Indoor lighting control method and system based on Internet of Things

Abstract

The present invention discloses an indoor lighting control method and system based on the Internet of Things, which relates to the field of intelligent lighting control technology, including: collecting indoor spectral distribution data through a spectrum analyzer, building a spectral feature fingerprint library using an edge computing unit, and matching and classifying the spectral feature fingerprints based on distributed computing nodes; inputting the spectral feature fingerprints into a preset light physiological mapping model to generate a spectral compensation sequence, which includes the target output ratio of spectra of different wavelengths; converting the spectral compensation sequence into an indoor lighting control instruction, executing the indoor lighting control instruction through an RGBW multi-color LED array and a microlens array, and adjusting the lighting environment of each area in the room. The present invention realizes accurate perception and intelligent regulation of complex light environments, and improves the intelligence level and user experience of indoor lighting.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent lighting control, and in particular to an indoor lighting control method and system based on the Internet of Things. Background Art

[0002] Indoor lighting control technology is an important part of building intelligence. Traditional indoor lighting control mainly relies on light intensity sensors to collect environmental parameters and perform simple switching or dimming control through preset thresholds. With the development of Internet of Things technology, intelligent lighting systems have gradually introduced multi-source data fusion and distributed computing architecture to achieve more sophisticated lighting environment adjustment. The current mainstream indoor lighting control system uses RGB three-color LED as an artificial light source, combined with natural light sensors and motion sensors to build a multi-level control network. This type of system usually adjusts the light environment based on two dimensions: illumination and color temperature, which to a certain extent improves the level of lighting intelligence.

[0003] However, the existing technology has the following technical problems: First, the traditional single-point spectrum acquisition method is difficult to accurately reflect the spectral distribution characteristics in complex indoor environments, especially in scenes where natural light and artificial light sources are mixed, and sampling deviations are prone to occur. Secondly, the existing feature extraction methods often use simple peak detection or Fourier transform, which has problems of feature loss and misjudgment when dealing with multi-light source mixed scenes. In addition, most systems use fixed compensation strategies and static feature libraries, lack the ability to differentiate between different types of light sources, and are difficult to adapt to dynamically changing lighting needs. At the same time, in terms of spectral regulation, the existing systems mainly focus on the adjustment of illumination and color temperature, ignoring the impact of spectral composition on human visual comfort. At the execution level, traditional LED lighting systems use ordinary scattering lenses for light distribution, which makes it difficult to achieve precise spatial light field regulation. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the present invention provides an indoor lighting control method and system based on the Internet of Things, which can solve the problems mentioned in the background technology.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an indoor lighting control method based on the Internet of Things, comprising: collecting indoor spectral distribution data, using an edge computing unit to build a spectral feature fingerprint library containing multiple spectral feature fingerprints, and matching and classifying the spectral feature fingerprints based on distributed computing nodes; the spectral feature fingerprints contain wavelength composition information of natural light and artificial light sources; the spectral feature fingerprints are input into a preset photophysiological mapping model to generate a spectral compensation sequence; the spectral compensation sequence is converted into an indoor lighting control instruction, and the indoor lighting control instruction is executed through an RGBW multi-color LED array and a microlens array to adjust the lighting environment of each area in the room.

[0007] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, the indoor spectral distribution data is collected by a spectrum analyzer and converted into a corresponding relationship between spectral intensity and wavelength; wherein the spectrum analyzer adopts an array spectral sensor to synchronously collect the indoor spectral distribution data at different positions indoors, decomposes the indoor spectral distribution data into monochromatic light of different wavelengths through a grating dispersion element, and uses a photoelectric detector array to measure the light intensity value at each wavelength to generate the corresponding relationship between the spectral intensity and the wavelength.

[0008] The construction of the spectral feature fingerprint library includes the following steps: the edge computing unit performs multi-scale analysis on the correspondence between the spectral intensity and the wavelength through wavelet transform to extract the wavelength composition information; the wavelength composition information is used as the feature quantity of the spectral feature fingerprint; the edge computing unit constructs the spectral feature fingerprint library based on the feature quantity, specifically, the feature quantity is grouped and stored according to the wavelength range, and an index relationship between the feature quantities is established. Among them, the wavelength composition information includes the spectral intensity peak and the wavelength corresponding to the spectral intensity peak.

[0009] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, the matching includes the following steps: calculating the spectral energy distribution characteristics of the wavelength composition information in the spectral feature fingerprint to be matched; the spectral energy distribution characteristics include continuous spectral energy and discrete spectral energy.

[0010] The spectral feature fingerprint is matched, specifically: if the continuous spectrum energy is greater than the discrete spectrum energy, the wavelength distribution curve of the continuous spectrum feature is subjected to correlation analysis with the continuous spectrum feature stored in the spectral feature fingerprint library, and the correlation coefficient of the wavelength distribution curve is calculated; if the correlation coefficient of the wavelength distribution curve is greater than the ratio of the continuous spectrum energy to the discrete spectrum energy, it is determined that the match is successful, and the spectral feature fingerprint is marked as a natural light feature fingerprint; if the correlation coefficient of the wavelength distribution curve is less than or equal to the ratio of the continuous spectrum energy to the discrete spectrum energy, it is determined that the match fails, and the spectral feature fingerprint is marked is the fingerprint to be analyzed; if the continuous spectrum energy is less than or equal to the discrete spectrum energy, the wavelength peak position of the discrete spectrum feature is analyzed with the discrete spectrum feature stored in the spectrum feature fingerprint library for peak overlap; if the peak overlap is greater than the ratio of the discrete spectrum energy to the continuous spectrum energy, it is determined that the match is successful, and the spectrum feature fingerprint is marked as the artificial light source feature fingerprint; if the peak overlap is less than or equal to the ratio of the discrete spectrum energy to the continuous spectrum energy, it is determined that the match fails, and the spectrum feature fingerprint is marked as the fingerprint to be analyzed; the peak overlap is the ratio of the energy in the overlapping wavelength range to the total energy.

