An electronic class list control system and method with emergency management and power supply deep isolation function

Abstract

The application discloses an electronic class board control system and method with emergency management and power supply deep isolation function, and belongs to the technical field of intelligent terminal control. The system comprises the following steps: boosting the input voltage through a boost circuit, providing a charging voltage difference for a battery by using a preset high potential and physically shutting off the battery discharge path; executing business operations and image acquisition under the main power supply domain; when an emergency trigger signal is monitored, activating the emergency power supply domain, constructing a mutual exclusion dead zone through a physical RC delay circuit, and first disconnecting the power supply of the main power supply domain to the shared storage resource and the bus connection in the dead zone, and then connecting the power supply of the emergency power supply domain and the bus connection. Through the application, the problem that the existing system cannot reliably take over the emergency resources and lacks physical isolation in power management when the main control is abnormal is solved.

Description

Technical Field

[0001] This invention relates to an electronic class sign control system and method with emergency management and deep power isolation functions, belonging to the technical field of intelligent terminal control. Background Technology

[0002] Electronic class signs, as important interactive terminals in smart campuses and intelligent buildings, typically integrate functions such as access control, attendance tracking, information dissemination, and image capture. In actual deployment, electronic class signs not only need to maintain stable operation under normal power supply, but also must reliably execute emergency commands (such as releasing access locks) and save critical status data in emergencies such as external power outages, main control system failures, or fires, in order to meet mandatory safety requirements for fire protection and security.

[0003] Existing electronic class sign control systems typically employ a single main control chip or a dual-core architecture of "main SoC + backup MCU" combined with a backup battery. In terms of power management, switching between the adapter and battery often relies on reverse connection protection diodes (which result in significant voltage drop and heat generation) or MOSFET switches controlled by the main control chip's GPIO. When the main control chip malfunctions, the GPIO state may become uncontrollable, causing battery power to flow back to the adapter or preventing switching to battery power. In multi-processor unit architectures, the main processing unit and emergency processing unit usually need to share resources such as Flash memory or real-time clock (RTC). Existing bus switching mechanisms heavily rely on software handshake protocols, whereby the main processing unit releases the bus and notifies the backup unit to take over via interrupts or software flags. However, when the main processing unit crashes due to electromagnetic interference, memory overflow, or program abnormalities, its internal software logic stagnates, failing to complete the handshake and bus release, causing the backup unit to be locked by the physical bus and unable to access shared resources. Furthermore, existing access control and fire alarm linkage systems often rely on the main control chip to parse signals and issue commands via software, posing a risk of software failure preventing door locks from releasing.

[0004] The common deficiency of the aforementioned existing technologies lies in the fact that the key safety logic of the system, such as power path management, resource scheduling, and emergency response, is highly dependent on the software operation status of the main control system, and the underlying hardware lacks physical-level isolation mechanisms and deterministic timing control. This leads to a core technical problem: how to eliminate dependence on software logic under extreme conditions such as abnormal main power supply domain or software failure of the main control unit, and achieve reliable isolation of power discharge paths and deterministic takeover of shared storage resources through purely hardware physical mechanisms, thereby ensuring the absolutely reliable execution of the electronic class sign's emergency functions.

[0005] Therefore, a new solution is needed to address this problem. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide an electronic class sign control system and method with emergency management and deep power isolation functions, which solves the problems of existing systems where emergency resources cannot be reliably taken over when the main control fails and power management lacks physical isolation.

[0007] The technical problem to be solved by this invention is achieved by the following technical solution:

[0008] A control method for an electronic class sign with emergency management and deep power isolation functions, characterized in that the electronic class sign includes a business processing unit and an emergency processing unit, wherein the business processing unit and the emergency processing unit share storage resources and are isolated from each other by power supply domain, and the method includes the following steps:

[0009] The input voltage of the external adapter is boosted to a preset high potential by a boost circuit. The preset high potential is used to provide a charging voltage difference for the backup battery pack. At the same time, the preset high potential is coupled to the gate of the switch in the battery discharge path so as to physically shut off the battery discharge path when the adapter is online.

[0010] Under the main power supply domain, the service processing unit performs normal service operations, which include controlling the image sensor powered by the main power supply domain to acquire images; the image sensor is electrically isolated from the emergency power supply domain of the emergency processing unit.

[0011] When an emergency trigger signal is detected, the emergency power supply domain is activated, and a mutual exclusion dead zone is constructed through a physical RC delay circuit. Within the mutual exclusion dead zone, the power supply and bus connection of the main power supply domain to the shared storage resources are first disconnected, and then the power supply and bus connection of the emergency power supply domain to the shared storage resources are reconnected, so that the emergency processing unit obtains absolute control over the shared storage resources.

