LED eye protection lighting control method and system

Abstract

The present invention relates to the field of lighting control technology and discloses an LED eye-protection lighting control method and system. The method obtains a panoramic bird's-eye view of the room and uses a pre-built lighting control area detection model to identify three types of lighting control areas: the display device area (ROI1), the operating plane area (ROI2), and the environmental transition zone area (ROI3). Subsequently, the status of the ROI1 and ROI2 areas is identified, and corresponding lighting control strategies are generated based on the identification results to achieve precise lighting control of different areas. The present invention can effectively solve the problem of conflicting lighting requirements between the display area and the working area in traditional lighting solutions, avoid screen reflections and uneven desktop illumination caused by changes in ambient light, and significantly improve users' visual comfort in various scenarios.

Description

Technical Field

[0001] The present invention relates to the field of lighting control technology, and more specifically, to a method and system for controlling LED eye-protection lighting. Background Art

[0002] As people's demands for visual health and lighting quality continue to increase, the application of lighting control technology in various indoor scenarios has become increasingly critical. Although existing lighting control systems can meet basic lighting needs to a certain extent, they still have many limitations when facing complex and changing indoor environments and the special lighting requirements of different functional areas.

[0003] Chinese patent application CN103124454B proposes a lighting control device. Its core principle is to use a human information detection unit to determine the presence and movement speed of a person within a lighting zone based on image data captured by a camera. Based on this information, the lighting zone is divided into occupied and unoccupied areas, and differentiated lighting control is implemented in these areas. However, this lighting control device oversimplifies its assessment of human activity, relying solely on movement speed. This makes it difficult to accurately identify the specific lighting requirements of different functional areas in a room (such as display areas and operating surfaces), hindering the demand for refined lighting control in modern indoor environments.

[0004] A Chinese patent with authorization announcement number CN117452834B discloses an intelligent control system for museum LED lighting. This system uses a museum detection and division module to mark verification areas as either strongly or weakly regulated, and uses a regional lighting monitoring module and a coordination evaluation module to achieve intelligent control and coordination status detection and evaluation of the museum's LED lighting. However, existing technologies primarily target the specific display needs of museums, focusing on operating status monitoring of lighting fixtures and overall lighting coordination evaluation, and pay insufficient attention to the personalized lighting needs of different functional areas within the room. It also lacks the precise identification and status perception of display device areas and operating surface areas, making it difficult to apply to the fine-grained control of display device and operating surface lighting in everyday scenarios such as offices, homes, and schools.

[0005] When dealing with the lighting needs of different functional areas in a room, existing technologies generally have the problem of inaccurate identification of differences in regional lighting needs, especially when display devices and operating surfaces coexist. This makes it difficult to provide comfortable and uniform operating surface lighting while ensuring display effects. Summary of the Invention

[0006] This invention is applicable to a variety of indoor scenarios, including but not limited to office spaces where projectors are used for conference presentations, homes where laser TVs or projection equipment are used for movie viewing and entertainment, and learning spaces where electronic display devices are used for teaching or learning activities. In these scenarios, display devices and operating surfaces often coexist and require different lighting conditions.

[0007] To overcome the aforementioned shortcomings of the prior art, the present invention provides an LED eye-protection lighting control method and system. By acquiring a panoramic overhead view of the interior and utilizing a pre-built lighting control area detection model, the system accurately identifies the display device area (ROI1), the operating plane area (ROI2), and the environmental transition zone area (ROI3). Real-time status recognition is performed on the ROI1 and ROI2 areas, and a lighting control strategy is dynamically generated based on the recognition results. This method effectively resolves the conflicting lighting requirements between the display area and the operating plane area in traditional lighting solutions, avoiding screen glare and uneven desktop illumination caused by changes in ambient light. It significantly improves user visual comfort in a variety of scenarios, achieving an organic combination of eye protection, energy saving, and comfortable lighting, and providing intelligent and efficient lighting solutions for offices, homes, and learning environments.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] The LED eye protection lighting control method includes:

[0010] Obtain a panoramic bird's-eye view of the interior, and identify three types of lighting control areas using the panoramic bird's-eye view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environment transition zone area ROI3;

[0011] Perform status recognition on ROI1 and ROI2, and generate a lighting control strategy based on the status recognition results of ROI1 and ROI2.

[0012] Furthermore, the lighting control area detection model adopts the YOLOv5 framework.

[0013] Furthermore, the method for identifying the state of the ROI1 region includes: monitoring the brightness LA of the ROI1 region in real time, and when LA exceeds a first brightness threshold L tha , the ROI1 area is determined to be activated; otherwise, the ROI1 area is determined to be inactivated.

[0014] Furthermore, the method for identifying the state of the ROI2 region includes: detecting human activity signals in the ROI2 region, and if no human activity signals are detected for a continuous period of Δt1, determining that the ROI2 region is in an idle state, otherwise it is in a working state.

[0015] Furthermore, the method for generating a lighting control strategy based on the status recognition results of the ROI1 area and the ROI2 area includes:

[0016] When ROI1 is activated and ROI2 is in working state, the regional light distribution strategy is started;

[0017] When ROI1 is not activated but ROI2 is in working state, the ROI2 fill light strategy is started;

[0018] If ROI2 is in idle state, it enters energy saving mode.

[0019] Furthermore, the regional light distribution strategy is to perform anti-glare control on ROI1, perform fill light control on ROI2, and perform gradient control on ROI3.

[0020] Furthermore, the method for controlling fill light in the ROI2 area includes:

[0021] Construct a material reflection characteristic table for ROI2 area, obtain the material of the operation plane of ROI2 area, query the material reflection characteristic table to obtain the critical reflection angle and half-height width reflection angle of the corresponding material; adjust the illumination angle of the main light source in ROI2 area according to the critical reflection angle and half-height width reflection angle of the corresponding material.

[0022] Furthermore, the method for performing fill light control on the ROI2 region further includes: classifying the working state of the ROI2 region, and performing fill light control according to the classification result of the working state of the ROI2 region.

[0023] Furthermore, the method for classifying the working state of the ROI2 area includes:

[0024] An ROI2 image is obtained by cutting out the indoor panoramic bird's-eye view. Based on the ROI2 image and the pre-built operation plane object detection model, paper documents and electronic devices on the operation plane of the ROI2 are identified.

[0025] If an electronic device is identified, determining whether the electronic device is in working state;

[0026] The working state of the ROI2 area is classified according to the recognition results of the paper documents and electronic devices on the operation plane of the ROI2 area and whether the electronic devices are in the working state.

