprint image. The STM32F4 main controller then initiates the fingerprint feature matching process, which includes: fingerprint image acquisition, feature point extraction, and template comparison. In the feature point extraction stage, the system uses a minutiae matching algorithm to identify endpoints and bifurcation points in the fingerprint grayscale image and extract their coordinates, orientation angles, and type information to form a feature point set. Subsequently, this feature point set is compared with the internally stored registered fingerprint templates to calculate the similarity score. The matching threshold is set to 85% similarity; if the score exceeds this threshold, the user is considered legitimate. The entire matching process is completed within 2 seconds, meeting the time requirement for rapid recovery. After successful verification, the STM32F4 main controller immediately sends a connection command to the video isolation management module and the network isolation management module to restore video and network connections and drive the status indicator module to switch to a solid green light.

[0029] In addition, the system has a dual-mode recovery mechanism, which retains the periodic automatic recovery function for facial recognition in addition to the rapid recovery of fingerprint recognition. The STM32F4 main controller will still periodically wake up the facial recognition detection module to perform a quick scan in the protection state. If the facial features of an authorized user are detected, the recovery process will also be triggered to ensure that normal use can still be restored through facial recognition when the fingerprint module fails.

[0030] The main control logic flowchart for system initialization and multi-module collaborative operation comprehensively demonstrates the complete state machine of the main controller from power-on initialization to the collaborative operation of various modules. After power-on, the system first enters the initialization phase, completing the configuration of each hardware module and loading software parameters; then it enters the main loop, periodically performing face status monitoring, and deciding whether to maintain normal use or activate protection based on the monitoring results; in protection mode, the system switches to low-power listening mode, waiting for fingerprint or face recovery events; once verification is successful, the system immediately returns to normal use and restarts periodic monitoring. The entire process ensures the optimal balance between security, real-time performance, and user experience.

[0031] Example 2 Please see Figures 1-3 This embodiment provides a desktop computer information security management system, which applies a desktop computer information security management method as described above. The system includes a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a status indication module. The face recognition detection module is used to acquire and preprocess images of the area in front of the computer, and to detect face regions in grayscale images based on a cascaded classifier algorithm to determine the user's presence in real time. The fingerprint recognition detection module is used to capture the user's fingerprint image through an optical fingerprint sensor and to perform feature extraction and matching using a minutiae matching algorithm to verify the user's identity. The video isolation management module is used to connect or disconnect the video signal path under the control of the STM32F4 main controller, so as to realize the normal transmission of video signal of the display or the black screen state. The network isolation management module is used to connect or disconnect the physical network connection under the control of the STM32F4 main controller to ensure that the network connection remains unobstructed or physically disconnected. The STM32F4 main controller is used for system initialization and configuration of parameters of each module, establishing communication links and control logic, periodically triggering face status monitoring and controlling video and network isolation modules according to the results, and processing fingerprint verification to quickly restore the system to normal state. The status indicator module is used to adjust the LED indicator light according to the number of users present.

[0032] All content not described in detail in this specification is prior art known to those skilled in the art, and the model parameters of each electrical appliance are not specifically limited; conventional equipment can be used. Electrical control components not mentioned in this technical solution are not shown in the figures because they are prior art, and will not be described further here.

[0033] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for managing information security of a desktop computer, characterized in that: An application is made in a desktop computer information security management system, the system comprising a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a status indication module; the method comprises: Step 1: System initialization and module preparation. After the STM32F4 main controller is powered on, it completes the hardware initialization and software parameter configuration of the face recognition detection module, fingerprint recognition detection module, video isolation management module and network isolation management module, and establishes the communication link and control logic between the modules. Step 2: Periodic face status monitoring. The face recognition detection module periodically collects and preprocesses images of the area in front of the computer, and performs face region detection on the grayscale images based on the cascaded classifier algorithm to determine the user's presence status in real time. Step 3: Maintaining normal operation. When the presence of a user is detected, the STM32F4 main controller sends a keep-on command to the video isolation management module and the network isolation management module to ensure that the video signal is transmitted normally to the display and the network connection remains uninterrupted. Step 4: User leaving seat determination and protection activation. When no user facial features are detected within a continuous monitoring period, the STM32F4 main controller determines that the user has left the seat and then sends a disconnect command to the video isolation management module and the network isolation management module, thereby turning off the display screen and physically disconnecting the network. Step 5: The quick recovery mechanism is activated. After the user returns, they can authenticate their identity through the fingerprint recognition detection module. Feature matching is completed within 2 seconds, and the main controller is triggered to reconnect the video and network, restoring the computer to normal operation.