[0012] The present invention is further configured such that: the shared storage resource includes a shared memory and a real-time clock (RTC);

[0013] The step of first disconnecting the power supply and bus connection of the main power supply domain to the shared storage resources within the mutual exclusion dead zone, and then reconnecting the power supply and bus connection of the emergency power supply domain to the shared storage resources, includes:

[0014] By utilizing the level threshold difference during the charging and discharging process of the physical RC delay circuit, the first load switch is controlled to disconnect the main power supply domain from the power supply rail of the shared storage resource, and the second load switch is controlled to connect the emergency power supply domain to the power supply rail.

[0015] Simultaneously, the dead-time timing control node signal output by the physical RC delay circuit is used to drive the multiplexer to forcibly switch the SPI pin of the shared memory from the service processing unit to the emergency processing unit;

[0016] By forcibly pulling down the gate potential of the I2C level conversion circuit on the service processing unit side through a hardware shielding circuit, the access channel of the service processing unit to the RTC is physically blocked.

[0017] The present invention is further configured to include an active isolation sampling step across power supply domains:

[0018] When the emergency processing unit is in an inactive state, the sampling circuit from the main power supply domain to the ADC pin of the emergency processing unit is physically disconnected by a high-level sampling switch to prevent current backflow.

[0019] When the emergency processing unit is activated, it actively outputs an enable signal to turn on the high-level sampling switch to collect the operating voltage status of the main power supply domain.

[0020] The present invention is further configured to include a fire alarm linkage step with hardware and software decoupling:

[0021] The external fire alarm signal is directly connected to the hardware disable terminal of the electric lock drive circuit. When the fire alarm signal is triggered, the door lock power circuit is immediately cut off from the physical level.

[0022] Meanwhile, the fire trigger status is synchronously fed back to the business processing unit through a high-impedance isolation bypass circuit, so that the business processing unit can maintain the execution of emergency log solidification during operation, thereby decoupling safety execution from logic management.

[0023] The present invention is further configured such that: the routine business operation performed by the business processing unit further includes: after the image sensor acquires the image, performing an image motion blur compensation step based on double exposure physical consistency verification, specifically including:

[0024] Acquire a first image and a second image, wherein the first image and the second image are image frames acquired by the image sensor in the same scene based on different exposure parameters;

[0025] Perform frequency domain transformation based on the first image and the second image to obtain the corresponding first spectrum information and second spectrum information;

[0026] The frequency domain attenuation characteristics are calculated based on the first and second spectral information, and consistency verification is performed based on the frequency domain attenuation characteristics to obtain the verification results.

[0027] Based on the verification results, an image compensation strategy is determined, and the compensated image is output.

[0028] The present invention is further configured such that: the calculation of frequency domain attenuation characteristics includes:

[0029] Calculate the first low-frequency band energy of the first spectral information in the preset low-frequency band region and the first high-frequency band energy in the preset high-frequency band region, and determine the first attenuation depth accordingly;

[0030] Calculate the second low-frequency band energy in the preset low-frequency band region and the second high-frequency band energy in the preset high-frequency band region based on the second spectral information, and determine the second attenuation depth accordingly;

[0031] The consistency verification includes: querying a preset mapping relationship based on the first attenuation depth and the second attenuation depth to obtain a first blur length estimate and a second blur length estimate; when the difference between the ratio of the first blur length estimate and the second blur length estimate and the ratio of the exposure time of the first image and the second image is less than a preset threshold, the verification result is determined to be passed.

[0032] The present invention is further configured such that when the verification result is passed, the image compensation strategy includes:

[0033] Perform directional projection analysis on the first or second spectral information to obtain an estimated motion direction.

[0034] A frequency domain representation of the fuzzy kernel is generated based on the first fuzzy length estimate or the second fuzzy length estimate and the motion direction estimate;

[0035] A Wiener filter operator is constructed based on the frequency domain representation of the fuzzy kernel and preset noise adjustment parameters;

[0036] The first image or the second image is converted to the frequency domain, and after point-by-point operation with the Wiener filter operator, it is converted back to the spatial domain to obtain the compensated image.

[0037] The present invention is further configured such that when the verification result is unsuccessful, the image compensation strategy includes:

[0038] Calculate the local variance of corresponding pixel positions in the first image and the second image;

[0039] The fusion weights of the first image and the second image at the same pixel position are determined based on the local variance.

[0040] The first image and the second image are weighted and summed according to the fusion weights to obtain a fused image as the compensated image.

[0041] The present invention is further configured such that: the method also includes a collaborative step between the image compensation strategy and the underlying hardware control;

[0042] During the execution of the image compensation strategy, when a power switching signal is detected, the execution of the image compensation strategy is paused, the intermediate parameters of the current processing state are stored in memory, and the execution is resumed according to the intermediate parameters after the power switching is completed.