[0027] An LED eye-protection lighting control system, which is used to implement the above-mentioned LED eye-protection lighting control method, comprises:

[0028] Lighting control area identification module: used to obtain a panoramic indoor view and identify three types of lighting control areas based on the panoramic indoor view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environmental transition zone area ROI3;

[0029] Light distribution strategy generation module: used to identify the status of ROI1 and ROI2 areas, and generate lighting control strategies based on the status identification results of ROI1 and ROI2 areas.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention effectively solves the contradiction between the lighting requirements of the display area and the operating plane area in traditional lighting solutions by intelligently identifying different functional areas in the room and dynamically adjusting the lighting strategy according to their status. This method can not only accurately adjust the illumination according to the actual use status of the display device and the operating plane, avoiding the screen reflection problem caused by changes in ambient light, but also optimize the light source angle and power according to the material characteristics, ensuring that the displayed content is clearly visible while providing a uniform and comfortable lighting environment for the operating plane. In addition, the method also takes into account energy efficiency and automatically switches to energy-saving mode when no one is using it, reducing energy consumption. The entire solution realizes the organic combination of eye protection, energy saving and comfortable lighting, and provides intelligent and efficient lighting solutions for various scenarios such as office, home and study, which significantly improves the user's visual experience and work and study efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 This is a flow chart of the principle of the LED eye protection lighting control method of the present invention;

[0034] Figure 2 This is a flow chart of a method for constructing a material reflection characteristic table in the LED eye protection lighting control method of the present invention;

[0035] Figure 3 This is a flow chart of a method for determining activation of ROI1 in the LED eye protection lighting control method of the present invention;

[0036] Figure 4 This is a flow chart of a method for detecting human activity signals in ROI2 area in the LED eye protection lighting control method of the present invention;

[0037] Figure 5 This is a functional module diagram of the LED eye protection lighting control system in the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] Example 1

[0040] See also Figure 1 As shown, this embodiment provides an LED eye protection lighting control method, including:

[0041] Step S1000: Obtain a panoramic overhead view of the interior, and identify three types of lighting control areas using the panoramic overhead view and a pre-built lighting control area detection model; the three types of lighting control areas are the display device area ROI1, the operation plane area ROI2, and the environmental transition zone area ROI3; perform status recognition on ROI1 and ROI2, and generate a lighting control strategy based on the status recognition results of ROI1 and ROI2;

[0042] Furthermore, step S1000 includes:

[0043] Step S1100: Acquire a panoramic bird's-eye view of the room, and identify three types of lighting control areas using the panoramic bird's-eye view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environment transition zone area ROI3;

[0044] The method for obtaining the indoor panoramic bird's-eye view is as follows:

[0045] Fisheye cameras installed at the four corners of the ceiling are used to collect indoor scene images, which are then subjected to distortion correction and image stitching to obtain a panoramic bird's-eye view of the interior.

[0046] Specifically, fisheye cameras are installed at the four corners of the ceiling. Fisheye cameras have a wide-angle field of view, capable of covering the entire indoor space. Fisheye cameras were chosen because they offer a wide field of view, allowing a single device to capture images of the entire indoor environment. The fisheye camera's mounting position and angle are carefully designed to ensure its field of view covers all critical indoor areas, including display devices, operating surfaces, and environmental transition zones. The fisheye camera captures images of the indoor scene at a set frame rate (for example, 30 frames per second). The captured images include all areas of the room, such as display devices (projector screens, TV screens, etc.), operating surfaces (desks, desks, etc.), and environmental transition zones. During image acquisition, camera parameters (such as exposure time and gain) are dynamically adjusted based on indoor lighting conditions to ensure image quality. Due to the wide-angle nature of the fisheye camera, the captured images exhibit significant barrel distortion. Distortion correction is performed using software algorithms to process the images, resulting in images that more closely resemble the geometry of the real scene. The specific implementation method is to use the distortion correction function in image processing libraries such as OpenCV, input distortion parameters (such as radial distortion coefficients and tangential distortion coefficients), and perform pixel-by-pixel correction on the image. The distortion parameters are pre-acquired through the camera calibration process to ensure the accuracy of the correction. The images captured and corrected by the four fisheye cameras are stitched together to generate a panoramic bird's-eye view of the indoor environment. The stitching process involves image registration and fusion to ensure that the images captured by different cameras are seamlessly connected at the boundaries. The specific implementation method is to use image feature matching (such as SIFT or ORB algorithms) to find common feature points between images, and then align and fuse the images through projective transformation. The stitched panoramic bird's-eye view has high resolution and complete indoor coverage.

[0047] The lighting control area detection model uses the YOLOv5 framework, a deep learning-based object detection model. YOLOv5 strikes a good balance between real-time performance and accuracy, making it suitable for indoor lighting control scenarios. The YOLOv5 model architecture consists of a backbone network, an intermediate layer, and a detection head. The backbone network extracts image features, the intermediate layer performs feature fusion, and the detection head performs object detection. The input is a panoramic overhead view of the indoor environment, and the output is the coordinates and categories of the detected lighting control areas (ROI1, ROI2, and ROI3). The training data consists of panoramic overhead views of various indoor environments. Each image is annotated with the display device area (ROI1), the operating plane area (ROI2), and the environmental transition zone (ROI3). The annotations include bounding boxes and category labels for the areas. The training dataset covers a variety of indoor scenes, such as offices, homes, and schools, to ensure the model's generalization ability. During training, the annotated images are fed into the model, and loss functions (such as intersection over union loss and classification loss) are calculated between the model output and the ground-truth labels. Model parameters are updated through backpropagation and optimization algorithms (such as the Adam optimizer) until the model's detection accuracy reaches the target. During training, the model uses data augmentation techniques (such as random cropping, rotation, and color adjustment) to improve its adaptability to different lighting conditions and environmental changes.

[0048] Traditional methods for identifying indoor lighting control areas often rely on manual annotation or simple image processing algorithms. These methods are prone to false positives and missed detections in complex indoor environments. By using the YOLOv5 model, combined with a panoramic indoor bird's-eye view, it can automatically and accurately identify display device areas, operating planes, and environmental transition zones. The YOLOv5 model's real-time and high-precision capabilities make it an excellent choice for indoor lighting control scenarios.

[0049] Step S1100 uses a fisheye camera to capture a panoramic bird's-eye view of the indoor environment. Its wide-angle feature ensures the integrity of the panoramic bird's-eye view of the indoor environment. Distortion correction and image stitching technology ensure the geometric accuracy and visual effect of the image, and can generate a high-quality panoramic bird's-eye view of the indoor environment. The YOLOv5 model extracts image features through a deep learning method and can adapt to changes in different indoor environments. The use of the YOLOv5 model significantly improves the detection accuracy and efficiency of the lighting control area and reduces the workload and errors of manual labeling. The high-quality indoor panoramic bird's-eye view and accurate regional detection results provide reliable basic data for the subsequent construction of material reflection characteristic tables and state identification. Step S1100 provides accurate regional information for the subsequent generation of lighting distribution strategies by obtaining a panoramic bird's-eye view of the indoor environment and identifying lighting control areas. If this step is missing, the subsequent generation of lighting distribution strategies will lack accurate regional information, resulting in poor lighting distribution effects and an inability to meet the lighting needs of different areas.

[0050] Step S1200, constructing a material reflection characteristic table for ROI2;

[0051] Further, if Figure 2 As shown, step S1200 includes:

[0052] Step S1210 , under a standard light source, collecting the spectral reflectance R(θ) of a typical material in ROI2 at different incident angles θ, and drawing a θ-R(θ) curve;

[0053] Step S1220, analyzing the θ-R(θ) curve, extracting key characteristic parameters of spectral reflectance, and constructing a material reflection characteristic table; the key characteristic parameters of spectral reflectance include critical reflection angle and half-height width reflection angle;

[0054] Specifically, representative material samples in ROI2 are first placed under a standard light source to ensure consistent and stable lighting conditions. The standard light source is selected based on relevant lighting standards, such as CIE standard illuminant D65, to simulate common indoor lighting conditions. The material samples are measured using a high-precision spectrometer with sufficient resolution and wavelength range for accurate analysis of the visible spectrum. During measurement, the incident angle θ is varied, typically at 5° intervals between 0° and 60°, to acquire spectral reflectance data at multiple angles. Within the 0° to 60° range, the intensity of reflected light varies significantly, effectively reflecting the reflective properties of the material. Beyond 60°, the intensity of reflected light becomes increasingly flat, potentially leading to increased measurement error. Especially at angles approaching 90°, where the light is nearly parallel to the surface, the intensity of reflected light increases dramatically, leading to unstable and unreliable measurement results. According to Fresnel's law, the intensity of reflected light varies with increasing angle of incidence. In the range of 0° to 60°, the intensity of the reflected light changes more linearly, providing more valid data points and helping to accurately extract the key characteristic parameters of spectral reflectance. In common indoor lighting environments, the incident angle of light usually does not exceed 60°. For example, ceiling lamps are usually designed to shine downward, and the incident angle of light is generally between 0° and 60°. Incident angles exceeding 60° are relatively rare in practical applications and may cause excessive light dispersion or produce unnecessary shadows and reflections.