2. The desktop computer information security management method according to claim 1, characterized in that: Step 1 includes: The STM32F4 main controller includes flash memory and random access memory; The STM32F4 main controller sends a reset signal to the face recognition detection module and waits for a ready response to confirm that the power supply voltage of the face recognition detection module is stable and the data interface is in an idle state. Then, the fingerprint recognition detection module is configured with registers, the acquisition mode is set to continuous trigger, the template storage area address is mapped to random access memory, and the preset matching threshold parameters are loaded. The STM32F4 main controller initializes the channel selection of the two video switching chips in the video isolation management module, sets the default output channel to the on state, and verifies the switching control pin level logic. For the network isolation management module, the STM32F4 main controller detects the driving capability of the electromagnetic relay drive circuit to ensure that the contacts can be reliably engaged or disengaged under subsequent control commands. In addition, the STM32F4 main controller also establishes an interrupt service routine entry point, which is bound to the face recognition timer interrupt, fingerprint recognition event interrupt, and status indicator update task, respectively.

3. The desktop computer information security management method according to claim 1, characterized in that: Step 2 includes the following steps: The face recognition detection module includes an image sensor and an OV7670 camera, and transmits image data with the STM32F4 main controller through a 16-bit parallel data interface; During the data acquisition phase, the STM32F4 main controller sends a start acquisition command to the face recognition detection module through the general input / output interface. The OV7670 camera then activates the light source and begins scanning line by line to generate the raw image data stream. The STM32F4 main controller triggers image preprocessing immediately after receiving the raw image data stream. The image preprocessing includes automatic white balance correction, gamma correction, and grayscale conversion, which are used to restore color, enhance details, and simplify data in the raw image data stream to generate a grayscale image. Then, a cascaded classifier algorithm based on Haar features is called to detect face regions in the grayscale image. The cascaded classifier algorithm is composed of multiple strong classifiers connected in series. Each level is a weighted combination of weak classifiers trained by AdaBoost. It only enters the next level of detection when the current level determines that the face is a candidate region. During the detection process, the cascaded classifier algorithm traverses the sliding window positions in the grayscale image. The window size starts at 24×24 pixels and is gradually enlarged by a scaling factor of 1.1 until it covers the entire image. If a face region that conforms to the Haar feature distribution is detected in any window, the center coordinates and scale parameters of the face region are recorded and marked as a valid face feature. The entire facial feature analysis process is carried out with the support of a single-user facial feature template library, which stores only the frontal face feature vector of an authorized user. If a valid facial feature is detected within a single cycle, the STM32F4 main controller marks the cycle as the user is present; otherwise, it marks the cycle as the user has left the seat.

4. The desktop computer information security management method according to claim 3, characterized in that: Step 3 includes the following steps: The video isolation management module includes a CV10211E1 chip, a CV10211E2 chip, a DP port, an HDMI interface, and a VGA interface. The input terminals of the CV10211E1 chip and the CV10211E2 chip are respectively connected to the corresponding video output interfaces of the desktop computer host, and the output terminals are connected to an external monitor. Under normal use, the STM32F4 main controller transmits low-level signals to the input terminals of the CV10211E1 chip and the CV10211E2 chip through general-purpose input / output pins, keeping their internal analog switches on and allowing the video signal to pass through without loss. Meanwhile, the network isolation management module uses an electromagnetic relay to control the physical layer network connection. In the disconnected state, the network interface resistance is greater than 100 megohms. The normally open contact of the electromagnetic relay is connected in series in the Ethernet physical layer line of the desktop computer host. The STM32F4 main controller controls the relay coil current through a driving transistor. During the user's continuous presence, the STM32F4 main controller maintains the coil energized, the contacts are closed, and the network signal is transmitted normally. In addition, when the user is present, the STM32F4 main controller synchronously updates the status indicator module, driving the green LED to remain constantly lit, providing the user with intuitive feedback that the system is in a safe operating state.