[0043] When the fire alarm signal is detected, the execution of the image compensation strategy is immediately stopped, the unfinished processing results are discarded and the acquired images are saved to ensure that the door lock release operation of the hardware prohibition terminal has the highest priority.

[0044] An electronic class sign control system with emergency management and deep power isolation functions is characterized by comprising:

[0045] The electronic class sign itself;

[0046] A camera module, located inside the electronic class sign body, is used to capture images;

[0047] Both the business processing unit and the emergency processing unit are located inside the electronic class sign body. The business processing unit and the emergency processing unit share storage resources and are isolated from each other in terms of power supply domain. The camera module is powered by the main power supply domain where the business processing unit is located and is electrically isolated from the emergency power supply domain of the emergency processing unit.

[0048] The power management circuit includes a boost circuit and a battery discharge path switching transistor, which is used to boost the adapter input voltage to turn off the battery discharge path.

[0049] The hardware switching circuit, including a physical RC delay network and a multiplexer, is used to build a mutual exclusion dead zone and switch the bus connection of shared memory resources during emergency triggering.

[0050] The business processing unit includes a processor and a memory storing a computer program, wherein the processor executes the computer program to implement the above-described method.

[0051] The beneficial effects of this invention are:

[0052] 1. By boosting the adapter input voltage through a boost circuit and using this high potential to physically force the switch of the battery discharge path to be turned off, the reverse flow of battery power to the adapter is effectively prevented. At the same time, a millisecond-level seamless power supply switch is achieved when the adapter is removed.

[0053] 2. Upon detecting an emergency trigger signal, a mutual exclusion dead zone is constructed using a physical RC delay circuit. Utilizing the pure hardware charging and discharging timing, the power supply and bus connection of the main power domain to the shared storage resources are physically disconnected first, and then the emergency power domain is connected. This mechanism completely eliminates the dependence on the main control system software handshake protocol, ensuring that even in extreme cases of main control unit crashes or logical anomalies, the emergency processing unit can still obtain absolute control over the shared resources, significantly improving the reliability of the system's emergency response. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method of the present invention.

[0055] Figure 2 This is a flowchart of the image motion blur compensation sub-process of the present invention.

[0056] Figure 3 This is a flowchart of the emergency power supply domain activation and dead zone switching of the present invention.

[0057] Figure 4 This is the system architecture diagram of the present invention. Detailed Implementation

[0058] To facilitate a clear understanding of the technical means, creative features, objectives, and effects of this invention, the invention will be further described below in conjunction with specific illustrations.

[0059] To address the technical problems in existing technologies where electronic class signs used as access control terminals lack integrated power management, cannot reliably take over emergency resources after the main control system crashes, have a risk of failure due to reliance on software logic for fire alarm linkage, and cannot self-verify motion blur compensation results in image acquisition scenarios, this application provides an electronic class sign control system and method with emergency management and deep power isolation functions. (Reference) Figure 1 This method is applied to an electronic class sign, which includes a business processing unit and an emergency processing unit. The business processing unit and the emergency processing unit share storage resources and are isolated from each other by power supply domains. The method includes the following steps.

[0060] S1: Power supply boost and battery discharge path management.

[0061] After the electronic class sign is connected to the external adapter, the 12V voltage input from the external adapter is boosted to a preset high potential (exemplarily, the preset high potential is 15V) through a boost circuit. This preset high potential has two functions.

[0062] S11: As the input source of the charging management chip, it provides a charging voltage difference for the backup battery pack. The backup battery pack is exemplarily a 3-cell lithium battery pack with a full-charge voltage of 12.6V and a preset high potential of 15V, which is higher than 12.6V, thus enabling normal charging of the battery pack.

[0063] S12: The gate of the PMOS transistor coupled to the battery discharge path via a diode. When the adapter is online, this preset high potential forces the gate of the PMOS transistor to be in the off state through this coupling link, thereby physically shutting off the battery discharge path during adapter connection and preventing battery power from flowing back to the adapter.

[0064] S13: A 10kΩ pull-down resistor is integrated into the control link with a preset high potential. This pull-down resistor rapidly releases the charge accumulated on the PMOS transistor gate when the adapter is removed, achieving millisecond-level switching from adapter power to battery power. A physical manual switch is also provided on the outside of the chassis. This manual switch can physically cut off the gate control signal, enabling manual shutdown of the backup battery discharge path during equipment maintenance.

[0065] S2: The service processing unit under the main power supply domain performs normal service operations and image motion blur compensation based on dual-exposure physical consistency verification.

[0066] refer to Figure 2 Under the main power supply domain, the service processing unit performs normal service operations. Normal service operations include controlling the image sensor to acquire images. This image sensor is powered by the main power supply domain and is electrically isolated from the emergency power supply domain of the emergency processing unit. Normal service operations also include an image motion blur compensation step based on dual-exposure physical consistency verification. This compensation step is performed after the image sensor acquires the image, and specifically includes the following sub-steps S21 to S24.