[0055] The collected θ-R(θ) curve is analyzed to extract key characteristic parameters of spectral reflectance, including the critical reflection angle and half-width reflection angle. The critical reflection angle refers to the starting angle at which the reflected light intensity begins to increase significantly when the incident angle increases to a certain angle. This is closely related to the surface microstructure and optical properties of the material. The half-width reflection angle refers to the range of incident angles corresponding to half the reflected light intensity, reflecting the material's ability to scatter light. These characteristic parameters are accurately extracted from the θ-R(θ) curve through curve fitting and mathematical analysis methods. When constructing a material reflectance characteristic table, the extracted key parameters such as the critical reflection angle and half-width reflection angle are stored in a table indexed by material type. This reflectance characteristic table provides important data support for subsequent fill light control, ensuring that the fill light strategy can be optimized and adjusted according to the material characteristics.

[0056] Prior art fill light control for the operating surface often fails to fully consider the reflective properties of the material, resulting in suboptimal fill light effects and problems such as dark areas or strong reflected light on the desktop. This invention, by collecting and analyzing detailed spectral reflectance data for typical materials in ROI2, constructs a material reflectance property table, providing precise material parameters for fill light control. This technical solution addresses the problems of uneven fill light and excessive or insufficient reflectance that arise from traditional methods that ignore material differences. By meticulously measuring and analyzing the spectral reflectance properties of typical materials in ROI2, the constructed material reflectance property table accurately reflects the reflective behavior of different materials at various incident angles. This provides critical data support for subsequent fill light control strategies, enabling precise adjustments based on the characteristics of different materials. For example, for highly reflective materials, the light source power can be appropriately reduced to avoid glare; for low-reflective materials, the light source power can be increased to improve illumination. This fill light control based on material reflectance properties effectively improves lighting uniformity and comfort in the operating surface, reduces visual fatigue, and enhances overall lighting quality. For example, in an office environment, work surfaces (such as desks) are made of a variety of materials, including paper documents and laptops. The material reflection characteristic table constructed in step S1200 accurately captures the reflection characteristic parameters of these materials. During fill light control, light source parameters are adjusted based on the material reflection characteristic table to ensure uniform illumination in the paper document area while preventing strong reflections from electronic device screens, thereby improving work efficiency and visual comfort.

[0057] Step S1200 closely coordinates with the subsequent step S1412. In step S1412, when performing classified fill light control based on the operating state of ROI2, the light source's illumination angle and power must be adjusted based on the parameters in the material reflection characteristics table. Without the material reflection characteristics table constructed in step S1200, the fill light control in step S1412 will lack critical data support, failing to optimize fill light for different materials. This will result in poor fill light performance and an inability to meet the lighting requirements of the operating surface area. Without step S1200, subsequent fill light control (such as step S1412) will not be able to obtain accurate material reflection characteristic parameters, resulting in an inability to optimize the fill light control strategy based on the material characteristics. For example, after identifying paper documents and electronic devices on the operating surface in ROI2, the material reflection characteristics table must be consulted to obtain the critical reflection angle of the corresponding materials to adjust the light source's illumination angle. Without this table, accurate fill light control cannot be achieved, leading to problems such as insufficient illumination in the paper document area or excessive reflection on the electronic device screen. Step S1200 provides necessary data support for subsequent fill light control by constructing a material reflection characteristic table; ensures that the fill light strategy can be optimized and adjusted according to the actual material characteristics of the operating plane area, thereby improving the intelligence level of the lighting system and user experience.

[0058] Step S1300 , performing status identification on the ROI1 region and the ROI2 region;

[0059] Furthermore, step S1300 includes:

[0060] Step S1310, as Figure 3 As shown, the brightness LA of ROI1 is monitored in real time. When LA exceeds the first brightness threshold L tha When , it is determined that ROI1 is activated; otherwise, it is determined that ROI1 is not activated;

[0061] Step S1320 , detecting a human activity signal in the ROI2 area. If a human activity signal is detected, the ROI2 area is determined to be in an active state, otherwise it is determined to be in an idle state.

[0062] The method for detecting human activity signals in the ROI2 area is as follows: a millimeter-wave radar is deployed in the ROI2 area. If the radar echo amplitude is lower than a preset echo amplitude threshold and the duration is greater than t3, it is determined that a human activity signal exists; otherwise, it is determined that no human activity signal exists.

[0063] Specifically, in step S1310, LA does not only refer to the luminous brightness of the display device (such as a projection screen or a TV screen), but the overall received brightness of the screen surface, including ambient light reflection and device self-luminescence. The purpose of monitoring this parameter is to accurately judge the lighting conditions in the display device area, because ambient light has a great influence on the display effect of the display device. In actual application scenarios, for example, in a conference room with large windows and no obstructions, when the sun is directly shining, even if the luminous brightness of the projection screen itself remains unchanged, the overall received brightness LA of the screen surface will increase significantly due to the enhanced reflection of ambient light, which may cause the display effect of the picture to deteriorate and a "whitening" phenomenon to occur. In order to realize real-time monitoring of LA, a photosensor is arranged in the ROI1 area. The photosensor can convert the optical signal into an electrical signal. By processing and analyzing the electrical signal, the brightness LA of the ROI1 area can be obtained in real time. When LA exceeds the first brightness threshold L tha , the ROI1 region is determined to be activated; otherwise, the ROI1 region is determined to be inactivated. The first brightness threshold L tha It is determined based on factors such as the brightness range of the display device that is comfortable for the human eye and the ambient brightness required for normal display. In existing technologies, some lighting control methods fail to accurately distinguish between the display device's own brightness and the overall brightness under the influence of ambient light. This results in an inability to precisely control lighting when ambient light changes, affecting the user's viewing experience. However, this step, by precisely monitoring LA and comparing it with the threshold, can promptly detect abnormal lighting conditions in the display device area, providing a basis for subsequent anti-glare control and other operations. This effectively prevents degradation of display device image quality due to excessive ambient light and improves the display device's visibility under varying ambient light conditions.

[0064] In step S1320, a millimeter wave radar is deployed in the ROI2 area to detect human activity signals. The millimeter wave radar uses electromagnetic waves in the millimeter wave frequency band to detect targets. When the human body moves in the ROI2 area, it will reflect the electromagnetic waves emitted by the millimeter wave radar, causing the echo amplitude received by the radar to change. Figure 4As shown, if the radar echo amplitude is lower than the preset echo amplitude threshold and the duration is greater than t3, a human activity signal is determined to be present. Otherwise, a human activity signal is determined to be absent and the radar echo amplitude continues to be acquired, continuously monitoring the human activity signal in ROI2. The preset echo amplitude threshold and time t3 are determined based on the echo signal characteristics generated by human activity in real-world scenarios. For example, in an office, if a human activity signal is detected while an employee is working at a desk, it can be determined that ROI2, where the desk is located, is in active use. Existing technologies often lack accuracy in determining whether an operating plane is occupied or rely on complex equipment. This step uses millimeter-wave radar to detect human activity signals, enabling precise determination of the operating plane's usage status, providing an accurate basis for subsequent lighting strategy adjustments. If ROI2 is idle, energy-saving mode can be promptly activated to reduce lighting energy consumption. If it is active, targeted fill lighting can be provided based on the specific situation, meeting the user's lighting needs in different operating scenarios and improving lighting intelligence and energy efficiency.