[0011] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, the marking result of the spectral feature fingerprint being marked as a natural light feature fingerprint, an artificial light source feature fingerprint or a fingerprint to be analyzed is used as a matching result, and the spectral feature fingerprint is classified based on the matching result, specifically: if the spectral feature fingerprint is marked as the natural light feature fingerprint, it is classified as a natural light-dominated type, and the continuous spectrum feature is used as the main feature; if the spectral feature fingerprint is marked as the artificial light source feature fingerprint, it is classified as an artificial light source-dominated type, and the discrete spectrum feature is used as the main feature; if the spectral feature fingerprint is marked as the fingerprint to be analyzed, the spectrum analyzer is used again to sample the light source multiple times to obtain multiple indoor spectral distribution data, the average spectral distribution of the multiple indoor spectral distribution data is calculated, and the matching and classification steps are repeated based on the average spectral distribution; if it is marked as the fingerprint to be analyzed again, it is stored in the spectral feature fingerprint library as a new feature type.

[0012] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, the calculation of the spectral energy distribution characteristics includes: dividing the wavelength range of the wavelength composition information into a number of wavelength intervals, and calculating the spectral intensity integral value in each of the wavelength intervals; defining the area where the spectral intensity integral value change rate of adjacent wavelength intervals is less than a first preset threshold as a continuous spectrum area, and defining the area where the spectral intensity integral value change rate of adjacent wavelength intervals is greater than or equal to the first preset threshold as a discrete spectrum area; the continuous spectrum energy is the integral sum of the spectral intensities in the continuous spectrum area, and the discrete spectrum energy is the integral sum of the spectral intensities in the discrete spectrum area.

[0013] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, wherein: the spectral compensation sequence includes target output ratios of spectra of different wavelengths, and the photophysiological mapping model is obtained by training based on pupil response data; the spectral feature fingerprint is input into a preset photophysiological mapping model to generate a spectral compensation sequence, comprising the following steps: inputting the spectral feature fingerprint into the photophysiological mapping model for mapping processing; wherein the photophysiological mapping model is constructed using a deep neural network, using the pupil response data of the human eye under different spectral conditions as training samples, and establishing a mapping relationship between spectral features and visual comfort; when the spectral feature fingerprint is marked as natural light-dominated, the photophysiological mapping model calculates the target light output ratio according to the energy distribution of the continuous spectrum feature. a target energy compensation value; when the spectral feature fingerprint is marked as artificial light source dominated, the photophysiological mapping model calculates the target wavelength compensation value according to the peak wavelength of the discrete spectral feature; the target energy compensation value and the target wavelength compensation value constitute a mapping result; the spectral compensation sequence is generated according to the mapping processing result, specifically, the target spectral energy distribution is calculated based on the mapping processing result, and compared with the current spectral energy distribution; if the spectral feature fingerprint is marked as natural light dominated, the energy compensation amount is generated based on the target energy compensation value; if the spectral feature fingerprint is marked as artificial light source dominated, the wavelength compensation amount is generated based on the target wavelength compensation value; the wavelength compensation amount and the energy compensation amount together constitute the spectral compensation sequence.

[0014] As a preferred solution of the indoor lighting control method based on the Internet of Things described in the present invention, wherein: converting the spectral compensation sequence into an indoor lighting control instruction comprises the following steps: calculating the driving current parameters of each color LED of RGBW according to the wavelength compensation amount and the energy compensation amount in the spectral compensation sequence; if the spectral feature fingerprint is marked as natural light-dominated, calculating the target luminous flux of each wavelength interval based on the energy compensation amount, and decomposing it into driving current values ​​of four channels of RGBW; if the spectral feature fingerprint is marked as artificial light-dominated, calculating the output power of each characteristic wavelength based on the wavelength compensation amount; The method comprises the steps of: adjusting the driving current of each LED unit in the RGBW multicolor LED array based on the indoor lighting control instruction; wherein the RGBW multicolor LED array is arranged in a matrix, and each LED unit is controlled by an independent driving circuit; each microlens unit in the microlens array corresponds to the corresponding LED unit one by one, and directional light distribution is achieved by adjusting the focal length and inclination angle of the microlens unit to form a controllable indoor light field distribution.

[0015] To further solve the above technical problems, the present invention provides the following technical solutions: an indoor lighting control system based on the Internet of Things, comprising: a spectral data processing module, used to collect indoor spectral distribution data through a spectrum analyzer, use an edge computing unit to build a spectral feature fingerprint library, and match and classify the spectral feature fingerprints based on distributed computing nodes; a photophysiological response modeling module, used to input the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence; a lighting control execution module, used to convert the spectral compensation sequence into an indoor lighting control instruction, execute the indoor lighting control instruction through an RGBW multi-color LED array and a microlens array, and adjust the lighting environment of each area in the room.

[0016] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the indoor lighting control method based on the Internet of Things when executing the computer program.

[0017] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the indoor lighting control method based on the Internet of Things as described above.

[0018] The beneficial effects of the present invention are as follows: the present invention realizes multi-point synchronous acquisition through the array spectrum sensor of the spectrum analyzer, and solves the sampling blind area problem existing in the traditional single-point measurement by cooperating with the spectroscopic effect of the grating dispersion element; in the feature extraction link, the edge computing unit is used to perform multi-scale analysis of wavelet transform, which overcomes the feature loss problem of conventional peak detection when processing complex light source mixtures; in terms of light source identification, the present invention adopts a comparison mechanism of continuous spectrum energy and discrete spectrum energy, combined with the correlation analysis of wavelength distribution curve and the peak overlap analysis, so as to improve the system's recognition accuracy of natural light and new LED light sources after being changed by building materials; in the spectrum compensation link, the photophysiological mapping model constructed based on the CNN-LSTM combination structure can distinguish and process different characteristics of natural light and artificial light sources, and avoids the problem of insufficient spectrum control accuracy caused by the traditional method of using unified processing for different types of light sources through differentiated compensation strategies; at the execution level, the one-to-one correspondence design of the RGBW multi-color LED array and the microlens array realizes accurate light supplement for areas with insufficient natural light and interference suppression for artificial light sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 This is a schematic diagram of the overall process of an indoor lighting control method based on the Internet of Things proposed by the present invention;

[0021] Figure 2 This is a schematic diagram of module interaction of an indoor lighting control system based on the Internet of Things proposed by the present invention;

[0022] Figure 3 This is a diagram of computer equipment in an indoor lighting control method based on the Internet of Things proposed by the present invention. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides an indoor lighting control method based on the Internet of Things.