[0067] S21: The image sensor acquires two frames of images in the same scene based on different exposure parameters, which are denoted as the first image and the second image, respectively.

[0068] S22: After acquiring the first image and the second image, perform a two-dimensional fast Fourier transform on the first image and the second image respectively to obtain the corresponding first amplitude spectrum and second amplitude spectrum, which are used as the first spectrum information and the second spectrum information respectively.

[0069] S23: Calculate the frequency domain attenuation characteristics based on the first and second frequency spectrum information, and perform consistency verification based on the frequency domain attenuation characteristics to obtain a verification result.

[0070] S24: Determine the image compensation strategy based on the verification result and output the compensated image.

[0071] When calculating the frequency domain attenuation characteristics, a circular region is divided with the center of the spectrum as the center and according to a preset low-frequency band radius as the preset low-frequency band region. An annular region is divided from the preset high-frequency band radius to the edge of the spectrum as the preset high-frequency band region. For the first amplitude spectrum, the sum of all amplitude values ​​within the preset low-frequency band region is calculated as the first low-frequency band energy, and the sum of all amplitude values ​​within the preset high-frequency band region is calculated as the first high-frequency band energy. Then, based on the first low-frequency band energy and the first high-frequency band energy, the formula is applied...

[0072]

[0073] Calculate the first attenuation depth ,in This indicates the energy of the first low-frequency band. This represents the energy of the first high-frequency band. The energy of the second low-frequency band is calculated in the same way for the second amplitude spectrum. Second high frequency band energy and according to the formula

[0074]

[0075] Calculate the second attenuation depth The values ​​of the first and second attenuation depths both range from -1 to 1.

[0076] During consistency verification, the system pre-stores a preset mapping relationship between attenuation depth and blur length, obtained through offline calibration. During calibration, a standard resolution test chart is mounted on a uniformly rotating platform, and multiple sets of images are captured at different exposure times under a known angular velocity. Each set of images corresponds to a known blur length. The attenuation depth of the short-exposure and long-exposure images in each set is calculated, and the corresponding data points for multiple sets of attenuation depths and blur lengths are recorded. Then, a monotonically increasing curve is fitted using cubic spline interpolation, and this curve is stored in the system as the preset mapping relationship. During online operation, the preset mapping relationship is queried based on the first attenuation depth to obtain the corresponding first blur length estimate, and the preset mapping relationship is queried based on the second attenuation depth to obtain the corresponding second blur length estimate. The ratio of the first blur length estimate to the second blur length estimate is then calculated, along with the ratio of the exposure time of the first image to the second image. When the difference between these two ratios is less than a preset threshold, the verification result is considered passed. The typical value of this preset threshold is 0.15, determined through error analysis during the calibration process described above. The principle behind this judgment is that if the blur in a double-exposure image is caused by the same constant-speed motion, then the ratio of the blur amount deduced from the attenuation depth is equal to the ratio of the exposure times; if this relationship does not hold, it indicates that the motion speed has changed significantly between the two exposures.

[0077] S241: When the verification result is passed, the first image compensation strategy is executed. First, the motion direction estimate is obtained. Specifically, the first amplitude spectrum or the second amplitude spectrum is projected along multiple candidate directions, the variance of the projection sequence in each direction is calculated, and the direction with the smallest variance is selected as the motion direction estimate. Then, based on the first or second blur length estimate and the motion direction estimate, a frequency domain representation of the blur kernel is generated, where the frequency domain response of the blur kernel is determined by the blur length and the motion direction. Subsequently, a Wiener filter operator is constructed based on the frequency domain representation of the blur kernel and a preset noise adjustment parameter. The noise adjustment parameter is the reciprocal of the signal-to-noise ratio, and its value ranges from 0.001 to 0.05. This parameter is adaptively adjusted according to the brightness level of the image: the global average brightness of the image is calculated, and 0.001 is taken when the average brightness is greater than 128, and 0.05 is taken when the average brightness is less than 64, with intermediate values ​​using linear interpolation. Finally, the first or second image is converted to the frequency domain using a two-dimensional fast Fourier transform, and then subjected to point-by-point complex multiplication with the Wiener filter operator. Finally, it is converted back to the spatial domain using an inverse two-dimensional fast Fourier transform to obtain the compensated image.

[0078] S242: When the verification result is unsuccessful, perform the following hierarchical rollback process.

[0079] S242a (First-level fallback – Adaptive fusion): Switches to the second image compensation strategy. Calculates the local variance of corresponding pixel positions in the first and second images within a preset window of size 3×3 pixels. Local variance reflects the strength of neighborhood texture; images with larger local variance receive higher fusion weights at that pixel position. Specifically, follow the formula...