5. The desktop computer information security management method according to claim 4, characterized in that: Step 4 includes the following steps: The user leaving the seat determination condition is that no valid facial feature is detected for two consecutive monitoring cycles. The STM32F4 main controller is configured with a counter to record the number of consecutive user leaving the seat cycles. Whenever a monitoring cycle ends and no valid facial feature is detected, the counter is incremented by 1. If a valid facial feature is detected, the counter is reset to zero. When the counter value reaches 2, the STM32F4 main controller immediately triggers the user leaving the seat determination. Upon detecting that the user has left their seat, the STM32F4 main controller first outputs a high-level signal to the input of the video isolation management module. The internal analog switches of the CV10211E1 and CV10211E2 chips quickly disconnect, cutting off the video signal path, and the display immediately enters a black screen state with no signal. At the same time, the STM32F4 main controller shuts off the drive current of the electromagnetic relay, and the relay contacts spring open under the action of spring force, physically disconnecting the Ethernet line, so that the external network cannot communicate with the host through this interface. In the disconnected state, the network interface resistance is greater than 100 megohms, cutting off the underlying network attack path. After protection is activated, the red LED of the STM32F4 main controller drive status indicator module will remain lit, alerting the user that the system has entered protection mode.

6. The desktop computer information security management method according to claim 5, characterized in that: Step 5 includes the following steps: The fingerprint recognition detection module uses an optical fingerprint sensor. When a user's finger touches the sensing area of ​​the fingerprint recognition detection module, the optical sensor immediately captures the light and dark image formed by the fingerprint ridges and valleys, generating a fingerprint grayscale image. A minutiae matching algorithm is used to identify the endpoints and bifurcation points in the fingerprint grayscale image and extract their coordinates, orientation angles, and type information to form a feature point set. The feature point set is compared with the internally stored registered fingerprint templates to calculate the similarity score between the two. If the score exceeds the matching threshold, the user is deemed a legitimate user. After successful verification, the STM32F4 main controller immediately sends a connection command to the video isolation management module and the network isolation management module to restore the video and network connection and drive the status indicator module to switch to a solid green light.

7. The desktop computer information security management method according to claim 6, characterized in that: The desktop computer is equipped with a dual-mode recovery mechanism, which retains the periodic automatic recovery function for facial recognition in addition to the fast recovery of fingerprint recognition. In the protected state, the STM32F4 main controller will still periodically wake up the facial recognition detection module to perform a fast scan. If the facial features of an authorized user are detected, the recovery process will also be triggered to ensure that normal use can still be restored through facial recognition when the fingerprint module fails.

8. A desktop computer information security management system, employing the desktop computer information security management method as described in claims 1-7, characterized in that: The system includes a face recognition detection module, a fingerprint recognition detection module, a video isolation management module, a network isolation management module, an STM32F4 main controller, and a status indication module; The face recognition detection module is used to acquire and preprocess images of the area in front of the computer, and to detect face regions in grayscale images based on a cascaded classifier algorithm to determine the user's presence in real time. The fingerprint recognition detection module is used to capture the user's fingerprint image through an optical fingerprint sensor and to perform feature extraction and matching using a minutiae matching algorithm to verify the user's identity. The video isolation management module is used to connect or disconnect the video signal path under the control of the STM32F4 main controller, so as to realize the normal transmission of video signal of the display or the black screen state. The network isolation management module is used to connect or disconnect the physical network connection under the control of the STM32F4 main controller to ensure that the network connection remains unobstructed or physically disconnected. The STM32F4 main controller is used for system initialization and configuration of parameters of each module, establishing communication links and control logic, periodically triggering face status monitoring and controlling video and network isolation modules according to the results, and processing fingerprint verification to quickly restore the system to normal state. The status indicator module is used to adjust the LED indicator light according to the number of users present.

Images