[0026] Figure 1 The overall process diagram of an indoor lighting control method based on the Internet of Things is shown, including the following steps:

[0027] S1: Collect indoor spectral distribution data, use the edge computing unit to build a spectral feature fingerprint library containing multiple spectral feature fingerprints, and match and classify the spectral feature fingerprints based on distributed computing nodes.

[0028] Specifically, the spectral characteristic fingerprint includes wavelength composition information of natural light and artificial light sources.

[0029] S1.1: Collect the indoor spectrum distribution data through a spectrum analyzer, and convert the indoor spectrum distribution data into a correspondence between spectrum intensity and wavelength.

[0030] Specifically, the spectrum analyzer adopts an array spectrum sensor to synchronously collect the indoor spectrum distribution data at different positions indoors, decomposes the indoor spectrum distribution data into monochromatic light of different wavelengths through a grating dispersion element, and uses a photodetector array to measure the light intensity value at each wavelength to generate the corresponding relationship between the spectrum intensity and the wavelength.

[0031] Exemplarily, a spectrum analyzer equipped with a 2048-pixel CCD array is used to synchronously collect spectral data at 5 measurement points. The wavelength range of the spectrum analyzer is 380-780nm, and the resolution is 0.5nm. The sampling frequency of each measurement point is 1Hz, and stable data is obtained by continuous sampling for 30 seconds. The spectral analysis results show that in the spectral data measured in the window area (measurement point 1), the continuous spectrum component of natural light has an average energy of 226.7μW / cm²·nm in the range of 450-650nm, while the discrete spectrum of fluorescent lamps measured in the area far from the window (measurement point 5) is mainly concentrated at three wavelengths of 435nm (57.3μW / cm²·nm), 546nm (63.2μW / cm²·nm) and 611nm (48.7μW / cm²·nm).

[0032] It should be noted that the traditional indoor spectral acquisition method usually adopts single-point measurement, which requires staff to take samples multiple times at different locations in the room, which is not only time-consuming but also prone to sampling blind spots. In addition, conventional spectrometers often use scanning measurements, which makes it difficult to capture complete spectral information when the light source changes rapidly. In the present invention, S1.1 uses an array spectral sensor to achieve multi-point synchronous acquisition, and cooperates with the grating dispersion element and the photodetector array to obtain complete spectral data at different locations in the room at the same time. This solution avoids the data deviation caused by inconsistent sampling time in traditional methods and improves the spatiotemporal consistency of spectral data. Especially in the scene where natural light and artificial light sources are mixed, it can accurately capture the spectral change characteristics of different regions, laying a reliable data foundation for subsequent feature extraction and classification.

[0033] S1.2: Utilize the edge computing unit to extract the wavelength composition information from the indoor spectral distribution data and construct the spectral feature fingerprint library.

[0034] Specifically, the edge computing unit performs multi-scale analysis on the correspondence between the spectral intensity and the wavelength through wavelet transform, extracts the wavelength composition information, and uses the wavelength composition information as the feature quantity of the spectral feature fingerprint; the edge computing unit constructs the spectral feature fingerprint library based on the feature quantity, including grouping and storing the feature quantity according to the wavelength range, and establishing an index relationship between the feature quantities. The wavelength composition information includes the spectral intensity peak and its corresponding wavelength

[0035] Exemplarily, in the edge computing unit, the db4 wavelet transform is used to perform a five-level decomposition of the correspondence between spectral intensity and wavelength, and the low-frequency approximate coefficient (A5) and high-frequency detail coefficient (D1-D5) are extracted respectively. For natural light samples, the energy of the A5 coefficient accounts for 85.3% of the total energy, indicating that the spectral distribution is mainly continuous components; while for LED light source samples, the energy of the D3 and D4 coefficients accounts for 64.7%, reflecting obvious discrete spectrum characteristics. The wavelength composition information is used as a feature quantity and stored in groups according to the wavelength interval of 20nm. A library containing 25 main spectral feature fingerprints is constructed, including 12 natural light types, 8 LED types, and 5 fluorescent lamp types. The index relationship between feature quantities is realized through a hash table, and the average retrieval time is less than 5ms.

[0036] Preferably, the spectral feature extraction in the prior art usually adopts a simple peak detection or Fourier transform method, which is prone to feature loss or misjudgment when dealing with complex light source mixing situations. The present invention introduces wavelet transform in S1.2 for multi-scale analysis, which can simultaneously obtain the characteristic information of the spectral signal at different scales. The present invention can be used to process complex scenes where natural light and artificial light sources coexist, because wavelet transform can effectively separate spectral components of different characteristic scales. By constructing a feature fingerprint library by storing the feature quantities in groups according to wavelength intervals and establishing an index relationship, not only the efficiency of feature matching is improved, but also flexible data structure support is provided for the dynamic update of spectral features. This feature extraction and storage scheme overcomes the technical problem that traditional methods are difficult to accurately characterize the characteristics of complex spectral mixing, enabling the system to more accurately distinguish and identify different types of light source combinations.

[0037] S1.3: Matching and classifying the spectral feature fingerprints based on distributed computing nodes.