[0065] Step S1300 accurately determines the status of ROI1 and ROI2 by monitoring the brightness LA of ROI1 and detecting human activity signals in ROI2. This solves the problem of the prior art in accurately determining the actual usage status of the display device area and the operating surface area. It avoids maintaining the original lighting status when it is not needed or when the lighting is too strong, resulting in energy waste or a negative impact on the visual experience. For example, during the day when a conference room is unoccupied and sunny, without this step, the lighting in the display device area and the operating surface area may remain on without adjusting the brightness, resulting in energy waste. With this step, the zone status can be determined in a timely manner, and the lighting strategy can be adjusted to reduce energy consumption. By identifying the status of different zones, the lighting system can better adapt to various application scenarios. Whether in a conference room, office, or home study, lighting control can be adjusted based on the actual zone status. This solves the problem of the poor versatility of existing lighting systems, enabling the present lighting control method to be widely applied in a variety of locations to meet the needs of different users. For example, in a home study, when a user is using a computer (ROI1 is activated and ROI2 is in operation) or reading a book (ROI1 is inactive and ROI2 is in operation), the lighting system can adjust the light distribution based on the different zone states to provide an appropriate lighting environment. Without step S1300, the entire lighting control solution would be unable to accurately determine the status of ROI1 and ROI2, resulting in a lack of basis for regional light distribution strategies and an inability to achieve precise lighting control. This could result in the display device area not being treated for anti-glare even in strong sunlight, or the operating surface area remaining illuminated at high brightness when unoccupied, impacting visual quality and wasting energy. Furthermore, subsequent functions such as energy-saving mode, strong light interference warning, and dynamic compensation mechanisms could not be effectively triggered and executed, reducing the intelligence and functionality of the entire lighting system and failing to meet user needs for eye protection, energy saving, and comfortable lighting.

[0066] Step S1400 , generating a lighting control strategy based on the status recognition results of the ROI1 and ROI2 regions;

[0067] Furthermore, step S1400 includes:

[0068] Step S1410: When ROI1 is activated and ROI2 is in working state, a regional light distribution strategy is started; the regional light distribution strategy is to perform anti-glare control on ROI1, fill light control on ROI2, and gradient control on ROI3;

[0069] Furthermore, step S1410 includes:

[0070] Step S1411, perform anti-glare control on the ROI1 area: according to the size and installation height of the display device in the ROI1 area, set the initial light output angle to the preset anti-glare angle to ensure that the main optical axis of the light source avoids the mirror reflection direction of the display device surface; when LA exceeds the preset second brightness threshold L tha2 When the output power of the ambient light source around ROI1 is automatically reduced, where L tha2 >L tha A surround low-glare light source is added to the edge of the ROI1 area, with the light output angle facing the ceiling or wall. The ambient light is supplemented by diffuse reflection to avoid strong contrast between the edge of the display device and the dark environment, which may cause visual fatigue.

[0071] Specifically, step S1411 primarily performs anti-glare control on ROI1 (the display device area), aiming to address the issue of glare on the display device screen caused by ambient light reflection, which affects the user's visual experience. The "preset anti-glare angle" is determined based on the vertical viewing angle range of the display device. Different types of display devices, such as projector screens, TV screens, and electronic whiteboards, have different vertical viewing angles. Taking a common projector screen as an example, assuming its vertical viewing angle range is measured to be 120°-160°, an appropriate preset anti-glare angle is determined by analyzing this range and taking into account human viewing habits and light reflection principles. This setting ensures that the main optical axis of the light source avoids specular reflection from the display device surface, preventing light from directly striking the center of the screen, thereby reducing the potential for glare at the source. For example, in a conference room, if the light source shines directly on the center of the projection screen, strong reflections will be generated on the screen, making it difficult for participants to see the content on the screen. Setting a preset anti-glare angle and adjusting the light source position and light output direction can effectively improve this situation.

[0072] The second brightness threshold L tha2 The setting is based on a large amount of experimental data and research on human visual comfort. tha2 When the ambient light reflection is too strong, it has a negative impact on the screen display effect and the human eye viewing experience. It is necessary to automatically control and reduce the output power of the ambient light source around the ROI1 area. For example, when the TV screen is used as a display device in the office, when the direct sunlight outside the window causes the overall received brightness LA of the screen surface to exceed L tha2 When the screen is turned off, the lighting system will automatically reduce the output power of the LED lights on the ceiling to reduce the reflection of ambient light on the screen, thereby reducing the intensity of the reflected light on the screen, so that LA matches the ambient brightness, alleviating problems such as frequent pupil contraction and decreased contrast between screen content and background caused by excessive screen reflection, reducing visual fatigue and improving visual comfort. tha2 The setting is higher than the first brightness threshold L thato ensure that stricter anti-glare control is triggered when the ambient light is too strong.

[0073] A surround low-glare light source is added to the edge of the ROI1 area, with the light output angled toward the ceiling or wall, utilizing diffuse reflection to supplement ambient light. This is because the strong contrast between the edge of the display device and the dark surroundings can cause visual fatigue and affect viewing quality. Light from the surround low-glare light source is evenly distributed throughout the surrounding environment after being directed toward the ceiling or wall. This prevents additional glare from directly hitting the screen while also appropriately increasing the ambient brightness at the edge of the display device, reducing the brightness difference between the edge and the dark environment. For example, in scenarios where electronic whiteboards are used at home, the edges of the electronic whiteboard appear particularly abrupt in a dark environment, which can easily cause visual fatigue after prolonged viewing. By adding a surround low-glare light source, the ambient brightness at the edge of the electronic whiteboard is increased, significantly improving the visual effect.

[0074] For example, when a projector is in use in a conference room, step S1411 sets the initial light output angle based on the size and installation height of the display device to prevent direct light from hitting the projection screen. Simultaneously, the system monitors ambient light brightness in real time. When the ambient light is too strong, it automatically reduces the power of the surrounding light source to ensure clear and visible projection content. The added surround low-glare light source supplements ambient light through diffuse reflection, making the brightness of the projection screen edge more coordinated with the surrounding environment, reducing visual fatigue for participants.

[0075] In the prior art, anti-glare control often only focuses on the luminous parameters of the display device itself, while ignoring the influence of ambient light. This results in excessively strong light reflected from the surface of the display device when the ambient light is too strong, causing glare problems. The present invention effectively solves this problem by comprehensively considering the size, installation height and ambient light conditions of the display device, setting the initial light output angle and dynamically adjusting the power of the ambient light source. At the same time, the additional surround low-glare light source supplements the ambient light through diffuse reflection, further reducing the brightness contrast between the edge of the display device and the surrounding environment. By accurately setting the initial light output angle and dynamically adjusting the power of the ambient light source, step S1411 effectively prevents light from being directly reflected into the human eye, reducing glare. This not only improves the visibility of the display device, but also reduces visual fatigue caused by strong light reflection. For example, when using a projection screen in a conference room with direct sunlight, reducing the power of the surrounding ambient light source can prevent the screen from "whitening" due to excessive ambient light, thereby improving the clarity of the projected content. The added surround low-glare light source supplements the ambient light through diffuse reflection, making the brightness of the edge of the display device more coordinated with the surrounding environment, further reducing visual fatigue and improving the overall lighting comfort.

