An ai screening method for an old-age lighting spectrum and a computer device

Abstract

The application discloses an AI screening method for old-adapted lighting spectrum and computer equipment, relates to the field of old-adapted spectrum screening, and comprises the following steps: collecting step spectrum distribution power data of each independent channel of a target product, wherein the spectrum distribution power data comprises red light spectrum power distribution, green light spectrum power distribution, blue light spectrum power distribution, cold white light spectrum power distribution and warm white light spectrum power distribution; synthesizing spectrum power distribution of different formulas according to the step spectrum distribution power data; calculating spectrum evaluation parameters of different formulas according to the spectrum power distribution of different formulas; comparing the spectrum evaluation parameters of different formulas with preset old-adapted spectrum standards to generate comparison results; marking the spectrum meeting the standards as "effective old-adapted spectrum", and storing the spectrum meeting the standards in a solidified spectrum library; and the application has the advantages of scientifically and reasonably screening old-adapted spectrum.

Description

Technical Field

[0001] This invention relates to the field of spectral screening, and more particularly to an AI screening method and computer device for age-friendly lighting spectra. Background Technology

[0002] Currently, existing age-friendly lighting products on the market have many shortcomings in their spectral design and screening methods, including: (1) Most age-friendly lighting products lack unified and scientific standards for spectral parameters, and have chaotic color temperature coverage (2700K-6500K). They are not designed specifically for the visual physiological characteristics of the elderly, such as yellowing of the lens, pupil constriction, and increased light scattering in the eye, as well as non-visual needs for daytime rhythm regulation. Some products have an excessively high proportion of short-wavelength blue light below 460nm, which can easily exacerbate the risk of retinal damage in the elderly.

[0003] (2) The evaluation dimensions of the light source are incomplete, focusing only on the general color rendering index Ra and ignoring the R9 index, which is crucial for the reproduction of high saturated red. At the same time, the melanopsin ratio (M / P ratio) related to physiological rhythm is not considered, resulting in insufficient high saturated color reproduction and insufficient daytime rhythm stimulation when the elderly see objects. Furthermore, there is no unified parameter calculation formula to support the algorithm's quantitative identification, and the screening results are highly subjective and inconsistent.

[0004] (3) The spectral screening method relies heavily on manual judgment and lacks an intelligent screening system. The spectral compatibility screening of multi-channel dimming lighting products is inefficient and inaccurate, and cannot quickly meet the dual visual and non-visual needs of the elderly.

[0005] Therefore, developing an age-friendly spectral screening model based on AI algorithms and with complete parameter calculation formulas is of great practical significance. Summary of the Invention

[0006] The problem to be solved by this invention is to provide an AI screening method for age-friendly lighting spectrum, which can achieve scientific and reasonable screening of age-friendly lighting spectrum.

[0007] To address the aforementioned technical problems, this invention provides an AI-based screening method for age-friendly lighting spectra, comprising: collecting step-by-step spectral power distribution data for each independent channel of a target product, wherein the spectral power distribution data includes red light spectral power distribution, green light spectral power distribution, blue light spectral power distribution, cool white light spectral power distribution, and warm white light spectral power distribution; synthesizing spectral power distributions for different formulations based on the step-by-step spectral power distribution data; calculating spectral evaluation parameters for different formulations based on the spectral power distributions of the different formulations, wherein the spectral evaluation parameters include color temperature, color rendering index, black-plastic ratio, spectral energy ratio of each band, and luminous efficacy threshold, wherein the color rendering index includes a general color rendering index and a color rendering index for a highly saturated red color sample; comparing the spectral evaluation parameters of different formulations with a preset age-friendly spectral standard to generate a comparison result; marking the spectra that meet the comparison results as "effective age-friendly spectra" and storing the compliant spectra in a fixed spectral library.

