A classroom electrical appliance energy-saving control device based on video image detection

Abstract

The present application relates to the technical field of electrical control and image processing, and particularly relates to a classroom electrical appliance energy-saving control device based on video image detection, which comprises a classroom monitoring server, an energy-saving control terminal and a campus network. The classroom monitoring server acquires classroom monitoring images, processes the images and judges whether there is anyone in the classroom. The energy-saving control terminal communicates with the classroom monitoring server, and after receiving a control instruction, the energy-saving control terminal adjusts the current in the electrical circuit to make the air switch reach the overcurrent action condition, and triggers the air switch to be turned off. The campus network is used for realizing data transmission between the classroom monitoring server and the energy-saving control terminal, and connecting classroom monitoring camera equipment. The present application realizes automatic energy-saving control of classroom electrical appliance equipment; without additional human body detection hardware, the system deployment cost is reduced; meanwhile, the original use mode and safety characteristics of the electrical system are maintained, the randomness of classroom use is adapted to, and energy waste is reduced.

Description

Technical Field

[0001] This invention relates to the fields of electrical control and image processing technology, specifically to a classroom electrical energy-saving control device based on video image detection. Background Technology

[0002] With the expansion of higher education institutions and the development of campus information technology, classrooms are generally equipped with lighting, air conditioning, multimedia equipment, and computers. Most campuses also have comprehensive monitoring systems and campus networks. However, due to the highly mobile and unpredictable nature of classroom use, with people frequently entering and leaving at different times, lighting, air conditioning, and other electrical appliances continue to run even when classrooms are unoccupied, resulting in energy waste.

[0003] Existing technologies typically employ infrared sensors, timer control devices, or smart home control methods for appliance management. However, infrared sensors are susceptible to changes in ambient temperature and obstructions, resulting in insufficient detection stability, and require additional hardware. Timer control relies on preset times, making it difficult to adapt to the randomness of classroom usage. Smart home control depends on equipment and applications from different manufacturers, leading to poor system compatibility and limited adaptability to existing appliances. Furthermore, some technologies achieve electrical control by replacing or modifying traditional circuit breakers, but this usually requires adding extra sensors or actuators, altering the original circuit breaker structure, resulting in high equipment costs, poor versatility, and difficulty in widespread application within existing electrical systems.

[0004] Under current conditions, the campus already possesses surveillance camera equipment and network infrastructure, but these resources have not been effectively utilized for electrical energy-saving control. Therefore, how to determine the status of people in classrooms using existing surveillance images without adding extra detection hardware or altering the original circuit breaker structure, and how to implement power-off control of electrical circuits when conditions are met, in order to adapt to the randomness of classroom use and reduce energy waste, has become a technical problem that needs to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a classroom electrical energy-saving control device based on video image detection to solve the above-mentioned technical problems, thereby realizing automated energy-saving control of classroom electrical equipment; reducing system deployment costs without the need for additional human body detection hardware; and maintaining the original usage and safety characteristics of the electrical system, adapting to the randomness of classroom use and reducing energy waste.

[0006] The objective of this invention can be achieved through the following technical solutions: A classroom electrical energy-saving control device based on video image detection, including The classroom monitoring server is used to periodically acquire monitoring images of the classroom, process the monitoring images and determine whether there are people in the classroom. When it is determined that there are no people in the classroom, it generates control commands in combination with the preset energy-saving time period. The energy-saving control terminal is installed in the classroom electrical circuit and communicates with the classroom monitoring server through the campus network to receive control commands. After receiving the control command, the energy-saving control terminal controls the current in the electrical circuit to make the air switch reach the overcurrent action condition, triggering the air switch to open and cut off the power supply to the classroom electrical appliances. The campus network is used for data transmission between the classroom monitoring server and the energy-saving control terminal. The classroom monitoring server is connected to the monitoring camera equipment in the classroom through the campus network.