[0080]

[0081] and

[0082]

[0083] Calculate the fusion weights, where and Let $x$ and $y$ represent the local variances of the first and second images at pixel position $(x,y)$, respectively. Use a small constant to prevent the denominator from being zero. Then follow the formula.

[0084]

[0085] The first and second images are weighted and summed to obtain a fused image, which is then used as the compensated image.

[0086] S242b (Second-Level Backoff – Quality Assessment and Reacquisition Guidance): Performs a no-reference image quality assessment on the fused image to obtain a quality score. This quality score is calculated using a sharpness evaluation function based on the product of gray-level variances. The calculation method is as follows: for image $I(x,y)$, calculate...

[0087]

[0088] Where $Q$ is the sharpness score, and $I(x,y)$ represents the gray value of the image at pixel position $(x,y)$. Represents the grayscale values ​​of horizontally adjacent pixels. The Q value represents the grayscale value of vertically adjacent pixels. A larger Q value indicates higher image sharpness. When the Q value is lower than a preset quality threshold, the fused image is deemed to be of insufficient quality, and a re-acquisition prompt signal is generated. This signal is used to prompt the user to adjust the exposure parameters or stabilize the device and re-acquire the image.

[0089] S242c (Third-level rollback - watermark mark): When the verification result is still unsuccessful after a preset number of consecutive re-acquisitions (exemplarily three times), the original acquired image is retained and a preset compensation failure watermark mark is superimposed on the original acquired image. The content of the watermark mark is "Original image - automatic compensation not effective", which is used to indicate that the image has not been compensated.

[0090] S3: Coordination between image compensation strategy and underlying hardware control.

[0091] During the execution of the image compensation strategy, the system continuously monitors power switching signals and fire alarm linkage signals.

[0092] S31: When a power switching signal is detected, it indicates that the external adapter has been removed and the system is switching from adapter power to battery power. At this time, the execution of the image compensation strategy is paused, and the intermediate parameters of the current processing state are stored in memory. These intermediate parameters include the completed frequency domain transformation results and the calculated intermediate variables. After the power switching is complete, i.e., after the PMOS transistor completes battery path conduction, the image compensation strategy is resumed based on the intermediate parameters stored in memory.

[0093] S32: Upon detecting a fire alarm signal, immediately halt the execution of the image compensation strategy, discard all unfinished processing results, save the acquired images (specifically, only save the original acquired images), and record the fire alarm trigger and image processing interruption events in the log. At this time, the electronic signboard's hardware cutoff circuit has the highest priority and executes the door lock release operation; the interruption of the image compensation strategy will not cause any delay in door lock release.

[0094] S4: Emergency Trigger and Emergency Power Supply Domain Activation Takeover.

[0095] S41: Reference Figure 3 When an emergency trigger signal is detected, the system activates the emergency power supply domain. The activation of the emergency power supply domain is independent of the system's current power source.

[0096] S42: Upon activation, a mutual exclusion dead time is constructed using a physical RC delay circuit. This physical RC delay circuit consists of a resistor and a capacitor connected in series, with the node connected to the input of a Schmitt trigger. The product of the resistor and capacitor determines the delay time. The Schmitt trigger converts the slow charging and discharging curve of the capacitor into a steep digital level transition. When the capacitor voltage reaches the positive threshold voltage of the Schmitt trigger, the output flips, thus outputting a control signal with a defined delay time, which is the dead time timing control node signal. By configuring the parameter values ​​of the resistor and capacitor, the time for the capacitor to charge to the positive threshold voltage is approximately 10ms.

[0097] S43: Within this mutual exclusion dead zone, utilizing the level threshold difference during the charging and discharging process of the physical RC delay circuit, the first load switch is controlled to disconnect the power supply rail from the main power supply domain to the shared memory resource, while the second load switch is controlled to connect the emergency power supply domain to the power supply rail. This disconnect-then-connect switching process generates a physical dead zone of approximately 10ms, which helps to avoid current backflow or level contention between the two independent power supply domains during the switching instant. Simultaneously, using this dead zone timing control node signal, the multiplexer is driven to force the SPI pin of the shared memory to switch from the service processing unit to the emergency processing unit. The shared memory resource includes the shared memory and the RTC. In a conventional dual-core system, the standby core's takeover of shared resources typically relies on a software handshake protocol. The successful execution of the software handshake depends on the main core still having responsiveness. When the main core loses responsiveness due to interference or program abnormalities, the software handshake cannot be completed. This embodiment implements resource takeover through a physical RC delay circuit and hard-wired logic, avoiding handshake failure and resource takeover failure caused by main core software abnormalities. Simultaneously, a hardware shielding circuit forces the gate potential of the I2C level conversion circuit on the service processing unit side down to ground, physically blocking the service processing unit's access to the RTC. For example, this hardware shielding circuit can be implemented using an NPN transistor (such as a 2N3904). After power switching, bus switching, and channel blocking, even if the service processing unit still has residual power or logic anomalies, it cannot interfere with the emergency processing unit's access to the shared storage resources. The emergency processing unit thus gains independent and absolute control over the shared storage resources.