[0008] As an improvement to the above scheme, the preset age-friendly spectrum standards include: a color temperature greater than or equal to 3500K and less than or equal to 4500K; a general color rendering index Ra greater than or equal to 95; a color rendering index of greater than or equal to 90 for highly saturated red color samples; a black-to-visual ratio greater than or equal to 0.8; a luminous efficacy threshold greater than or equal to 270 lm / W; a proportion of short-wavelength blue light below 460nm greater than or equal to 5% and less than or equal to 10%; a proportion of cyan light between 460-500nm greater than or equal to 20% and less than or equal to 35%; and a proportion of red light between 590-650nm greater than or equal to 40% and less than or equal to 50%.

[0009] As an improvement to the above scheme, the preset age-friendly spectrum standard is adjusted in special age-friendly scenarios as follows: the melanopsin ratio is greater than or equal to 0.8 and less than or equal to 0.9; the luminous efficacy threshold is greater than or equal to 280 lm / W; and the proportion of red light in the 590-650nm range is greater than or equal to 50% and less than 60%.

[0010] As an improvement to the above scheme, the method for calculating the color temperature based on the spectral power distribution of the different formulations includes: converting the spectral power distribution data into tristimulus values; locating the chromaticity coordinates of the corresponding spectrum based on the tristimulus values; and calculating the color temperature of the spectrum based on the chromaticity coordinates.

[0011] As an improvement to the above scheme, the method for calculating the general colorimetric index based on the spectral power distribution of the different formulations includes: The general color rendering index is calculated using the following formula:

[0012]

[0013] in, This indicates the general color rendering index. The specific color rendering index represents a single standard color sample. Indicates the first and second halves of the test light source and the reference light source. The color difference value of a standard color sample.

[0014] As an improvement to the above scheme, the method for calculating the color rendering index of a highly saturated red sample based on the spectral power distribution of the different formulations includes: The color rendering index of a highly saturated red color sample is calculated using the following formula:

[0015] in, The color rendering index indicates the color rendering index of a highly saturated red color sample. Indicates the first under the test light source and the reference light source The color difference value of a standard color sample.

[0016] As an improvement to the above scheme, the method for calculating the melanopsin ratio based on the spectral power distribution of the different formulations includes: Calculate the melanoplasm ratio using the following formula:

[0017] in, Indicates melanopsin ratio, This represents the spectral power distribution of the light source under test at wavelength λ. This represents the CIE standard melanopsin light response function at wavelength λ. This represents the CIE standard photometric efficiency function at wavelength λ.

[0018] As an improvement to the above scheme, the method for calculating the spectral energy ratio of each band based on the spectral power distribution of the different formulations includes: Calculate using the following formula:

[0019] in, Indicates the band as The proportion, This represents the spectral power distribution of the light source under test at wavelength λ.

[0020] As an improvement to the above scheme, the method for calculating the luminous efficacy threshold based on the spectral power distribution of the different formulations includes: The luminous efficacy threshold is calculated using the following formula:

[0021] in, This indicates the maximum light-emitting efficiency of photopic vision. This represents the CIE standard photometric efficiency function at wavelength λ. This represents the spectral power distribution of the light source under test at wavelength λ. This indicates the luminous efficacy threshold.

[0022] Accordingly, the present invention also provides a computer device, the computer device including a storage device and a processor, the storage device storing a computer program, and the processor executing the computer program to implement the AI ​​screening method for age-friendly lighting spectrum as described in any of the above claims.

[0023] Implementing this invention has the following beneficial effects: By integrating spectral data acquisition, spectral synthesis, parameter quantification calculation, rule matching, and result output, a complete AI screening process has been constructed. This process enables automated and high-throughput screening of the spectra of multi-channel dimming lighting products, significantly improving screening efficiency and accuracy, and overcoming the subjectivity and poor consistency issues of traditional manual screening. At the same time, based on a complete parameter system (including color temperature, color rendering index, black-plastic ratio, band proportion, and luminous efficacy), this method ensures that the screened spectra can fully meet the visual and non-visual needs of the elderly, thus guaranteeing the health, comfort, and functionality of lighting from the source. More preferably, the present invention further enhances the applicability and flexibility of the method by fine-tuning the preset parameter thresholds for specific age-friendly scenarios (such as Alzheimer's care and nighttime awakenings). Specifically, increasing the proportion of red light can enhance visual comfort and potential biological benefits, adjusting the melanopsin ratio range can optimize the rhythm regulation effect to adapt to special physiological states, and increasing the luminous efficacy threshold ensures energy saving and practicality in special scenarios, thereby expanding the application scope of the present invention, enabling it to accurately adapt to diverse age-friendly lighting needs, and enhancing the product's scenario-based adaptability. Attached Figure Description