[0007] Furthermore, the classroom monitoring server includes an image acquisition module, an image processing and analysis module, a logic judgment module, and an instruction sending module. The image acquisition module is connected to the monitoring camera equipment in the classroom through the campus network to acquire classroom monitoring images at regular intervals. The image processing and analysis module is used to preprocess the monitoring images and detect personnel. The logic judgment module makes comprehensive decisions based on the personnel detection results and the system clock. The instruction sending module is used to convert the comprehensive decisions into control instructions and send them to the energy-saving control terminal through the campus network.

[0008] Furthermore, the energy-saving control terminal includes a microcontroller, a controllable switch, a rectifier, and a load controller. The microcontroller is used to receive output control commands and generate control signals to control the controllable switch to turn on. The rectifier is used to provide DC voltage to the load controller. The load controller includes an inductor and a power transistor. The inductor and the power transistor are connected in series to form a current regulation loop and are connected to the branch where the controllable switch is located. The microcontroller outputs a pulse signal with a set frequency to the power transistor.

[0009] Furthermore, the image preprocessing module performs the following processing steps on the monitoring images: grayscale conversion and noise reduction; for monitoring images under dim lighting conditions, the grayscale values ​​are transformed from a non-uniform distribution concentrated in a certain grayscale region to a uniform distribution using a histogram equalization algorithm.

[0010] Furthermore, the personnel testing process includes: Human targets are identified in preprocessed images using a deep learning-based object detection model. The AM-FM image decomposition method is introduced to extract the instantaneous frequency components of the image; The LeNet classifier is used to assist in the detection of features of human body parts such as hair and back of the head; The instantaneous frequency components of the AM-FM image decomposition method are fused with the detection output of the deep learning target detection model for judgment. When any detection channel confirms the presence of human features, it is determined that there is a person.

[0011] Furthermore, the personnel testing process also includes: Spot filtering is performed using a multi-feature judgment method. The grayscale value and background difference value of the monitoring image are collected, and grayscale threshold, difference threshold and size threshold are set. The size threshold is set to the normal human body size range. If the grayscale value exceeds the grayscale threshold or the difference value with the background exceeds the difference threshold and the size exceeds the normal human body range, it is judged as a spot and the spot is filtered out.

[0012] Furthermore, the workflow of the logic judgment module and the instruction sending module includes: When the personnel detection result is no one, the current system time is obtained, and the current system time is compared with multiple preset energy-saving time periods, wherein the energy-saving time period includes at least one preset time interval; When the current system time is within any energy-saving time period, a power-off control command is output; when the current system time is not within an energy-saving time period or when personnel are detected and identified as being present, no power-off control command is output. Generate a data packet containing the classroom ID and power outage control command, and send the data packet to the energy-saving control terminal of the corresponding classroom via the campus network's TCP / IP protocol.

[0013] Furthermore, the workflow of the energy-saving control terminal includes: After receiving the power-off control command from the classroom monitoring server, the microcontroller outputs a control signal to turn on the controllable switch in the branch connected between the air switch output and the load. After the controllable switch is turned on, the electrical energy in the electrical circuit is converted by the rectifier and applied to the current regulation circuit composed of the inductor and the power tube. The microcontroller outputs pulse signals to the power transistor to control the power transistor to periodically turn on and off, causing the current in the current regulation circuit to change and thus driving the branch current to change. When the current in the controllable switch branch exceeds the overcurrent threshold of the air switch in the line, the air switch enters the overcurrent operation state and thermally trips.