[0038] Specifically, the matching and classification includes the following steps:

[0039] S1.3.1: Calculate the spectral energy distribution characteristics of the wavelength composition information in the spectral feature fingerprint to be matched, wherein the spectral energy distribution characteristics include continuous spectral energy and discrete spectral energy.

[0040] Specifically, the calculation of the spectral energy distribution characteristics includes: firstly, dividing the wavelength range of the wavelength composition information into a plurality of wavelength intervals, calculating the spectral intensity integral value in each wavelength interval, defining the region where the spectral intensity integral value change rate of adjacent wavelength intervals is less than the first preset threshold as a continuous spectrum region, and defining the region where the spectral intensity integral value change rate of adjacent wavelength intervals is greater than or equal to the first preset threshold as a discrete spectrum region; the continuous spectrum energy is the sum of the spectral intensity integrals in the continuous spectrum region, and the discrete spectrum energy is the sum of the spectral intensity integrals in the discrete spectrum region. Among them, the spectral intensity integral value is the integral of the spectral intensity with the wavelength in the wavelength interval, expressed as the total energy in the wavelength interval; the spectral intensity integral value change rate is the ratio of the difference between the spectral intensity integral values ​​of adjacent wavelength intervals and the interval width.

[0041] For example, in a mixed light source scenario, we divide the wavelength range of 380-780nm into 20 wavelength intervals, each with a width of 20nm. The calculated spectral intensity integral values ​​of each interval are as follows (unit: μW / cm²):

[0042] 380-400nm: 18.5;

[0043] 400-420nm: 23.1;

[0044] 420-440nm: 84.7 (fluorescent lamp peak range);

[0045] 440-460nm: 46.2;

[0046] 460-480nm: 42.8;

[0047] 480-500nm: 45.3;

[0048] 500-520nm: 47.9;

[0049] 520-540nm: 49.2;

[0050] 540-560nm: 76.8 (fluorescent lamp peak range);

[0051] 560-580nm: 53.4;

[0052] The change rate of the spectral intensity integral value of adjacent wavelength intervals is calculated. For example, the change rate from 420-440nm to 440-460nm is (84.7-46.2) / 20nm = 1.925, which is significantly greater than the set first preset threshold of 0.2. Therefore, 420-440nm is identified as a discrete spectrum region. Through similar calculations, 420-440nm, 540-560nm and 600-620nm are identified as discrete spectrum regions, and the rest are continuous spectrum regions. The final calculation results show that the continuous spectrum energy is 487.9μW / cm², the discrete spectrum energy is 206.2μW / cm², and the ratio of continuous spectrum energy to discrete spectrum energy is 2.37.

[0053] It should be noted that the first preset threshold is used to distinguish between continuous spectrum and discrete spectrum features, and its value depends on the spectral characteristics of the light source. Natural light usually exhibits a smooth and continuous spectral distribution, while artificial light sources such as LEDs and fluorescent lamps have obvious spectral line characteristics. Based on the statistical analysis of a large amount of indoor spectral sampling data, when the spectral intensity of adjacent wavelength intervals changes by more than 20%, it usually indicates that there are significant spectral line features there. Therefore, the first preset threshold is preferably set to 0.2, that is, when the spectral intensity integral value of adjacent wavelength intervals changes by more than 20%, it is judged as a discrete spectrum feature. The selection of this value can not only effectively identify the characteristic spectral lines of LEDs, but also adapt to the fluctuation range of natural light under different weather conditions.

[0054] S1.3.2: Matching the spectral feature fingerprint, specifically including:

[0055] If the continuous spectrum energy is greater than the discrete spectrum energy, a correlation analysis is performed on the wavelength distribution curve of the continuous spectrum feature and the continuous spectrum features stored in the spectrum feature fingerprint library to calculate the correlation coefficient of the wavelength distribution curve.

[0056] If the correlation coefficient of the wavelength distribution curve is greater than the ratio of the continuous spectral energy to the discrete spectral energy, the match is determined to be successful, and the spectral feature fingerprint is marked as a natural light feature fingerprint; if the correlation coefficient of the wavelength distribution curve is less than or equal to the ratio of the continuous spectral energy to the discrete spectral energy, the match is determined to have failed, and the spectral feature fingerprint is marked as a fingerprint to be analyzed.

[0057] If the continuous spectrum energy is less than or equal to the discrete spectrum energy, the wavelength peak position of the discrete spectrum feature is analyzed with the discrete spectrum feature stored in the spectral feature fingerprint library for peak overlap; wherein the peak overlap is the ratio of the energy within the overlapping wavelength range to the total energy.

[0058] If the peak overlap is greater than the ratio of the discrete spectral energy to the continuous spectral energy, the match is determined to be successful, and the spectral feature fingerprint is marked as the artificial light source feature fingerprint; if the peak overlap is less than or equal to the ratio of the discrete spectral energy to the continuous spectral energy, the match is determined to be unsuccessful, and the spectral feature fingerprint is marked as the fingerprint to be analyzed.

[0059] For example, for a sampling point dominated by natural light, the calculated continuous spectrum energy is 325.8μW / cm², the discrete spectrum energy is 98.2μW / cm², and the ratio is 3.32. The wavelength distribution curve of its continuous spectrum feature is correlated with the standard day spectrum (D65) in the spectral feature fingerprint library, and the Pearson correlation coefficient is calculated to be 0.89, which is greater than the inverse of the ratio of continuous spectrum energy to discrete spectrum energy, 0.30. Therefore, the spectral feature fingerprint is successfully matched and marked as a natural light feature fingerprint.