[0076] Step S1412, performing fill light control on the ROI2 area;

[0077] Furthermore, step S1412 includes:

[0078] Step S14121: Obtain the material of the operating plane in ROI2, query the material reflection characteristic table to obtain the critical reflection angle and half-width reflection angle of the corresponding material; adjust the illumination angle of the main light source in ROI2 according to the critical reflection angle and half-width reflection angle of the corresponding material;

[0079] Specifically, step S14121 aims to improve the lighting quality in ROI2 (the operating plane) by adjusting the main light source's illumination angle based on the material's reflective properties, addressing the issue of glare from the operating plane affecting the visual experience. The "critical reflection angle" refers to the angle at which the intensity of light reflected from the material surface increases significantly, potentially disrupting vision. The "full width at half maximum" (FWHM) reflects the intensity distribution of reflected light. Together, these two factors reflect the material's reflective properties.

[0080] The material type of the operating surface in ROI2 can be determined by pre-marking the operating surface material or by using image recognition technology combined with a material feature database. For example, in an office environment, the material information of different desks can be entered into the system in advance. When the material type is needed, the system can be queried to obtain it. If image recognition technology is used, a camera captures an image of the operating surface, analyzes the image's texture, color, and other features, and compares them with samples in the material feature database to determine the material type.

[0081] After determining the material type, query the material reflectance properties table to obtain the corresponding material's critical reflection angle and half-width reflection angle. For a wooden tabletop, for example, assume its critical reflection angle is 40° and its half-width reflection angle is 30°. Adjust the main light source's illumination angle in ROI2 based on these parameters. This principle is based on the law of light reflection. When the incident angle of light equals the critical reflection angle, the reflected light is strongest, resulting in noticeable glare, which affects visual quality. Therefore, adjusting the main light source's illumination angle away from the critical reflection angle can effectively reduce glare. Specifically, adjust the main light source's illumination angle to a value within a certain range of the critical reflection angle based on actual conditions. For example, for the wooden tabletop described above, the main light source's illumination angle can be adjusted to a value away from 40°, such as 25° or 55°. The half-width reflection angle also provides a reference for adjustment. A smaller half-width reflection angle indicates that reflected light is concentrated within a smaller angle range, requiring more precise adjustment to avoid the area near the critical reflection angle. A larger half-width reflection angle allows for more flexibility in the adjustment range.

[0082] Existing technologies often fail to carefully consider the material's reflective properties when adjusting the light source's illumination angle, resulting in strong reflections at certain angles and affecting visual comfort. This invention, by querying a material reflectivity table, accurately obtains the critical reflection angle and half-width reflection angle for different materials and adjusts the light source's illumination angle accordingly, effectively avoiding strong reflections.

[0083] Step S14122: classify the working state of the ROI2 area, and perform fill light control according to the classification result of the working state of the ROI2 area;

[0084] Furthermore, step S14122 includes:

[0085] Step S141221: obtaining an ROI2 image from the indoor panoramic bird's-eye view, and identifying paper documents and electronic devices on the ROI2 operation plane based on the ROI2 image and a pre-built operation plane object detection model;

[0086] Specifically, the indoor panoramic bird's-eye view captures indoor scene images using fisheye cameras mounted at the four corners of the ceiling. After distortion correction and image stitching, it captures the visual information of the entire indoor space. The ROI2 region image is cropped based on the predetermined coordinate range of the ROI2 region in the indoor panoramic bird's-eye view. An image cropping algorithm can be used to separate the image portion containing the work plane from the panoramic view based on the ROI2 region's boundary coordinates. For example, in an office scenario, the location of the ROI2 region, where the desk is located, can be determined through preliminary analysis of the office area's layout. This input into the cropping algorithm yields an ROI2 region image containing only the desk. This avoids detecting electronic devices in irrelevant areas of the room, improving detection accuracy and ensuring more targeted recognition of paper documents and electronic devices on the work plane. The cropped ROI2 region image is then analyzed using a work plane object detection model based on the YOLOv5 framework. This model is trained on a custom dataset containing image samples of common office paper documents (such as A4 paper and notebooks) and electronic devices (such as laptops, tablets, and mobile phones). During training, the model learns the characteristics of these samples, including shape, color, and texture, enabling it to identify different objects. During recognition, the model processes and analyzes the pixel information in the ROI2 image, comparing it with the characteristics of the training samples to determine the presence of paper documents and electronic devices in the image and determine their location and category.

[0087] Step S141222: If an electronic device is identified, determine whether the electronic device is in working state;

[0088] The method for determining whether an electronic device is in an operating state comprises: identifying a display screen area of ​​the electronic device and extracting image features of the display screen area; inputting the extracted image features into a pre-trained electronic device operating state classification model to determine whether the electronic device is in an operating state;

[0089] Specifically, when identifying the display area of ​​an electronic device, a preliminary location is first performed based on the device's overall outline and morphological features. For example, for a common rectangular electronic device, the display area can be roughly determined by detecting edge features and geometric shapes in the image. Image segmentation techniques are then used to further precisely delineate the display area. Image segmentation can distinguish the display from other parts of the electronic device based on differences in features such as color and brightness. When extracting image features from the display area, the focus is on brightness, color distribution, and texture of the image content. Brightness features reflect the display's luminous intensity, color distribution reflects its color pattern, and texture features contain detailed information about the screen's content. These features are extracted using specific image feature extraction algorithms, such as the gray-level co-occurrence matrix for texture features and the color histogram for color distribution information. The extracted image features are then fed into a pre-trained electronic device operating state classification model. This model is built using a convolutional neural network (such as MobileNet or EfficientNet) and trained on a dataset containing image samples of various electronic devices with their displays in both on and off states. During the training process, the model learns the differences in display image features when on and off, so that when it receives new display image features, it can accurately determine whether the electronic device is in working state.

[0090] Step S141223: classify the working state of the ROI 2 area according to the recognition results of the paper documents and the electronic device on the operating plane of the ROI 2 area and whether the electronic device is in the working state;

[0091] Step S141224: perform fill light control according to the classification result of the working state of ROI2 area.

[0092] Specifically, if only paper documents and no electronic devices are present in ROI2, this is considered the paper document working state. The primary requirement is to provide a lighting environment suitable for reading and writing paper documents. Increase the main light source's output angle and power in ROI2 to ensure more even light coverage on the paper surface, improving illumination. By properly adjusting the light source's position and angle, shadows are avoided and the reflection uniformity on the paper surface is improved. Adjust the color temperature to a neutral setting to reduce eye irritation from blue light, making reading and writing more comfortable for users and enhancing the reading and writing experience. If only electronic devices are in operation, this is considered the electronic device working state. The focus is on reducing screen glare and ensuring appropriate color reproduction. Lower the main light source's output angle to prevent direct light from hitting the electronic screen, reducing screen glare and improving on-screen visibility. Appropriately reduce the output power to prevent excessive ambient light from exacerbating screen glare. Increase the light source's color temperature to improve on-screen color reproduction, allowing users to see more realistic image colors. Auxiliary light sources are placed around electronic devices to provide localized fill light, reducing the brightness contrast between the electronic screen and the surrounding environment and reducing eye fatigue. For example, when using a computer in the office, this fill light setting allows employees to see the screen content more clearly and improve work efficiency.