[0098] S5: Active isolation sampling across power supply domains.

[0099] S51: When the emergency processing unit is inactive, a high-level sampling switch composed of a PMOS transistor physically disconnects the sampling circuit from the main power supply domain to the ADC pin of the emergency processing unit. The source of this high-level sampling switch receives the monitored main power supply domain, the drain is connected to the ADC pin of the emergency processing unit via a voltage divider resistor, and the gate is connected to a high level through a 100kΩ pull-up resistor and controlled by the general-purpose input / output port of the emergency processing unit. When the emergency processing unit is not powered on, its general-purpose input / output port is in a high-impedance state, the gate of the high-level sampling switch is pulled high by the pull-up resistor, and the switch is in the open state, thereby preventing the current from the main power supply domain from flowing back into the unpowered emergency processing unit through the protection diode inside the ADC pin.

[0100] S52: When the emergency processing unit is activated and powered on, the emergency processing unit actively outputs a low level from the general-purpose input / output port. This low level pulls the gate of the high-level sampling switch low, turns on the high-level sampling switch, and thus collects the working voltage status of the main power supply domain.

[0101] S6: Fire alarm linkage with hardware and software decoupling.

[0102] S61: The external fire alarm signal is directly connected to the hardware disable terminal of the electric lock drive circuit. This fire alarm signal is introduced through a 4-pin pluggable terminal block. The positive terminal of the fire alarm signal is directly connected to the system voltage via a fuse, while the negative terminal serves as the physical enable condition for the relay drive. When the fire alarm signal is triggered, the relay coil circuit is physically disconnected, and the drive-level chip immediately cuts off the door lock power circuit at the physical level. This response is independent of any software instructions or the microcontroller's operating state. In conventional solutions, the door lock release command is output by the microcontroller via software. If the microcontroller fails to execute the command due to an anomaly, the fire alarm linkage will fail. This embodiment achieves direct hardware-level response by directly connecting the fire alarm signal to the hardware disable terminal, without relying on software instructions.

[0103] S62: While the hardware executes the forced disconnection, a high-impedance isolated bypass circuit synchronously feeds back the fire alarm trigger status to the service processing unit. This high-impedance isolated bypass circuit has a bypass sampling circuit designed at the negative terminal of the fire alarm signal. It generates a fire status signal through a 100kΩ voltage divider resistor and connects it to the general-purpose input / output interrupt pin of the service processing unit. Upon receiving this feedback signal, the service processing unit, while maintaining normal operation, executes emergency log saving, records the specific time of the fire alarm trigger, and synchronizes the alarm information to the remote management platform via the network. This mechanism decouples the hardware layer's execution of door lock release from the software layer's recording of event status.

[0104] refer to Figure 4The present invention also provides an electronic class sign control system with emergency management and deep power isolation functions. The system includes an electronic class sign body, a camera module, a business processing unit, an emergency processing unit, a power management circuit, and a hardware switching circuit.

[0105] The electronic class sign body provides the physical support structure for the system. The camera module is located inside the electronic class sign body, with its image acquisition end facing outwards to capture images. The camera module is connected to the service processing unit via a data bus, transmitting the acquired image data to the service processing unit for processing.

[0106] Both the business processing unit and the emergency processing unit are located inside the electronic class sign body, and are powered separately through physically isolated power supply domains. The business processing unit is connected to the main power supply domain, and the emergency processing unit is connected to the emergency power supply domain. During operation, the business processing unit's internal functional logic includes: image acquisition control logic, used to execute the dual-exposure image acquisition control described in S21; frequency domain transformation logic, used to execute the two-dimensional fast Fourier transform described in S22; consistency verification logic, used to execute the frequency domain attenuation characteristic calculation and consistency verification described in S23; and compensation output logic, used to execute the first image compensation strategy described in S241 or the second image compensation strategy described in S242a.

[0107] The service processing unit and the emergency processing unit share storage resources via a bus switching circuit. These shared storage resources include shared memory and an RTC (Real-Time Controller). The clock and data signal lines of both units are connected to the shared storage resources via a multiplexer. The switching control terminal of this multiplexer is connected to the output of the physical RC delay circuit to receive dead-time timing control node signals. The power supply domains of the service processing unit and the emergency processing unit are isolated by load switches. A first load switch is connected in series between the main power supply domain and the power supply rail of the shared storage resource, and its enable terminal is connected to the first control output of the physical RC delay circuit. A second load switch is connected in series between the emergency power supply domain and the power supply rail of the shared storage resource, and its enable terminal is connected to the second control output of the physical RC delay circuit. The gate of the I2C level conversion circuit on the service processing unit side is also connected to a hardware shielding circuit. This hardware shielding circuit is, exemplarily, an NPN transistor (such as a 2N3904), whose base is connected to the status signal terminal of the emergency power supply domain, its collector is connected to the gate of the I2C level conversion circuit, and its emitter is grounded.