[0024] Figure 1 This is a flowchart of an embodiment of the AI ​​screening method for age-friendly lighting spectrum according to the present invention; Figure 2 This is a schematic diagram of the preset age-friendly spectrum standard of the AI ​​screening method for age-friendly lighting spectrum of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. It is hereby declared that the directional terms such as up, down, left, right, front, back, inside, and outside used in this text are based solely on the accompanying drawings and are not intended to specifically limit the invention.

[0026] This invention breaks through the traditional paradigm of manual screening of age-friendly spectra and constructs an AI spectrum screening model based on the visual physiological characteristics and non-visual needs of the elderly. By clarifying the calculation formulas of each core lighting parameter, quantitative judgment is achieved. Combined with the intelligent matching of AI algorithms, the efficient and accurate screening of age-friendly spectra is realized, while ensuring the visual health, visual experience and physiological rhythm regulation needs of the elderly.

[0027] like Figure 1 As shown, Figure 1 The flowchart illustrates an embodiment of the AI ​​screening method for age-friendly lighting spectrum according to the present invention, which includes: S1. Collect step spectral distribution power data of each independent channel of the target product; It should be noted that the spectral power distribution data includes the power distribution of red light, green light, blue light, cool white light, and warm white light, i.e. , , , , ; Specifically, for target multi-channel dimming lighting products (this embodiment takes RGBCW five-channel LED products as an example), data acquisition needs to be performed in a dark room environment using a high-precision spectroradiometer or spectrophotometer; When acquiring the spectral power distribution data, PWM dimming is supported for each independent channel with a dimming accuracy of 1%. During acquisition, each channel needs to be adjusted in 1% increments from 0% to 100% brightness (or from minimum drive current to rated drive current), and the spectral power distribution data of that channel in the visible light wavelength range of 380nm-780nm is recorded at each step point. The collected spectral distribution power data is used to construct a basic spectral database and imported into the data storage module of the AI ​​algorithm. This database fully records the spectral "fingerprint" of each channel at different brightness levels, providing a basic unit for subsequent spectral synthesis.

[0028] S2. Synthesize the spectral power distribution of different formulations based on the stepped spectral power distribution data; Based on the principle of optical superposition, when multiple channels emit light simultaneously, the total spectral power distribution synthesized can be approximated as the linear sum of the spectral power distributions of each channel at the current brightness ratio; specifically, the spectral power distribution of different formulations can be achieved according to the following formula:

[0029] in: , , , , These represent the brightness scaling factors for the red, green, blue, cool white, and warm white channels, respectively, with values ​​ranging from 0 to 1; Indicates wavelength; Indicates wavelength as The total spectral power distribution; , , , , These represent the independent spectral power distributions of the red, green, and blue LED channels, as well as the independent spectral power distributions of the cool white and warm white LED channels.

[0030] This invention can generate a massive number of different brightness ratio combinations within a huge parameter space (five-dimensional space in this embodiment) using AI algorithms, pseudo-random number generation, grid search, or more optimized algorithms (such as genetic algorithms). , , , , For each generated combination, the algorithm calls the above formula to synthesize a corresponding total spectrum. (λ).