[0093] When there are only electronic devices in a non-working state, it is an idle state of electronic devices. The light output angle and light output power of the main light source are further reduced to provide only basic lighting. The auxiliary light sources around the electronic devices are turned off to avoid energy waste and meet energy-saving requirements. When there are paper documents and electronic devices in a working state at the same time, it is a mixed working state. It is necessary to take into account the lighting requirements of paper document reading and electronic device use. For the paper document area, refer to the paper document working state for fill light, and for the electronic device area, refer to the electronic device working state for fill light. The lighting requirements of the two are comprehensively balanced. At the same time, the reflection of paper documents and the reflection of electronic screens are controlled to meet the lighting requirements of different tasks. When there are paper documents and electronic devices in a non-working state at the same time, it is a mixed idle state. It is necessary to balance the illumination required for paper documents and the energy-saving requirements of electronic devices when they are idle. For the paper document area, ensure the paper surface illumination, and reduce the light output power for the electronic device area to avoid energy waste. When there are neither paper documents nor electronic devices present, the room is in an idle state. The main light source's angle and power are significantly reduced to a comfortable, basic lighting level, meeting the minimum illumination requirement in a non-working state. The color temperature is adjusted to warm light to create a relaxing atmosphere, providing both basic lighting and a sense of comfort. Compared to full-power lighting, this significantly reduces energy consumption.

[0094] Existing technologies often only simplistically distinguish between occupied and unoccupied areas when classifying the working state of the operating plane, failing to accurately identify different object types and states. This results in a single fill-light strategy that fails to meet diverse needs. For example, in school classrooms, students use both paper textbooks and electronic devices for learning, and traditional lighting cannot simultaneously meet the optimal lighting requirements of both. This step effectively addresses this problem by accurately classifying the working state of ROI2 and implementing targeted fill-light control. For paper document work, this solution provides lighting suitable for reading and writing; for electronic device work, it reduces screen glare and optimizes color reproduction. This ensures a comfortable visual experience for users in various operating scenarios, improves learning and work efficiency, and meets the diverse lighting needs of the operating plane. In step S14122, the parameters of the lighting source are appropriately adjusted for different working states to avoid energy waste. When the electronic device is idle or in standby mode, the lighting intensity is reduced and unnecessary auxiliary light sources are turned off, reducing power consumption. Compared to traditional lighting systems that maintain high brightness during non-operating hours or when the device is idle, this solution achieves a balance between energy conservation and lighting performance, reducing long-term operating costs and improving energy efficiency. Based on the recognition and classification of the working status of objects and electronic devices on the operating surface, the lighting system can automatically adjust the lighting strategy, with greater intelligence and adaptability. Whether in an office, school, home study, or other scenarios, the lighting can be adjusted in real time according to actual conditions to meet the ever-changing needs of users.

[0095] Step S14122 coordinates with step S1320 in detecting human activity signals in ROI2. When the area is detected as being in the active state, the active state classification and fill light control in this step are initiated; when the area is in the idle state, the system enters energy-saving mode, thus forming an integrated whole for the entire lighting control scheme. Compared to operating each step independently, this synergistic effect allows for more precise adjustment of lighting based on the environment and user needs, enabling dynamic intelligent control of the lighting system. This significantly improves the performance and practicality of the lighting system, ensuring the effective implementation of the entire lighting control scheme and achieving the goals of eye protection, energy saving, and comfortable lighting. Without step S14122, the entire lighting control scheme would be unable to effectively differentiate and provide targeted fill light for different work scenarios in ROI2. This can lead to severe visual impairment caused by screen glare when using electronic devices, insufficient or uneven illumination when reading paper documents, and reduced learning and work efficiency. Furthermore, the lighting energy consumption cannot be adjusted according to the active state, resulting in energy waste.

[0096] Step S1413, perform gradient control on ROI3: the edge illumination of ROI1 is the starting gradient illumination E start , the edge illumination of ROI2 area is the end gradient illumination E end , in ROI3 area (transition zone between ROI1 area and ROI2 area), a transition zone from Estart to E end The linear or nonlinear illuminance gradient can avoid the sudden brightness change in adjacent areas causing the burden of light adaptation of the eyes. According to the illuminance value E of ROI3 area monitored in real time by the ambient light sensor C , set the upper limit of gradient illumination E C,max and the lower limit E C,min By adjusting the power or angle of the light source in ROI3, E C Always smoothly transition within the gradient range.

[0097] Specifically, the ROI3 area (environmental transition zone) is gradually controlled to avoid sudden brightness changes in the adjacent areas of ROI1 area (display device area) and ROI2 area (operation plane area), reduce the burden of light adaptation on the human eye, and optimize the transition effect of the overall indoor lighting environment. start " is the illumination value at the edge of ROI1 area, " ends the gradient illumination E end " is the illuminance value at the edge of ROI2. The determination of these two values ​​is based on the illuminance requirements of ROI1 and ROI2 when they are working normally, and the range of human eye's adaptation to brightness changes. For example, the appropriate illuminance of ROI1 when the display device is in normal use is Xlux. Considering the transition requirements between the edge of the display device and the surrounding environment, the illuminance of the edge of ROI1 is determined to be Elux. start =XY lux (Y is the illuminance difference determined according to the actual situation); the appropriate illuminance of ROI2 area in different working states is different. Assuming that the appropriate illuminance in the paper document working state is Z lux, also considering the transition with ROI3 area, determine the edge illuminance E of ROI2 area end is ZW lux (W is the corresponding illuminance difference). "Linear or nonlinear illuminance gradient" means that in ROI3 area, the illuminance gradient changes from E start to E end The linear gradient means that the illumination changes from E to start Change to E end Non-linear gradients can be based on actual needs, such as simulating the changing laws of natural light, and adopt exponential, logarithmic and other change methods to achieve an illumination transition effect that is more in line with the visual habits of the human eye.

[0098] The implementation process is as follows: First, an ambient light sensor is placed in the ROI3 area to monitor the illumination value E in the ROI3 area in real time. C The ambient light sensor converts the received light intensity into an electrical signal or digital signal and transmits it to the lighting control system. C The upper limit of the gradient illumination E is set based on the environmental factors (such as the changes in ambient light during the day and night). C,max and the lower limit E C,minFor example, during the day, when the ambient light is strong, in order to make the transition between indoor and outdoor lighting more natural, C,max Set it higher; at night, when the ambient light is weak, C,min The lighting control system then adjusts the C Value and E C,max 、E C,min By adjusting the power or angle of the light source in ROI3, E C Always smoothly transition within the gradient range. C Close to E C,max When E C Close to E C,min When the light source is too bright, increase the power of the light source or adjust the angle to increase the light intensity.

[0099] Many existing lighting systems fail to properly manage lighting transitions between different functional areas, resulting in significant brightness differences between adjacent areas. This leads to frequent adjustments to brightness when the eye switches between these areas, which can easily cause visual fatigue. For example, in some conference rooms, the brightness difference between the projection screen and the surrounding work area is significant. Participants' eyes need to constantly adjust to the brightness changes while viewing the projection and taking notes, impacting the meeting experience and work efficiency. In a home environment, sudden changes in brightness between the living room TV area and the sofa reading area can also cause user discomfort. This step effectively addresses these issues by implementing a gradual illumination control within ROI3. By implementing a gradual illumination control within ROI3, sudden brightness changes between adjacent areas are avoided. When switching between different areas, the eye does not need to quickly adapt to large brightness changes, effectively reducing the burden of light adaptation. This gradual illumination control creates a softer, more natural indoor lighting, enhancing the overall lighting comfort of the indoor environment. Whether working, studying, or living, users experience a more comfortable lighting atmosphere and increase their satisfaction with the environment. The gradient control of ROI3 organically combines the lighting effects of ROI1 and ROI2, making the entire lighting system more complete and coordinated. The lighting transition between different areas is natural, avoiding abruptness and improving the quality of the lighting system.