[0014] Furthermore, energy-saving time periods are preset and adjusted according to the school's schedule, and special settings are made for holidays.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention acquires and analyzes classroom monitoring images, and combines personnel detection results with preset energy-saving time periods to achieve automatic power-off control of the classroom when it is unoccupied, adapting to the randomness of classroom use. 2. This invention utilizes existing surveillance camera equipment on campus for personnel status recognition, achieving a control method that eliminates the need for additional human detection hardware, thus reducing system deployment costs; and by connecting classroom monitoring servers and energy-saving control terminals through the campus network, it enables centralized control of electrical equipment in multiple classrooms. 3. This invention achieves electrical circuit disconnection control by adjusting the branch current in the existing electrical circuit to trigger the operation of the air switch, without requiring modification to the original air switch structure; Attached Figure Description

[0016] Figure 1 This is a structural block diagram of a classroom electrical energy-saving control device based on video image detection. Figure 2 A flowchart of the classroom monitoring server's workflow; Figure 3 This is a diagram of the internal structure of the energy-saving control terminal.

[0017] The components include: 1. Classroom monitoring server; 11. Image acquisition module; 12. Image processing and analysis module; 13. Logic judgment module; 14. Command sending module; 2. Campus network; 3. Control terminal; 31. MCU; 32. Controllable switch; 33. Rectifier; 34. Load controller; 35. WIFI module. Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0019] like Figure 1 and Figure 2 The classroom electrical energy-saving control device shown includes: Classroom monitoring server 1 is used to periodically acquire classroom surveillance images, process the images, and determine whether anyone is in the classroom. When no one is detected, it generates control commands based on a preset energy-saving time period. By processing the classroom surveillance images and determining the presence of people, and generating control commands based on the preset energy-saving time period, it achieves automatic power-off control for classrooms when no one is present, solving the problem of electrical equipment running idle for extended periods due to high personnel turnover. By utilizing existing surveillance cameras on campus to acquire and analyze image information, it achieves personnel status determination without the need for additional human detection devices, reducing system deployment costs.

[0020] Energy-saving control terminal 3 is installed in the classroom electrical circuit and communicates with the classroom monitoring server 1 through the campus network 2 to receive control commands. After receiving the control command, the energy-saving control terminal 3 controls the current in the electrical circuit to make the air switch reach the overcurrent action condition, triggering the air switch to open and cut off the power supply to the classroom electrical appliances. By connecting the classroom monitoring server 1 and the energy-saving control terminal 3 through the campus network 2, centralized control and unified management of electrical equipment in multiple classrooms can be realized. By adjusting the branch current in the existing electrical circuit to trigger the air switch, the electrical circuit disconnection control can be realized. Campus network 2 is used for data transmission between classroom monitoring server 1 and energy-saving control terminal 3. Classroom monitoring server 1 is connected to the monitoring camera equipment in the classroom through campus network 2.

[0021] In this embodiment, without adding additional human detection hardware, the existing surveillance camera images are used to determine the status of people in the classroom, and the electrical circuits are controlled when no one is present and it is during an energy-saving period. At the same time, the power-off control of electrical appliances is achieved by adjusting the branch current to trigger the air switch, avoiding modification of the original air switch structure and reducing equipment replacement and modification costs.

[0022] Specifically, in this embodiment, the classroom monitoring server 1 includes an image acquisition module 11, an image processing and analysis module 12, a logic judgment module 13, and an instruction sending module 14. The image acquisition module 11 is connected to the monitoring camera equipment in the classroom via the campus network 2 and is used to acquire classroom monitoring images at regular intervals. The image processing and analysis module 12 is used to preprocess the monitoring images and detect personnel. The logic judgment module 13 makes a comprehensive decision based on the personnel detection results and the system clock. The instruction sending module 14 is used to convert the comprehensive decision into control instructions and send them to the energy-saving control terminal 3 via the campus network 2. The classroom monitoring server 1 and the energy-saving control terminal 3 communicate via a wired network or a Wi-Fi module 35.

[0023] Specifically, in this embodiment, the energy-saving control terminal 3 includes a microcontroller 31, a controllable switch 32, a rectifier 33, and a load controller 34. The microcontroller 31 is used to receive output control commands and generate control signals to control the controllable switch 32 to turn on. The rectifier 33 is used to provide DC voltage to the load controller 34. The load controller 34 includes an inductor H1 and a power transistor LQ. The inductor H1 and the power transistor LQ are connected in series to form a current regulation loop and are connected to the branch where the controllable switch 32 is located. The microcontroller 31 outputs a pulse signal with a set frequency to the power transistor LQ.