[0060] For example, in an indoor environment mainly illuminated by T5 fluorescent lamps, the measured discrete spectrum energy of the spectral feature is 156.4μW / cm², the continuous spectrum energy is 67.9μW / cm², and the ratio is 2.30. The main discrete spectrum peaks of the spectrum are located at 435nm, 546nm and 611nm, and the peak overlap analysis is performed with the standard peak positions (434nm, 545nm and 610nm) of typical T5 fluorescent lamps in the feature fingerprint library. Considering the wavelength deviation of ±5nm, the energy in the overlapping wavelength range is calculated to be 139.8μW / cm², accounting for 89.4% of the total energy of the discrete spectrum. This overlap is greater than the inverse of the ratio of discrete spectrum energy to continuous spectrum energy (2.30), 0.43, so the system successfully matches this spectral feature fingerprint and marks it as an artificial light source feature fingerprint.

[0061] It should be noted that, at the level of light source feature recognition, the prior art usually uses a single energy threshold or a fixed feature template to judge the light source type. This method often makes misjudgments when dealing with complex indoor lighting environments. For example, when natural light enters the room through tinted glass, its spectral characteristics will change, and it is easy to be mistakenly identified as an artificial light source. The present invention proposes a hierarchical recognition mechanism based on spectral energy distribution characteristics. By calculating the proportional relationship between continuous spectral energy and discrete spectral energy, combined with the correlation analysis of wavelength distribution curves and peak overlap analysis, a multi-dimensional feature judgment system is established. Therefore, the present invention can not only accurately distinguish natural light after being changed by building materials, but also identify light sources with special spectral characteristics such as new LED lamps, thereby improving the accuracy. At the feature matching strategy level, traditional methods often use a fixed matching threshold, which is difficult to adapt to the lighting change characteristics of different seasons and different time periods. For example, the spectral distribution of natural light will become relatively flat on cloudy days, while a unique spectral peak will appear at sunset. The dynamic matching criterion of the present invention, that is, using the spectral energy ratio as an adaptive threshold, enables the matching standard to be automatically adjusted according to the current lighting conditions. When processing composite light sources, the dominant light source type can be accurately identified, providing a more accurate basis for subsequent lighting adjustment. This matching mechanism greatly improves the adaptability of the present invention under different environmental conditions, making lighting adjustment more intelligent and accurate.

[0062] S1.3.3: The marking result of the spectral feature fingerprint being marked as a natural light feature fingerprint, an artificial light source feature fingerprint, or a fingerprint to be analyzed is used as a matching result, and the spectral feature fingerprint is classified based on the matching result:

[0063] If the spectral feature fingerprint is marked as the natural light feature fingerprint, it is classified as natural light dominated type, and the continuous spectrum feature is used as the main feature;

[0064] If the spectral feature fingerprint is marked as the artificial light source feature fingerprint, it is classified as an artificial light source dominated type, and the discrete spectrum feature is used as the main feature.

[0065] If the spectral feature fingerprint is marked as the fingerprint to be analyzed, the spectrum analyzer is used again to sample the light source multiple times to obtain multiple indoor spectral distribution data, the average spectral distribution of the multiple indoor spectral distribution data is calculated, and the matching and classification steps are repeated based on the average spectral distribution; if it is marked as the fingerprint to be analyzed again, it is stored in the spectral feature fingerprint library as a new feature type.

[0066] For example, in an environment using a new quantum dot LED lamp, the first sampling analysis results show that the continuous spectrum energy is 112.6μW / cm², the discrete spectrum energy is 105.8μW / cm², and the ratio is 1.06. Because the two energies are close, the system attempts to perform a correlation analysis, but the calculated correlation coefficient is 0.68, which is less than the inverse of the energy ratio of 0.94. At the same time, the peak overlap is 0.71, which is also less than the inverse of the energy ratio of 0.94. Therefore, the spectral feature fingerprint is preliminarily marked as the fingerprint to be analyzed.

[0067] The light source was then sampled five additional times, each with an interval of 10 seconds, and the average spectral distribution of the five samples was calculated. The averaged data showed that the continuous spectrum energy was 110.2μW / cm², the discrete spectrum energy was 107.4μW / cm², the ratio was 1.03, and the peak overlap was 0.73, which was still less than the inverse of the energy ratio of 0.97. The system finally marked this spectral feature fingerprint as a fingerprint to be analyzed, and stored its feature data (including the main peaks at 455nm, 528nm and 638nm) as a new feature type in the spectral feature fingerprint library, named "quantum dot LED type" for subsequent identification.

[0068] Preferably, conventional methods usually use a static feature library, which cannot cope with new light sources or special lighting conditions. Therefore, the present invention proposes a feature verification mechanism based on multiple sampling and average spectral distribution. By performing a secondary confirmation on the fingerprint to be analyzed, it not only avoids misjudgment caused by instantaneous interference, but also can effectively identify and record new light source features. The present invention thus has the ability to continuously evolve and can continuously adapt to new lighting equipment and complex indoor light environments. In intelligent buildings, the self-updating ability of this feature library ensures that the system can always maintain the best recognition performance, laying the foundation for achieving precise lighting control.

[0069] S2: Inputting the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence.

[0070] Specifically, the spectral compensation sequence includes target output ratios of spectra of different wavelengths, and the photophysiological mapping model is trained based on pupil response data.

[0071] S2.1: inputting the spectral feature fingerprint into the photophysiological mapping model for mapping processing;

[0072] Specifically, the photophysiological mapping model is constructed using a deep neural network, and the pupil response data of the human eye under different spectral conditions is used as a training sample to establish a mapping relationship between spectral features and visual comfort; when the spectral feature fingerprint is marked as natural light-dominated, the photophysiological mapping model calculates the target energy compensation value according to the energy distribution of the continuous spectrum feature; when the spectral feature fingerprint is marked as artificial light-dominated, the photophysiological mapping model calculates the target wavelength compensation value according to the peak wavelength of the discrete spectrum feature. The target energy compensation value and the target wavelength compensation value constitute a mapping result.