[0100] Step S1420: When the ROI1 area is not activated but the ROI2 area is in the working state, the ROI2 fill light strategy is started, wherein the ROI2 fill light strategy controls the fill light of the ROI2 area and maintains uniform basic illumination of the ROI3 area;

[0101] Specifically, step S1420 mainly starts the ROI2 fill light strategy for the scenario where ROI1 is not activated but ROI2 is in working state, aiming to provide appropriate lighting for the operation plane area while ensuring stable basic lighting in the environmental transition zone area.

[0102] “ROI1 area is not activated” means that the brightness LA of the display device area does not exceed the first brightness threshold L tha , indicating that the display device may not be in use or that ambient light has little impact on it. "ROI2 is in operation" indicates that the operating plane is currently in use. In this case, the ROI2 fill light strategy is activated. The ROI2 fill light strategy applies fill light control to ROI2, while maintaining uniform base lighting in ROI3. Fill light control for ROI2 is consistent with the "fill light control for ROI2" method in the regional light distribution strategy described in step S1412. Uniform base lighting is maintained in ROI3. This is because in this scenario, ROI1 is inactive, and the lighting requirements for the ambient transition zone are relatively low. Only a stable base lighting level is required to ensure a natural transition between ROI2 and the surrounding environment, avoiding sudden brightness changes that could cause eye discomfort. For example, in a classroom, when the projector is not in use (ROI1 is inactive) and students are studying at their desks (ROI2 is in operation), the lighting in ROI3 is maintained at an appropriate base brightness, preventing students from experiencing eye fatigue due to changes in ambient lighting while focusing on their desk work. Existing technologies often lack specificity when dealing with lighting scenarios with combinations of activation states for different areas. For example, in some places, regardless of whether the display device is in use, the lighting system uses a unified lighting mode. As a result, when ROI1 is inactive and ROI2 is working, ROI2 may experience insufficient or excessive lighting, as well as uncoordinated lighting in the environmental transition zone, affecting the user's visual experience and work and learning efficiency. This step effectively solves these problems by activating the ROI2 fill light strategy and stabilizing the lighting in ROI3. Without step S1420, in the scenario where ROI1 is inactive but ROI2 is in working state, the lighting system cannot provide appropriate lighting for ROI2 in a targeted manner, which may result in insufficient lighting in the operating plane area, affecting the user's work and learning efficiency. At the same time, the lighting in ROI3 cannot be reasonably adjusted according to this scenario, and the lighting may be too strong or too weak, disrupting the coordination of the entire indoor lighting.

[0103] Step S1430: If the ROI2 area is in an idle state, enter the energy-saving mode.

[0104] Specifically, when ROI2 (the operating surface) is idle, the lighting parameters of each area are adjusted to enter energy-saving mode. The specific implementation process is as follows: First, the light output angle of the lighting source in ROI1 (the display area) is reduced to provide only basic lighting. This is because when ROI2 is unoccupied, the lighting demand for the display area is also reduced. For example, after office hours, ROI1, where the conference room projection screen is located, generally does not require high-brightness lighting. Lowering the light output angle can reduce light energy waste while maintaining a certain basic brightness to meet basic safety and visual requirements. Second, the fill light in ROI2 is turned off and the illumination in ROI2 is reduced. When ROI2 is idle, it means no one is reading, writing, or operating electronic devices in that area, making fill light unnecessary. Disabling fill light and reducing illumination directly reduces energy consumption. For example, in a school classroom, if the lighting on the desks remains at the same level during recess, energy waste can be avoided. Disabling fill light and reducing illumination can prevent this. Finally, the gradient range of ROI3 (the environmental transition zone) was reduced, and the upper limit of the gradient illumination in ROI3 was lowered. Since the lighting intensity in ROI1 and ROI2 has been reduced, ROI3, as a transition zone, does not need to maintain a large gradient range and a high upper limit of illumination. This can further reduce energy consumption while ensuring that the lighting transition between areas remains relatively natural.

[0105] Existing technologies often fail to adjust lighting in a timely manner when areas are idle, resulting in significant energy waste. For example, traditional office lighting systems often keep lighting fully on even when certain areas are unoccupied after hours, resulting in significant energy waste. This step effectively addresses this issue by accurately detecting the idle state of ROI2 and promptly adjusting the lighting parameters for each area. By disabling fill lighting in ROI2 and reducing the lighting intensity in ROI1 and ROI3, energy consumption is directly reduced. Reducing the light output angle and power of the lighting source reduces the workload of the lighting equipment, helping to extend its service life.

[0106] Step S1500: determining whether there is strong light interference. If there is strong light interference, triggering a strong light interference warning and dynamic compensation mechanism;

[0107] Furthermore, step S1500 includes:

[0108] Step S1510: If the illumination value E C Greater than the preset illumination threshold E th , and the duration exceeds t2 seconds, it is determined that there is strong light interference;

[0109] Step S1520: If strong light interference exists, a strong light interference warning and dynamic compensation mechanism is triggered.

[0110] The dynamic compensation mechanism is: increase the light output angle of the lighting source in ROI1 area, and at the same time increase the upper limit of the gradual illumination in ROI3 area; continuously monitor the illumination value E in ROI3 area. C , if the illumination value E C Less than or equal to the illumination threshold E th , and the duration exceeds t2 seconds, the original light output angle of ROI1 area and the original gradient illumination upper limit of ROI3 area are restored.

[0111] Specifically, the illuminance value E C It is obtained through real-time monitoring of the ambient light sensor deployed in ROI3 area (environmental transition zone). The preset illumination threshold E th It is determined based on the human eye's adaptation range to different light intensities and the lighting requirements of the display device during normal use. For example, in a normal office environment, after a large number of experiments and actual tests, it is determined that when the illuminance value in the ROI3 area exceeds a certain value for a long time, it will have a negative impact on the viewing effect of the display device and the comfort of the human eye. This specific value is E th The duration of t2 seconds is to avoid misjudging strong light interference due to short-term light fluctuations, ensuring the accuracy of judgment. For example, in a meeting room, when sunlight suddenly shines through the window, the ambient light sensor detects the illuminance value E in ROI3 area. C Rapidly rise and exceed E th , and the duration exceeds t2 seconds, it is determined that there is strong light interference.

[0112] If there is strong light interference, the strong light interference warning and dynamic compensation mechanism will be triggered; triggering the warning can be achieved in a variety of ways, such as sounding an alarm on the control terminal, displaying a prompt message on the display screen, etc., to remind the user that there is currently strong light interference, which may affect the visual effect. The specific implementation of the dynamic compensation mechanism is: increasing the light output angle of the lighting source in the ROI1 area (display device area) and expanding the anti-glare protection range of the ROI1 area. This is because strong light interference may increase the reflected light on the surface of the display device. Increasing the light output angle can make the light of the light source more evenly distributed and reduce the impact of reflected light on the human eye. At the same time, increase the upper limit of the gradual illumination of the ROI3 area to accelerate the transition of light adaptation and reduce the burden on eye adjustment. Under strong light interference, the human eye suddenly enters a strong light environment from a normal lighting environment and needs to adapt to light. Increasing the upper limit of the gradual illumination can make the eyes adapt to light changes faster and reduce discomfort. Continuously monitor the illumination value E of the ROI3 area C , if the illumination value E C Less than or equal to the illumination threshold E th , and the duration exceeds t2 seconds, the original light output angle of ROI1 and the upper limit of the original gradient illumination of ROI3 are restored to avoid energy waste caused by overcompensation. For example, in the above conference room, when the sunlight is blocked by the curtains, the illumination value of ROI3 is EC Lowered to E th The lighting parameters will automatically return to normal settings after the following t2 seconds.