[0031] S3. Calculate the spectral evaluation parameters of different formulations based on the spectral power distribution of the different formulations; It should be noted that the spectral evaluation parameters include color temperature, color rendering index, melanopsin ratio, spectral energy ratio of each band, and luminous efficacy threshold. This step is the core step of the invention. It involves calculating a complete set of age-appropriate spectral evaluation parameters from the spectra of the different formulations synthesized in step S2 using a built-in standardized formula. The spectral evaluation parameters will be described below: I. Color Temperature Calculating color temperature is not a single formula, but a multi-step derivation process. The fundamental principle is: by measuring the spectral power distribution of the light source, calculating the coordinate position of the light source on the chromaticity diagram, and then finding the blackbody radiation trajectory closest to that coordinate point, the temperature corresponding to this blackbody trajectory is the correlated color temperature of the light source; the specific steps are as follows: (1) Convert the spectral power distribution data into tristimulus values;

[0032]

[0033]

[0034] in: This represents the spectral power distribution of the light source under test, i.e., the light intensity at each wavelength λ. , , This represents the CIE 1931 standard colorimetric observer color matching function; Indicates the wavelength interval used in the calculation; This represents the integration of all visible light wavelengths (typically from 380 nm to 780 nm). The tristimulus value representing the human eye's tricolor response (red); The tristimulus value representing the human eye's three-color response (green); The tristimulus value represents the tricolor response (blue) of the human eye.

[0035] (2) Locate the chromaticity coordinates of the corresponding spectrum based on the tristimulus values; The tristimulus values ​​are converted into chromaticity coordinates (x, y) that are independent of brightness and purely represent color, so as to be located on a two-dimensional chromaticity map;

[0036]

[0037] in: The tristimulus value representing the human eye's tricolor response (red); The tristimulus value representing the human eye's three-color response (green); The tristimulus value representing the human eye's tricolor response (blue); and This indicates the coordinate position of the color on the CIE1931 chromaticity diagram.

[0038] (3) Calculate the color temperature of the spectrum based on the chromaticity coordinates.

[0039] Calculating the color temperature of the spectrum based on the chromaticity coordinates involves converting the chromaticity coordinates into color temperature values ​​in Kelvin. Since the blackbody radiation trajectory is a curve on the chromaticity diagram, directly calculating the color temperature from the coordinate points is very complex. Therefore, empirical formulas are usually used for approximate calculations. This invention uses the following empirical formula, as detailed below:

[0040]

[0041] in: and This indicates the coordinate position of the color on the CIE 1931 chromaticity diagram; This represents an intermediate variable, which is related to the direction of the "isotherm" from the chromaticity coordinate point to the blackbody locus; Indicates color temperature.

[0042] II. Color rendering index In this invention, the color rendering index includes a general color rendering index and a color rendering index for highly saturated red samples. The general color rendering index ensures that object colors appear highly realistic and natural. As the lens of an elderly person yellows, it absorbs some blue-violet light like a filter, leading to a yellowish and distorted color perception. A high Ra light source can compensate for this color difference, enabling the elderly to more accurately distinguish the color of medicines, the freshness of food, and facial color, reducing the risk of misjudgment. Meanwhile, the color rendering index of highly saturated red samples ensures that red objects are vibrant and eye-catching, greatly improving the recognition of important information and ensuring safety and health.

[0043] The general formula for calculating the color rendering index is as follows:

[0044]

[0045] in: Indicates the general color rendering index; A specific color rendering index that represents a single standard color sample; Indicates the first and second halves of the test light source and the reference light source. The color difference value of a standard color sample.

[0046] The color rendering index of the highly saturated red color sample is shown below:

[0047] in: The color rendering index indicates the color development index of a highly saturated red color sample; Indicates the first under the test light source and the reference light source The color difference value of a standard color sample.

[0048] III. Melanocyte Ratio Melanops are special photoreceptor cells in the retina, primarily responsible for receiving light signals and transmitting them to the brain's biological clock, regulating melatonin secretion, and thus affecting the sleep-wake cycle. Older adults often experience circadian rhythm disruptions and poor sleep quality. In these cases, a higher melanopsin ratio (≥0.8) indicates that light sources effectively stimulate melanops, providing strong non-visual biological stimulation during the day, helping to suppress melatonin, maintain daytime alertness and wakefulness, and in the evening, aiding in a smooth transition to nighttime rest, thereby improving sleep quality and stabilizing mood.