[0108] The power management circuit includes a boost converter, a charging management chip, and a battery discharge path switch. The input of the boost converter is connected to the external adapter interface, and its output is connected to the input of the charging management chip and the control terminal of the battery discharge path switch. The battery discharge path switch is connected in series between the positive output of the backup battery pack and the system's main power rail, with its gate connected as the control terminal to the output of the boost converter. When the adapter is online, a high-level signal at the output of the boost converter keeps this switch in the off state.

[0109] The hardware switching circuit includes a physical RC delay circuit (also known as a physical RC delay network) and a multiplexer. The input of the physical RC delay circuit is connected to the emergency trigger signal line, and its output is connected to the enable terminals of the first load switch, the second load switch, and the control terminal of the multiplexer. The physical RC delay circuit consists of resistors and capacitors connected in series, with the node connected to the input of a Schmitt trigger. The output of the Schmitt trigger serves as the output of the physical RC delay circuit. The multiplexer exemplarily uses the SN74CB3Q3257 chip, with its common terminal connected to the SPI pin of the shared memory, the first channel connected to the SPI pin of the service processing unit, and the second channel connected to the SPI pin of the emergency processing unit. The shared memory exemplarily uses the W25Q16 chip, and the RTC exemplarily uses the RX8025T chip. The I2C level conversion circuit exemplarily uses the BSS138PS chip. The high-order sampling switch exemplarily uses the SI2309 APMOS transistor.

[0110] In an alternative embodiment, the electronic class sign control system can be further divided into multiple functional modules in its logical architecture, including: a power management module for performing the boost and battery discharge path shutdown operations; a normal business and image compensation module for performing the double exposure image acquisition and image motion blur compensation strategy; an emergency takeover and dead zone control module for constructing a mutual exclusion dead zone through the physical RC delay circuit and performing a forced switch between power supply and bus when an emergency trigger signal is received; and a fire linkage decoupling module for performing the hardware-level door lock release and software-level log solidification operations.

[0111] The business processing unit includes a processor and a memory storing a computer program. When the processor executes the computer program, it implements the various steps of the electronic class sign control method with emergency management and deep power isolation functions.

[0112] The above values ​​and device models are for illustrative purposes only. Those skilled in the art can make adaptive adjustments based on actual circuit parameters and performance requirements, all of which fall within the protection scope of this invention.

[0113] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and various changes and modifications can be made without departing from the spirit and scope of the invention, all of which fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling an electronic class sign with emergency management and deep power isolation functions, characterized in that, The electronic class sign includes a business processing unit and an emergency processing unit. The business processing unit and the emergency processing unit share storage resources and are isolated from each other by power supply domain. The method includes the following steps: The input voltage of the external adapter is boosted to a preset high potential by a boost circuit. The preset high potential is used to provide a charging voltage difference for the backup battery pack. At the same time, the preset high potential is coupled to the gate of the switch in the battery discharge path so as to physically shut off the battery discharge path when the adapter is online. Under the main power supply domain, the service processing unit performs normal service operations, which include controlling the image sensor powered by the main power supply domain to acquire images; the image sensor is electrically isolated from the emergency power supply domain of the emergency processing unit. When an emergency trigger signal is detected, the emergency power supply domain is activated, and a mutual exclusion dead zone is constructed through a physical RC delay circuit. Within the mutual exclusion dead zone, the power supply and bus connection of the main power supply domain to the shared storage resources are first disconnected, and then the power supply and bus connection of the emergency power supply domain to the shared storage resources are reconnected, so that the emergency processing unit obtains absolute control over the shared storage resources. The shared storage resources include shared memory and a real-time clock (RTC). The step of first disconnecting the power supply and bus connection of the main power supply domain to the shared storage resources within the mutual exclusion dead zone, and then reconnecting the power supply and bus connection of the emergency power supply domain to the shared storage resources, includes: By utilizing the level threshold difference during the charging and discharging process of the physical RC delay circuit, the first load switch is controlled to disconnect the main power supply domain from the power supply rail of the shared storage resource, and the second load switch is controlled to connect the emergency power supply domain to the power supply rail. Simultaneously, the dead-time timing control node signal output by the physical RC delay circuit is used to drive the multiplexer to forcibly switch the SPI pin of the shared memory from the service processing unit to the emergency processing unit; By using a hardware shielding circuit, the gate potential of the I2C level conversion circuit on the service processing unit side is forcibly pulled low to physically block the service processing unit's access channel to the RTC.