[0024] Specifically, classroom monitoring server 1, according to pre-set scheduled tasks, initiates image capture requests to the monitoring cameras in the target classroom via campus network 2 at key time points each day or according to a fixed cycle, acquiring JPEG format image data of the current classroom. The image preprocessing module's image processing flow includes: grayscale conversion and noise reduction. After receiving the monitoring image, the image preprocessing module converts the color image to a grayscale image to reduce the amount of data computation. A Gaussian filtering algorithm is used to reduce noise in the image, removing interference caused by noise from the camera sensor or slight changes in lighting. For monitoring images under dim lighting conditions, a histogram equalization algorithm is used to transform the grayscale values ​​from a non-uniform distribution concentrated in a certain grayscale area to a uniform distribution, expanding the dynamic range of pixel grayscale values, enhancing image contrast, and making it easier to distinguish people from the background.

[0025] Specifically, the personnel testing process includes: Human targets are identified in preprocessed images using a deep learning object detection model; the deep learning object detection model is based on YOLOv4 or YOLOv7. (This is represented as...) ;in The maximum confidence level is calculated by representing the number of targets to be detected (N). This is used to represent the detection result intensity of the most likely human target in the current image. This enables the detection of the overall contour and salient features of a human body in a classroom scene, obtaining the target location and corresponding confidence information.

[0026] To address potential facial occlusion in classroom environments (such as students looking down at books or facing away from the camera) or insufficient lighting, the AM-FM image decomposition method is introduced to extract the instantaneous frequency components of the image. The calculation formula is as follows: ,in It represents the instantaneous amplitude, reflecting the local brightness or energy distribution of the image; Indicates instantaneous phase; instantaneous frequency component Given by the gradient of the phase function: ,in This represents an instantaneous frequency vector, used to describe local texture changes in an image. The spatial gradient representing the phase, and the instantaneous frequency component can characterize the fine-grained texture features of areas such as hair and the back of the head.

[0027] The LeNet classifier is used to assist in the detection of features on human body parts such as hair and the back of the head. The local feature regions extracted by AM-FM are input into the LeNet classifier to obtain the auxiliary detection results. , It represents a classification function; when the overall outline of the human body is not obvious, it serves as auxiliary information in the judgment, providing supplementary basis for personnel detection.

[0028] The instantaneous frequency components of the AM-FM image decomposition method are fused with the detection output of the deep learning object detection model for judgment, i.e., the AM-FM assisted detection result is obtained. maximum confidence level Calculations are performed after fusion: ; P represents the unified judgment value after fusion, and T represents the judgment threshold. When any detection channel confirms the presence of human characteristics, it is judged as a person.

[0029] Specifically, the personnel testing process also includes: Spot filtering is performed using a multi-feature judgment method. Grayscale values ​​and background difference values ​​of the monitoring image are collected, and a grayscale threshold is set. Used to filter areas with abnormally high brightness and differential thresholds. Used to detect regions that differ significantly from the background; and size threshold. The size threshold is set to the normal human body size range; This represents the grayscale value of the current image pixel. The grayscale values ​​of the background image pixels. Let the size of the k-th connected region be ; calculate Background difference value. For the k-th detected candidate region, define a decision function: in, Indicates the average gray value of the area. This represents the average difference between regions; =1: Identify it as a light spot and remove it; =0: Retained as a valid target; If its grayscale value exceeds the grayscale threshold or the difference between it and the background exceeds the difference threshold and its size exceeds the normal range for a human body, it is determined to be a light spot and is filtered out. Preferably, the grayscale threshold is set to 200 and the difference threshold is set to 30.