[0073] It should be noted that the photophysiological mapping model can be constructed using one or more combinations of convolutional neural networks CNN, recurrent neural networks RNN, long short-term memory networks LSTM or gated recurrent unit networks GRU. Among them, the CNN network is used to extract the local features and global features of the spectral feature fingerprint, the RNN network is used to capture the changing laws of pupil reactions at different times, and the LSTM network and the GRU network are used to establish long-term dependencies to overcome the timing differences between spectral features and pupil reactions. Preferably, in this embodiment, the photophysiological mapping model adopts a combined structure of CNN and LSTM, wherein CNN is used for spatial feature extraction and LSTM is used for temporal feature modeling. The combination of the two ensures both the accurate extraction of spectral features and the accurate prediction of pupil dynamic responses.

[0074] Specifically, the combined structure of CNN and LSTM includes: the CNN part is composed of multiple convolutional layers and pooling layers, which are used to extract multi-scale features from the spectral feature fingerprint, wherein the first convolutional layer extracts basic spectral edge and peak features, the second convolutional layer extracts the spectral energy distribution pattern, and the third convolutional layer comprehensively extracts high-level spectral features; the feature output extracted by CNN is reorganized into a temporal feature sequence and input to the LSTM layer; the LSTM layer contains multiple memory units, each of which includes an input gate, a forget gate and an output gate, wherein the input gate controls the input amount of the current spectral feature, the forget gate is used to selectively retain or forget historical pupil response information, and the output gate is used to control the output of the current state; through this structure, unified processing of spatial analysis of spectral features and temporal modeling of pupil response is achieved.

[0075] Preferably, traditional indoor lighting control systems usually adopt simple light intensity mapping or fixed compensation strategies, which cannot accurately reflect the actual impact of spectral characteristics on the human visual system. In a complex light source environment, such as an office with both natural lighting and LED lighting, simple control strategies often lead to visual fatigue. The present invention adopts a CNN-LSTM combined structure in S2.1 to construct a photophysiological mapping model. Through the multi-layer feature extraction capability of CNN, the local detail features of the spectrum (such as the characteristic peak of the LED) and the global distribution features (such as the continuous spectrum of natural light) can be captured simultaneously, while the memory mechanism of LSTM can learn and adapt to the dynamic adaptation process of the human eye to different spectral combinations. This deep learning architecture not only solves the problem that traditional methods are difficult to handle complex light source combinations, but also can predict the long-term impact of spectral changes on visual comfort through time series modeling, laying the foundation for achieving truly physiologically friendly lighting control.

[0076] S2.2: generating the spectrum compensation sequence according to the mapping processing result;

[0077] Specifically, the target spectral energy distribution is calculated based on the mapping processing result and compared with the current spectral energy distribution; if the spectral feature fingerprint is marked as natural light-dominated, the energy compensation amount is generated based on the target energy compensation value to determine the relative proportion of each wavelength component; if the spectral feature fingerprint is marked as artificial light-dominated, the wavelength compensation amount is generated based on the target wavelength compensation value to adjust the output intensity of spectra of different wavelengths; the wavelength compensation amount and the energy compensation amount together constitute the spectral compensation sequence.

[0078] Preferably, the prior art often adopts a unified processing method when generating a spectral compensation strategy, and fails to consider the characteristic differences of different types of light sources. For example, it is difficult to achieve precise spectral control by simply adjusting the total light intensity when compensating for natural light, or only adjusting the color temperature for artificial light sources. The S2.2 step of the present invention proposes a differentiated compensation strategy: for natural light-dominated scenes, the relative proportion of each wavelength component is adjusted by the energy compensation amount to maintain the continuous spectrum characteristics of natural light; for artificial light source-dominated scenes, the output intensity of the characteristic wavelength is accurately adjusted by the wavelength compensation amount. This differentiated compensation mechanism based on light source characteristics not only improves the accuracy of spectral control, but also minimizes unnecessary energy consumption while maintaining the lighting effect. Especially in scenes where lighting modes need to be switched frequently (such as conference rooms, multi-function halls, etc.), this solution can achieve a smooth transition of the spectrum, avoid the mutation problem under the traditional control method, and greatly improve the user's visual comfort experience.

[0079] S3: Convert the spectrum compensation sequence into indoor lighting control instructions, execute the indoor lighting control instructions through the RGBW multi-color LED array and the microlens array, and adjust the lighting environment of each area in the room.

[0080] The RGBW multi-color LED array has an independently addressable characteristic.

[0081] S3.1: Convert the spectrum compensation sequence into the indoor lighting control instruction.

[0082] Specifically, the driving current parameters of each RGBW color LED are calculated according to the wavelength compensation amount and the energy compensation amount in the spectral compensation sequence; if the spectral feature fingerprint is marked as natural light-dominated, the target luminous flux of each wavelength interval is calculated based on the energy compensation amount, and is decomposed into driving current values ​​of four RGBW channels; if the spectral feature fingerprint is marked as artificial light source-dominated, the output power of each characteristic wavelength is calculated based on the wavelength compensation amount, and converted into a corresponding RGBW driving current combination.

[0083] It should be noted that in the RGBW multicolor LED array, R represents red LED, G represents green LED, B represents blue LED, and W represents white LED. Compared with the traditional RGB three-color LED, adding white LED not only improves the spectral energy utilization efficiency, but more importantly, it can better meet the continuous spectrum compensation requirements in natural light-dominated scenes in the present invention. This is because white LED has a relatively smooth spectral distribution characteristic and can be used as a basic light source to provide broadband lighting, while RGB three-color LED is used to accurately adjust the light intensity of a specific wavelength. The combination of the two can achieve more precise spectral control.

[0084] S3.2: Execute the indoor lighting control instruction through the RGBW multi-color LED array and the microlens array.

[0085] Specifically, the driving current of each LED unit in the RGBW multi-color LED array is adjusted based on the indoor lighting control instruction; wherein the RGBW multi-color LED array adopts a matrix arrangement, and each LED unit is controlled by an independent driving circuit; each microlens unit in the microlens array corresponds one-to-one to the corresponding LED unit, and by adjusting the focal length and inclination angle of the microlens unit, directional light distribution is achieved to form a controllable indoor light field distribution.