[0049] The melanopsin ratio is obtained by comparing the integral light source spectrum with the melanopsin light response function and photopic vision efficiency function. It is used to evaluate the potential impact of the light source on physiological rhythms. The specific calculation is shown below:

[0050] in: Indicates melanopsin ratio; This represents the spectral power distribution of the light source under test at wavelength λ; This represents the CIE standard melanopsin light response function at wavelength λ. This represents the CIE standard photometric efficiency function at wavelength λ.

[0051] IV. Spectral Energy Ratio of Each Band The spectral energy percentage of each band is obtained by calculating the integral ratio of the spectral power in a specific band to the total visible light spectral power. This invention pays particular attention to the percentage of short-wavelength blue light below 460nm, the percentage of cyan light in the 460-500nm range, and the percentage of red light in the 590-650nm range. The short-wavelength blue light below 460nm is one of the risk factors for age-related macular degeneration (AMD). By limiting its proportion to a low but non-zero level (not less than 5% to ensure color rendering), the risk of potential photochemical damage to the retina is minimized. The 460-500nm cyan light is the region with the highest sensitivity of melanopsin and is an "effective ingredient" for regulating physiological rhythms. By setting a high lower limit (≥20%), the light source is ensured to have sufficient rhythmic stimulation ability, while an upper limit (≤35%) is set to prevent excessive cyan light from causing visual discomfort or color distortion. The 590-650nm red light may have a positive effect on improving mitochondrial function and promoting cellular energy metabolism, and has potential health benefits for the elderly population. At the same time, the high proportion of red light ensures the comfort of lighting and possible positive biological regulation.

[0052] Specifically, the formula for calculating the spectral energy percentage of each band is as follows:

[0053] in: Indicates the band as The proportion; This represents the spectral power distribution of the light source under test at wavelength λ.

[0054] V. Luminous efficacy threshold Considering that the elderly often need longer periods of brighter lighting, high luminous efficacy means that more effective luminous flux can be generated while consuming the same amount of electricity. A higher luminous efficacy threshold not only meets the environmental protection requirements of green energy saving, but also reduces the burden of electricity bills. At the same time, it avoids the use of products with excessive power consumption in pursuit of a healthy spectrum, which can lead to problems such as high heat generation and shortened lifespan, thus ensuring the economy and sustainability of the "age-friendly lighting" solution.

[0055] The luminous efficacy threshold is calculated based on the photopic luminous efficacy function, which characterizes the energy efficiency of the light source. The specific calculation formula is shown below:

[0056] in: This indicates the maximum light-emitting efficiency of the light-emitting vision; Represents the CIE standard photometric efficiency function; This represents the spectral power distribution of the light source under test at wavelength λ; This indicates the luminous efficacy threshold.

[0057] S4. Compare the spectral evaluation parameters of different formulations with the preset age-appropriate spectral standard; Based on the visual physiological characteristics of the elderly, such as yellowing of the lens and pupil constriction, this invention constructs a multi-dimensional and quantitative preset age-friendly spectrum standard. The preset age-friendly spectrum standard transforms the special visual physiological defects of the elderly (such as yellowing of the lens and pupil constriction) and dual visual and non-visual needs (such as high-fidelity color reproduction and daytime rhythm stimulation) into hard thresholds that can be accurately executed by AI algorithms, thereby ensuring that each selected "effective age-friendly spectrum" can effectively avoid the risk of short-wave blue light damage and improve visual comfort and color accuracy. Specifically, such as Figure 2 As shown, the preset age-friendly spectral standard includes: The color temperature must be greater than or equal to 3500K and less than or equal to 4500K; Generally, the color rendering index Ra is greater than or equal to 95; The color rendering index of a highly saturated red sample is greater than or equal to 90; Melanoplasm ratio greater than or equal to 0.8; The luminous efficacy threshold is greater than or equal to 270 lm / W; The proportion of short-wavelength blue light below 460nm must be greater than or equal to 5% and less than or equal to 10%. The proportion of cyan light in the 460-500nm range must be greater than or equal to 20% and less than or equal to 35%. The proportion of red light in the 590-650nm range must be greater than or equal to 40% and less than or equal to 50%.