[0030] Specifically, the workflow of the logic judgment module and the instruction sending module includes: When the personnel detection result is no one, the current system time is obtained, and the current system time is compared with multiple preset energy-saving time periods, wherein the energy-saving time period includes at least one preset time interval; When the current system time is within any energy-saving time period, a power-off control command is output; when the current system time is not within an energy-saving time period or when personnel are detected and identified as being present, no power-off control command is output. A data packet containing the classroom ID and power outage control command is generated and sent to the corresponding classroom's energy-saving control terminal via the campus network's TCP / IP protocol. In this embodiment, the energy-saving time periods are set as follows: midday energy-saving period 12:00-14:00, evening energy-saving period 17:30-19:00, and nighttime energy-saving period 21:30-06:00 the next day. If the current time falls within any of these three time periods, the trigger condition is met. If the detection result is "occupied" or the time is not within the preset time period, no operation is performed, and the system waits for the next scheduled data collection by the teacher.

[0031] Specifically, the workflow of the energy-saving control terminal includes: After receiving the power-off control command from the classroom monitoring server, the microcontroller outputs a control signal to turn on the controllable switch in the branch connecting the air switch output terminal and the load. After the controllable switch is turned on, the electrical energy in the electrical circuit is converted by the rectifier and applied to the current regulation circuit composed of the inductor and the power transistor. The microcontroller outputs a pulse signal to the power transistor to control the power transistor to periodically turn on and off, causing the current in the current regulation circuit to change and thus driving the branch current to change. When the current in the controllable switch branch exceeds the overcurrent threshold of the air switch in the circuit, the air switch enters the overcurrent action state and thermally trips.

[0032] like Figure 3 As shown, S1, S2, and S3 represent the air switches installed in the circuit, and L1, L2, and L3 represent the loads in the circuit. A wire connects the output of the conventional air switch to the load appliance and then to the controllable switch 32. The microcontroller 31 outputs a control signal to turn on the controllable switch 32. The DC voltage obtained by the rectifier 33 is applied to the two ends of the load controller 34. The load controller 34 includes an inductor. A power transistor LQ and a microcontroller MCU31 generate a pulse signal with frequency f, which is applied to the power transistor LQ of the load controller. Under the action of the pulse signal, the inductor... resistance to The current flowing through the 32 branches of the controllable switch is Where U is the power supply voltage of the load controller line, and the microcontroller 31 adjusts the inductance by changing the frequency f. The inductive reactance changes the magnitude of the branch current. When the branch current exceeds the overcurrent threshold of the air switch in the line, the thermal trip of the air switch will activate, and the line will be broken.

[0033] Specifically, energy-saving time periods are preset and adjusted according to the school's schedule, and special arrangements are made for holidays to adapt to classroom usage at different times.

[0034] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0035] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0036] The present invention has been further described above with reference to specific embodiments. However, it should be understood that the specific description herein should not be construed as limiting the nature and scope of the present invention. Various modifications made to the above embodiments by those skilled in the art after reading this specification are all within the scope of protection of the present invention.

Claims

1. A classroom electrical energy-saving control device based on video image detection, characterized in that, include The classroom monitoring server is used to periodically acquire monitoring images of the classroom, process the monitoring images and determine whether there is anyone in the classroom. When it is determined that there is no one in the classroom, a control command is generated in combination with a preset energy-saving time period. An energy-saving control terminal is installed in the classroom electrical circuit and communicates with the classroom monitoring server through the campus network. It is used to receive control commands. After receiving the control commands, the energy-saving control terminal controls the current in the electrical circuit to make the air switch reach the overcurrent action condition, triggering the air switch to open and cut off the power supply to the classroom electrical appliances. The campus network is used for data transmission between the classroom monitoring server and the energy-saving control terminal. The classroom monitoring server is connected to the monitoring camera equipment in the classroom through the campus network.