[0086] It should be noted that traditional LED lighting systems usually use ordinary scattering lenses or reflectors for light distribution, which makes it difficult to achieve precise spatial light field control. The one-to-one correspondence between the microlens array and the RGBW multi-color LED array in the present invention enables the output light of each LED unit to be independently and accurately controlled. In natural light-dominated scenes, precise fill light can be achieved in areas with insufficient natural light by adjusting the focal length and inclination of the microlens; in artificial light-dominated scenes, the light distribution control of the microlens array can be used to reduce interference between different light sources and avoid glare.

[0087] In addition, the present invention considers the differentiated processing of light source types in the process of converting the spectral compensation sequence into RGBW driving current parameters. For natural light-dominated scenes, broadband fill light is mainly achieved by adjusting the driving current of white light LEDs, while fine-tuning the RGB three-color LEDs to optimize the spectral energy distribution; for artificial light-dominated scenes, the narrowband characteristics of RGB three-color LEDs are mainly utilized to accurately match the compensation requirements of the target wavelength. This differentiated control strategy not only improves the energy utilization efficiency of the system, but also effectively avoids unnecessary spectral interference.

[0088] In summary, the present invention realizes multi-point synchronous acquisition through the array spectrum sensor of the spectrum analyzer, and solves the sampling blind area problem existing in the traditional single-point measurement by cooperating with the spectroscopic effect of the grating dispersion element; in the feature extraction link, the edge computing unit is used to perform multi-scale analysis of wavelet transform, which overcomes the feature loss problem of conventional peak detection when processing complex light source mixtures; in terms of light source identification, the present invention adopts a comparison mechanism of continuous spectrum energy and discrete spectrum energy, combined with the correlation analysis of wavelength distribution curve and the peak overlap analysis, so as to improve the system's recognition accuracy of natural light and new LED light sources after being changed by building materials; in the spectrum compensation link, the photophysiological mapping model constructed based on the CNN-LSTM combination structure can distinguish and process the different characteristics of natural light and artificial light sources, and avoids the problem of insufficient spectrum control accuracy caused by the traditional method of using unified processing for different types of light sources through differentiated compensation strategies; at the execution level, the one-to-one correspondence design of the RGBW multi-color LED array and the microlens array realizes accurate fill light for areas with insufficient natural light and interference suppression of artificial light sources.

[0089] Example 2, reference Figure 2 , which is an embodiment of the present invention, provides an indoor lighting control system based on the Internet of Things.

[0090] Figure 2 This is a schematic diagram of the module interaction of the system, including:

[0091] A spectral data processing module is used to collect indoor spectral distribution data through a spectrum analyzer, build a spectral feature fingerprint library using an edge computing unit, and match and classify the spectral feature fingerprints based on distributed computing nodes;

[0092] A photophysiological response modeling module, used for inputting the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence;

[0093] The lighting control execution module is used to convert the spectrum compensation sequence into indoor lighting control instructions, execute the indoor lighting control instructions through the RGBW multi-color LED array and the microlens array, and adjust the lighting environment of each area in the room.

[0094] Example 3, reference Figure 3 , is an embodiment of the present invention, which is different from the previous embodiment in that: if the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0095] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0096] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0097] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An indoor lighting control method based on the Internet of Things, characterized in that: include: Collect indoor spectral distribution data, use edge computing units to build a spectral feature fingerprint library containing multiple spectral feature fingerprints, and match and classify the spectral feature fingerprints based on distributed computing nodes; the spectral feature fingerprints contain wavelength composition information of natural light and artificial light sources; the indoor spectral distribution data is collected by a spectrum analyzer and converted into a correspondence between spectral intensity and wavelength; Inputting the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence; Convert the spectrum compensation sequence into indoor lighting control instructions, execute the indoor lighting control instructions through the RGBW multi-color LED array and the microlens array, and adjust the lighting environment of each area in the room; The construction of the spectral feature fingerprint library includes the following steps: The edge computing unit performs multi-scale analysis on the correspondence between the spectral intensity and the wavelength through wavelet transform to extract the wavelength composition information; Using the wavelength composition information as a feature quantity of the spectral feature fingerprint; The edge computing unit constructs the spectral feature fingerprint library based on the feature quantity, specifically by grouping and storing the feature quantity according to the wavelength range, and establishing an index relationship between the feature quantities; Wherein, the wavelength composition information includes the spectral intensity peak and the wavelength corresponding to the spectral intensity peak; The matching comprises the following steps: Calculating the spectral energy distribution characteristics of the wavelength composition information in the spectral feature fingerprint to be matched; the spectral energy distribution characteristics include continuous spectral energy and discrete spectral energy; The spectral feature fingerprint is matched, specifically: If the continuous spectrum energy is greater than the discrete spectrum energy, a correlation analysis is performed on the wavelength distribution curve of the continuous spectrum feature and the continuous spectrum feature stored in the spectrum feature fingerprint library to calculate the correlation coefficient of the wavelength distribution curve; If the correlation coefficient of the wavelength distribution curve is greater than the ratio of the continuous spectrum energy to the discrete spectrum energy, it is determined that the match is successful, and the spectrum feature fingerprint is marked as a natural light feature fingerprint; if the correlation coefficient of the wavelength distribution curve is less than or equal to the ratio of the continuous spectrum energy to the discrete spectrum energy, it is determined that the match fails, and the spectrum feature fingerprint is marked as a fingerprint to be analyzed; If the continuous spectrum energy is less than or equal to the discrete spectrum energy, performing peak overlap analysis on the wavelength peak position of the discrete spectrum feature and the discrete spectrum feature stored in the spectrum feature fingerprint library; If the peak overlap is greater than the ratio of the discrete spectrum energy to the continuous spectrum energy, it is determined that the match is successful, and the spectrum feature fingerprint is marked as the artificial light source feature fingerprint; if the peak overlap is less than or equal to the ratio of the discrete spectrum energy to the continuous spectrum energy, it is determined that the match fails, and the spectrum feature fingerprint is marked as the fingerprint to be analyzed; The peak overlap is the ratio of the energy in the overlapping wavelength range to the total energy; The spectral feature fingerprint is classified based on the matching result, specifically: If it is determined to be the natural light characteristic fingerprint, it is classified as the natural light-dominated type, and the continuous spectrum feature is used as the main feature; If it is determined to be the artificial light source characteristic fingerprint, it is classified as the artificial light source dominant type, and the discrete spectrum feature is used as the main feature; If it is determined to be the fingerprint to be analyzed, the spectrum analyzer is used again to sample the light source multiple times to obtain multiple indoor spectral distribution data, the average spectral distribution of the multiple indoor spectral distribution data is calculated, and the matching and classification steps are repeated based on the average spectral distribution; if it is determined to be the fingerprint to be analyzed again, it is stored in the spectral feature fingerprint library as a new feature type.