[0058] Furthermore, for specific age-appropriate scenarios, a dedicated spectral formula library selected after fine-tuning parameter thresholds can be invoked. The adjustment of the preset age-appropriate spectral standards includes: Melanoplasm ratio greater than or equal to 0.8 and less than or equal to 0.9; The luminous efficacy threshold is greater than or equal to 280 lm / W; The proportion of red light in the 590-650nm range must be greater than or equal to 50% and less than 60%.

[0059] S5. Mark the spectra that meet the comparison results as "effective age-friendly spectra" and store the qualified spectra in the solidified spectrum library.

[0060] All spectral formulations that passed the rigorous screening in step S4 were marked as "effective age-friendly spectra" by the AI ​​model. The channel brightness ratios of these formulations ( , , , , The calculated ideal ratio values ​​are stored in the final effective spectrum library; then, the current or switching time of each LED channel is precisely controlled by techniques such as PWM (pulse width modulation) to reproduce a spectrum in the physical world that is highly consistent with the calculation results. Even better, the system can solidify these effective ratio combinations into different lighting modes (such as "daytime active mode" and "comfortable reading mode"). When the user selects the corresponding mode, the product's control system will automatically set the PWM dimming signal of each channel to the ratio corresponding to that mode, thereby stably outputting healthy and comfortable light suitable for the elderly.

[0061] Furthermore, for mass-produced products, by solidifying these successful "formulas," it can be ensured that each lamp can output almost identical optical characteristics (color temperature, color rendering index, etc.) in the same mode, thus guaranteeing the stability of product quality and the consistency of user experience.

[0062] During operation, the AI ​​screening method for age-friendly lighting spectra first collects step spectral distribution power data of each independent channel of the target multi-channel dimming product (such as an RGBCW five-channel LED) in a dark room environment using a high-precision spectroradiometer or spectrophotometer to construct a basic spectral database. Subsequently, the AI ​​algorithm generates a massive number of different channel brightness ratio combinations through pseudo-random number generation, grid search, or optimization algorithms (such as genetic algorithms), and linearly synthesizes the total spectral power distribution corresponding to each combination based on the principle of optical superposition. Next, for each synthesized spectrum, the built-in standardized formula is automatically called to calculate a complete set of spectral evaluation parameters, including color temperature, color rendering index (Ra and R9), black-plastic ratio, spectral energy ratio of each band, and luminous efficacy threshold. Then, these parameters are compared one by one with the preset age-friendly spectrum parameter thresholds to screen out spectral formulas that meet all parameters. Finally, the qualified formulas are marked as "effective age-friendly spectra," and the corresponding ratio is solidified into the product control system through PWM dimming technology to achieve stable output of different lighting modes such as "daytime active mode," ensuring consistent optical characteristics of each lamp.

[0063] Accordingly, the present invention also provides a computer device, the computer device including a storage device and a processor, the storage device storing a computer program, and the processor executing the computer program to implement the AI ​​screening method for age-friendly lighting spectrum as described above.

[0064] In summary, this invention achieves the quantification and intelligent screening of age-friendly lighting spectra by constructing an AI spectral screening model based on the visual physiological characteristics and non-visual needs of the elderly. This method solves the problems of missing parameter dimensions, strong subjectivity in screening, and low efficiency in existing technologies. Furthermore, through complete parameter calculation formulas and AI algorithms, it ensures the accuracy, consistency, and efficiency of the screening results. At the same time, based on the visual physiological characteristics of the elderly, such as yellowing of the lens and pupil constriction, the thresholds of core parameters such as the proportion of each wavelength, color temperature, and light effect are precisely limited. This not only reduces the risk of retinal damage from short-wavelength blue light, but also improves the high saturation color reproduction and rhythmic stimulation effect, thus comprehensively meeting the physiological and visual needs of the elderly.