2. The classroom electrical energy-saving control device based on video image detection as described in claim 1, characterized in that, The classroom monitoring server includes an image acquisition module, an image processing and analysis module, a logic judgment module, and an instruction sending module. The image acquisition module is connected to the monitoring camera equipment in the classroom through the campus network and is used to acquire classroom monitoring images at regular intervals. The image processing and analysis module is used to preprocess the monitoring images and detect personnel. The logic judgment module makes a comprehensive decision based on the personnel detection results and the system clock. The instruction sending module is used to convert the comprehensive decision into control instructions and send them to the energy-saving control terminal through the campus network.

3. The classroom electrical energy-saving control device based on video image detection as described in claim 1, characterized in that, The energy-saving control terminal includes a microcontroller, a controllable switch, a rectifier, and a load controller. The microcontroller receives and outputs the control command and generates a control signal to control the controllable switch to turn on. The rectifier provides DC voltage to the load controller. The load controller includes an inductor and a power transistor. The inductor and the power transistor are connected in series to form a current regulation loop, which is connected to the branch where the controllable switch is located. The microcontroller outputs a pulse signal with a set frequency to the power transistor.

4. A classroom electrical energy-saving control device based on video image detection as described in claim 2, characterized in that, The image preprocessing module performs the following processing steps on the monitoring images: grayscale conversion and noise reduction; for monitoring images under dim lighting conditions, the grayscale values ​​are transformed from a non-uniform distribution concentrated in a certain grayscale region to a uniform distribution using a histogram equalization algorithm.

5. A classroom electrical energy-saving control device based on video image detection as described in claim 4, characterized in that, The personnel detection process includes: Human targets are identified in preprocessed images using a deep learning-based object detection model. The AM-FM image decomposition method is introduced to extract the instantaneous frequency components of the image; The LeNet classifier is used to assist in the detection of features of human body parts such as hair and back of the head; The instantaneous frequency component of the AM-FM image decomposition method is fused with the detection output of the deep learning target detection model for judgment. When any detection channel confirms the presence of human features, it is determined that there is a person.

6. A classroom electrical energy-saving control device based on video image detection as described in claim 5, characterized in that, The personnel detection process also includes: Spot filtering is performed using a multi-feature judgment method. The grayscale value and background difference value of the monitoring image are collected, and grayscale threshold, difference threshold and size threshold are set. The size threshold is set to the normal human body size range. If the grayscale value exceeds the grayscale threshold or the difference value with the background exceeds the difference threshold and the size exceeds the normal human body range, it is determined to be a spot and the spot is filtered out.

7. A classroom electrical energy-saving control device based on video image detection as described in claim 2, characterized in that, The workflow of the logic judgment module and the instruction sending module includes: When the result of the personnel detection is that no one is present, the current system time is obtained, and the current system time is compared with multiple preset energy-saving time periods, wherein the energy-saving time periods include at least one preset time interval. When the current system time is within any of the energy-saving time periods, a power-off control command is output; when the current system time is not within the energy-saving time periods or when personnel are detected and identified as being present, the power-off control command is not output. Generate a data packet containing the classroom ID and power outage control command, and send the data packet to the energy-saving control terminal of the corresponding classroom via the TCP / IP protocol of the campus network.

8. A classroom electrical energy-saving control device based on video image detection as described in claim 3, characterized in that, The workflow of the energy-saving control terminal includes: After receiving a power-off control command from the classroom monitoring server, the microcontroller outputs a control signal to turn on the controllable switch in the branch connecting the output of the air switch and the load. After the controllable switch is turned on, the electrical energy in the electrical circuit is converted by the rectifier and applied to the current regulation circuit composed of the inductor and the power transistor. The microcontroller outputs a pulse signal to the power transistor to control the power transistor to periodically turn on and off, causing the current in the current regulation circuit to change and thus driving the branch current to change. When the current in the controllable switch branch exceeds the overcurrent threshold of the air switch in the line, the air switch enters the overcurrent operation state and thermally trips.

9. A classroom electrical energy-saving control device based on video image detection as described in claim 7, characterized in that, The energy-saving time periods are preset and adjusted according to the school's schedule, and special settings are made for holidays.

Images