2. The method according to claim 1, characterized in that, It also includes active isolation sampling steps across power supply domains: When the emergency processing unit is in an inactive state, the sampling circuit from the main power supply domain to the ADC pin of the emergency processing unit is physically disconnected by a high-level sampling switch to prevent current backflow. When the emergency processing unit is activated, it actively outputs an enable signal to turn on the high-level sampling switch to collect the operating voltage status of the main power supply domain.

3. The method according to claim 1, characterized in that, It also includes the fire alarm linkage steps with hardware and software decoupling: The external fire alarm signal is directly connected to the hardware disable terminal of the electric lock drive circuit. When the fire alarm signal is triggered, the door lock power circuit is immediately cut off from the physical level. Meanwhile, the fire trigger status is synchronously fed back to the business processing unit through a high-impedance isolation bypass circuit, so that the business processing unit can maintain the execution of emergency log solidification during operation, thereby decoupling safety execution from logic management.

4. The method according to claim 3, characterized in that, The execution of normal business operations by the business processing unit further includes: after the image sensor acquires the image, performing an image motion blur compensation step based on double exposure physical consistency verification, specifically including: Acquire a first image and a second image, wherein the first image and the second image are image frames acquired by the image sensor in the same scene based on different exposure parameters; Perform frequency domain transformation based on the first image and the second image to obtain the corresponding first spectrum information and second spectrum information; The frequency domain attenuation characteristics are calculated based on the first and second spectral information, and consistency verification is performed based on the frequency domain attenuation characteristics to obtain the verification results. Based on the verification results, an image compensation strategy is determined, and the compensated image is output.

5. The method according to claim 4, characterized in that, The calculation of frequency domain attenuation characteristics includes: Calculate the first low-frequency band energy of the first spectral information in the preset low-frequency band region and the first high-frequency band energy in the preset high-frequency band region, and determine the first attenuation depth accordingly; Calculate the second low-frequency band energy in the preset low-frequency band region and the second high-frequency band energy in the preset high-frequency band region based on the second spectral information, and determine the second attenuation depth accordingly; The consistency verification includes: querying a preset mapping relationship based on the first attenuation depth and the second attenuation depth to obtain a first blur length estimate and a second blur length estimate; when the difference between the ratio of the first blur length estimate and the second blur length estimate and the ratio of the exposure time of the first image and the second image is less than a preset threshold, the verification result is determined to be passed.

6. The method according to claim 5, characterized in that, When the verification result is passed, the image compensation strategy includes: Perform directional projection analysis on the first or second spectral information to obtain an estimated motion direction. A frequency domain representation of the fuzzy kernel is generated based on the first fuzzy length estimate or the second fuzzy length estimate and the motion direction estimate; A Wiener filter operator is constructed based on the frequency domain representation of the fuzzy kernel and preset noise adjustment parameters; The first image or the second image is converted to the frequency domain, and after point-by-point operation with the Wiener filter operator, it is converted back to the spatial domain to obtain the compensated image.

7. The method according to claim 5, characterized in that, When the verification result is unsuccessful, the image compensation strategy includes: Calculate the local variance of corresponding pixel positions in the first image and the second image; The fusion weights of the first image and the second image at the same pixel position are determined based on the local variance. The first image and the second image are weighted and summed according to the fusion weights to obtain a fused image as the compensated image.

8. The method according to claim 4, characterized in that, The method also includes a collaborative step between image compensation strategies and underlying hardware control: During the execution of the image compensation strategy, when a power switching signal is detected, the execution of the image compensation strategy is paused, the intermediate parameters of the current processing state are stored in memory, and the execution is resumed according to the intermediate parameters after the power switching is completed. When the fire alarm signal is detected, the execution of the image compensation strategy is immediately stopped, the unfinished processing results are discarded and the acquired images are saved to ensure that the door lock release operation of the hardware prohibition terminal has the highest priority.

9. An electronic class sign control system with emergency management and deep power isolation functions, characterized in that, include: The electronic class sign itself; A camera module, located inside the electronic class sign body, is used to capture images; Both the business processing unit and the emergency processing unit are located inside the electronic class sign body. The business processing unit and the emergency processing unit share storage resources and are isolated from each other in terms of power supply domain. The camera module is powered by the main power supply domain where the business processing unit is located and is electrically isolated from the emergency power supply domain of the emergency processing unit. The power management circuit includes a boost circuit and a battery discharge path switching transistor, which is used to boost the adapter input voltage to turn off the battery discharge path. The hardware switching circuit, including a physical RC delay network and a multiplexer, is used to build a mutual exclusion dead zone and switch the bus connection of shared memory resources during emergency triggering. The business processing unit includes a processor and a memory storing a computer program, wherein the processor executes the computer program to implement the method as described in any one of claims 1 to 8.

Images