[0065] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. An AI screening method for age-friendly lighting spectrum, characterized in that, include: Collect step spectral power distribution data for each independent channel of the target product. The spectral power distribution data includes red light spectral power distribution, green light spectral power distribution, blue light spectral power distribution, cool white light spectral power distribution, and warm white light spectral power distribution. Based on the stepped spectral power distribution data, spectral power distributions of different formulations are synthesized; The spectral evaluation parameters of different formulations are calculated based on the spectral power distribution of the different formulations. The spectral evaluation parameters include color temperature, color rendering index, melanopsin ratio, spectral energy ratio of each band and luminous efficacy threshold. The color rendering index includes the general color rendering index and the color rendering index of the high-saturation red color sample. The spectral evaluation parameters of different formulations are compared with preset age-appropriate spectral standards to generate comparison results; The spectra that meet the comparison results are marked as "effective age-friendly spectra" and the qualified spectra are stored in the curing spectrum library.

2. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, The preset age-appropriate spectral standards include: The color temperature must be greater than or equal to 3500K and less than or equal to 4500K; Generally, the color rendering index Ra is greater than or equal to 95; The color rendering index of a highly saturated red sample is greater than or equal to 90; Melanoplasm ratio greater than or equal to 0.8; The luminous efficacy threshold is greater than or equal to 270 lm / W; The proportion of short-wavelength blue light below 460nm must be greater than or equal to 5% and less than or equal to 10%. The proportion of cyan light in the 460-500nm range must be greater than or equal to 20% and less than or equal to 35%. The proportion of red light in the 590-650nm range must be greater than or equal to 40% and less than or equal to 50%.

3. The AI ​​screening method for age-friendly lighting spectrum as described in claim 2, characterized in that, In specific age-friendly scenarios, the preset age-friendly spectral standards include: Melanoplasm ratio greater than or equal to 0.8 and less than or equal to 0.9; The luminous efficacy threshold is greater than or equal to 280 lm / W; The proportion of red light in the 590-650nm range must be greater than or equal to 50% and less than 60%.

4. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, The method for calculating color temperature based on the spectral power distribution of the different formulations includes: The spectral power distribution data is converted into tristimulus values; The chromaticity coordinates of the corresponding spectrum are located based on the tristimulus values; The color temperature of the spectrum is calculated based on the chromaticity coordinates.

5. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, Methods for calculating the general colorimetric index based on the spectral power distribution of the different formulations include: The general color rendering index is calculated using the following formula: in, This indicates the general color rendering index. The specific color rendering index represents a single standard color sample. Indicates the first and second halves of the test light source and the reference light source. The color difference value of a standard color sample.

6. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, Methods for calculating the color rendering index of highly saturated red samples based on the spectral power distribution of the different formulations include: The color rendering index of a highly saturated red color sample is calculated using the following formula: in, The color rendering index indicates the color rendering index of a highly saturated red color sample. Indicates the first under the test light source and the reference light source The color difference value of a standard color sample.

7. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, The method for calculating the melanopsin ratio based on the spectral power distribution of the different formulations includes: Calculate the melanoplasm ratio using the following formula: in, Indicates melanopsin ratio, This represents the spectral power distribution of the light source under test at wavelength λ. This represents the CIE standard melanopsin light response function at wavelength λ. This represents the CIE standard photometric efficiency function at wavelength λ.

8. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, The method for calculating the spectral energy percentage of each band based on the spectral power distribution of the different formulations includes: Calculate using the following formula: in, Indicates the band as The proportion, This represents the spectral power distribution of the light source under test at wavelength λ.

9. The AI ​​screening method for age-friendly lighting spectrum as described in claim 1, characterized in that, The method for calculating the luminous efficacy threshold based on the spectral power distribution of the different formulations includes: The luminous efficacy threshold is calculated using the following formula: in, This indicates the maximum light-emitting efficiency of photopic vision. This represents the CIE standard photometric efficiency function at wavelength λ. This represents the spectral power distribution of the light source under test at wavelength λ. This indicates the luminous efficacy threshold.

10. A computer device, the computer device comprising a storage unit and a processor, characterized in that, The storage device stores a computer program, and when the processor executes the computer program, it implements an AI screening method for age-friendly lighting spectrum according to any one of claims 1 to 8.

Images