2. The indoor lighting control method based on the Internet of Things as claimed in claim 1, characterized in that: The spectrum analyzer adopts an array spectrum sensor to synchronously collect the indoor spectrum distribution data at different positions indoors, decomposes the indoor spectrum distribution data into monochromatic light of different wavelengths through a grating dispersion element, and uses a photodetector array to measure the light intensity value at each wavelength to generate the corresponding relationship between the spectrum intensity and the wavelength.

3. The indoor lighting control method based on the Internet of Things as claimed in claim 2, characterized in that: The calculation of the spectral energy distribution characteristics includes: Dividing the wavelength range of the wavelength composition information into a plurality of wavelength intervals, and calculating the spectral intensity integral value in each of the wavelength intervals; The region where the change rate of the spectral intensity integral value of the adjacent wavelength interval is less than the first preset threshold is defined as a continuous spectrum region, and the region where the change rate of the spectral intensity integral value of the adjacent wavelength interval is greater than or equal to the first preset threshold is defined as a discrete spectrum region; The continuous spectrum energy is the integrated sum of the spectrum intensities in the continuous spectrum region, and the discrete spectrum energy is the integrated sum of the spectrum intensities in the discrete spectrum region.

4. The indoor lighting control method based on the Internet of Things as claimed in claim 3, characterized in that: The spectral compensation sequence includes target output ratios of spectra of different wavelengths, and the photophysiological mapping model is trained based on pupil response data; The step of inputting the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence comprises the following steps: Inputting the spectral feature fingerprint into the photophysiological mapping model for mapping processing; wherein the photophysiological mapping model is constructed using a deep neural network, using pupil response data of human eyes under different spectral conditions as training samples, and establishing a mapping relationship between spectral features and visual comfort; When the spectral feature fingerprint is marked as natural light-dominated, the photophysiological mapping model calculates a target energy compensation value according to the energy distribution of the continuous spectrum feature; When the spectral feature fingerprint is marked as an artificial light source-dominated type, the photophysiological mapping model calculates a target wavelength compensation value according to a peak wavelength of the discrete spectral feature; The target energy compensation value and the target wavelength compensation value constitute a mapping result; Generating the spectral compensation sequence according to the mapping processing result, specifically calculating the target spectral energy distribution based on the mapping processing result, and comparing it with the current spectral energy distribution; If the spectral feature fingerprint is marked as natural light-dominated, generating an energy compensation amount based on the target energy compensation value; If the spectral characteristic fingerprint is marked as artificial light source dominated, generating a wavelength compensation amount based on the target wavelength compensation value; The wavelength compensation amount and the energy compensation amount together constitute the spectrum compensation sequence.

5. The indoor lighting control method based on the Internet of Things as claimed in claim 4, characterized in that: Converting the spectrum compensation sequence into indoor lighting control instructions comprises the following steps: Calculating the driving current parameters of the RGBW LEDs according to the wavelength compensation amount and the energy compensation amount in the spectrum compensation sequence; If the spectral feature fingerprint is marked as natural light-dominated, the target luminous flux of each wavelength interval is calculated based on the energy compensation amount, and is decomposed into driving current values ​​of four channels of RGBW; If the spectral characteristic fingerprint is marked as artificial light source dominated, the output power of each characteristic wavelength is calculated based on the wavelength compensation amount and converted into a corresponding RGBW driving current combination; Executing the indoor lighting control instruction by using the RGBW multi-color LED array and the microlens array includes the following steps: Adjusting the driving current of each LED unit in the RGBW multi-color LED array based on the indoor lighting control instruction; Wherein, the RGBW multi-color LED array is arranged in a matrix, and each LED unit is controlled by an independent driving circuit; Each microlens unit in the microlens array corresponds to a corresponding LED unit one by one. By adjusting the focal length and inclination angle of the microlens unit, directional light distribution is achieved to form a controllable indoor light field distribution.

6. An indoor lighting control system based on the Internet of Things, based on the indoor lighting control method based on the Internet of Things according to any one of claims 1 to 5, characterized in that: include, A spectral data processing module is used to collect indoor spectral distribution data through a spectrum analyzer, build a spectral feature fingerprint library using an edge computing unit, and match and classify the spectral feature fingerprints based on distributed computing nodes; A photophysiological response modeling module, used for inputting the spectral feature fingerprint into a preset photophysiological mapping model to generate a spectral compensation sequence; The lighting control execution module is used to convert the spectrum compensation sequence into indoor lighting control instructions, execute the indoor lighting control instructions through the RGBW multi-color LED array and the microlens array, and adjust the lighting environment of each area in the room.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the indoor lighting control method based on the Internet of Things described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the indoor lighting control method based on the Internet of Things described in any one of claims 1 to 5 are implemented.

Images