[0113] Existing technologies often fail to respond effectively and in a timely manner when faced with strong light interference, resulting in a reduced visual experience for users and eye fatigue. For example, in some classrooms, when sunlight is directly overhead, students will experience reflections when viewing electronic whiteboards or projection screens, affecting their learning outcomes, and traditional lighting systems are unable to automatically adjust lighting to address this issue. This step effectively addresses the impact of strong light interference on vision through real-time monitoring and dynamic compensation mechanisms. Without step S1500, the lighting system would be unable to respond in a timely manner when encountering strong light interference, resulting in severe reflections on the surface of the display device, making it difficult for users to see the screen content clearly, and affecting work and learning efficiency. At the same time, the human eye is in an uncomfortable state for a long time under strong light interference, which can easily cause visual fatigue and damage vision. In the entire solution, the early lighting optimization work will be ineffective due to strong light interference, and the lighting control in each area will not be able to effectively respond to sudden strong light conditions, and will not achieve the goals of improving visual comfort and protecting human eye health.

[0114] Example 2

[0115] This embodiment provides an LED eye protection lighting control system based on embodiment 1, such as Figure 5 As shown, including:

[0116] Lighting control area identification module: used to obtain a panoramic indoor view and identify three types of lighting control areas based on the panoramic indoor view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environmental transition zone area ROI3;

[0117] Light distribution strategy generation module: used to identify the status of ROI1 and ROI2 areas, and generate lighting control strategies based on the status identification results of ROI1 and ROI2 areas.

[0118] In the lighting control area recognition module, the method for obtaining the indoor panoramic bird's-eye view is: using fisheye cameras installed at the four corners of the ceiling to collect indoor scene images, performing distortion correction and image stitching on the indoor scene images, and obtaining the indoor panoramic bird's-eye view.

[0119] In the light distribution strategy generation module, the method for identifying the states of the ROI1 and ROI2 regions includes:

[0120] Step S1310: Real-time monitoring of the brightness LA of the ROI1 area. When LA exceeds the first brightness threshold L tha When , it is determined that ROI1 is activated; otherwise, it is determined that ROI1 is not activated;

[0121] Step S1320 , detecting a human activity signal in the ROI2 area. If a human activity signal is detected, the ROI2 area is determined to be in an active state, otherwise it is determined to be in an idle state.

[0122] In the light distribution strategy generation module, the method for generating the lighting control strategy according to the state recognition results of the ROI1 area and the ROI2 area includes:

[0123] Step S1410: When ROI1 is activated and ROI2 is in working state, a regional light distribution strategy is started; the regional light distribution strategy is to perform anti-glare control on ROI1, fill light control on ROI2, and gradient control on ROI3;

[0124] Step S1420: When the ROI1 area is not activated but the ROI2 area is in the working state, the ROI2 fill light strategy is started, wherein the ROI2 fill light strategy controls the fill light of the ROI2 area and maintains uniform basic illumination of the ROI3 area;

[0125] Step S1430: If the ROI2 area is in an idle state, enter the energy-saving mode.

[0126] The methods and systems of the present application may be implemented in many ways. For example, the methods and systems of the present application may be implemented using software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps used in the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified.

[0127] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0128] The above-described specific embodiments further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is merely a specific 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 shall be included within the scope of protection of the present invention.

Claims

1. LED eye protection lighting control method, characterized in that, The method comprises: Obtain a panoramic bird's-eye view of the interior, and identify three types of lighting control areas using the panoramic bird's-eye view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environment transition zone area ROI3; Perform status recognition on ROI1 and ROI2, and generate a lighting control strategy based on the status recognition results of ROI1 and ROI2; the status of ROI1 includes activated and inactivated; The method for identifying the state of the ROI2 region includes: detecting a human activity signal in the ROI2 region, and if no human activity signal is detected for a continuous Δt1 time, determining that the ROI2 region is in an idle state, otherwise it is in a working state; The method for detecting human activity signals in the ROI2 area is as follows: a millimeter-wave radar is deployed in the ROI2 area, and if the radar echo amplitude is lower than a preset echo amplitude threshold and the duration is greater than t3, it is determined that a human activity signal exists; otherwise, it is determined that no human activity signal exists; The lighting control strategy includes starting a regional light distribution strategy when ROI1 is activated and ROI2 is in working state; the regional light distribution strategy is to perform anti-glare control on ROI1, fill light control on ROI2, and gradient control on ROI3; The method for controlling fill light in ROI2 area includes: constructing a material reflection characteristic table for ROI2 area, obtaining the material of the operation plane of ROI2 area, querying the material reflection characteristic table to obtain the critical reflection angle and half-height width reflection angle of the corresponding material; adjusting the illumination angle of the main light source in ROI2 area according to the critical reflection angle and half-height width reflection angle of the corresponding material.

2. The LED eye protection lighting control method according to claim 1, characterized in that: The lighting control area detection model adopts the YOLOv5 framework.

3. The LED eye protection lighting control method according to claim 2, characterized in that: The method for identifying the state of the ROI1 region includes: monitoring the brightness LA of the ROI1 region in real time, and when LA exceeds a first brightness threshold L tha , the ROI1 area is determined to be activated; otherwise, the ROI1 area is determined to be inactivated.

4. The LED eye protection lighting control method according to claim 3, characterized in that: The lighting control strategy also includes: When ROI1 is not activated but ROI2 is in working state, the ROI2 fill light strategy is started; If ROI2 is in idle state, it enters energy saving mode.

5. The LED eye protection lighting control method according to claim 4, characterized in that: The method for performing fill light control on the ROI2 area further includes: classifying the working state of the ROI2 area, and performing fill light control according to the classification result of the working state of the ROI2 area.

6. The LED eye protection lighting control method according to claim 5, characterized in that: The method for classifying the working state of the ROI2 area includes: An ROI2 image is obtained by cutting out the indoor panoramic bird's-eye view. Based on the ROI2 image and the pre-built operation plane object detection model, paper documents and electronic devices on the operation plane of the ROI2 are identified. If an electronic device is identified, determining whether the electronic device is in working state; The working state of the ROI2 area is classified according to the recognition results of the paper documents and electronic devices on the operation plane of the ROI2 area and whether the electronic devices are in the working state.

7. An LED eye protection lighting control system, which is used to implement the LED eye protection lighting control method according to any one of claims 1 to 6, characterized in that: The system comprises: Lighting control area identification module: used to obtain a panoramic indoor view and identify three types of lighting control areas based on the panoramic indoor view and a pre-built lighting control area detection model; the three types of lighting control areas are display device area ROI1, operation plane area ROI2, and environmental transition zone area ROI3; Light distribution strategy generation module: used to identify the status of ROI1 and ROI2 areas, and generate lighting control strategies based on the status identification results of ROI1 and ROI2 areas.

Images