CN118130996A - LED temperature optimization method and system based on multispectral detection - Google Patents
LED temperature optimization method and system based on multispectral detection Download PDFInfo
- Publication number
- CN118130996A CN118130996A CN202410533460.4A CN202410533460A CN118130996A CN 118130996 A CN118130996 A CN 118130996A CN 202410533460 A CN202410533460 A CN 202410533460A CN 118130996 A CN118130996 A CN 118130996A
- Authority
- CN
- China
- Prior art keywords
- led
- temperature
- luminous intensity
- image
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 238000005457 optimization Methods 0.000 title claims abstract description 33
- 239000013598 vector Substances 0.000 claims abstract description 111
- 230000003595 spectral effect Effects 0.000 claims abstract description 69
- 238000004458 analytical method Methods 0.000 claims abstract description 41
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000004020 luminiscence type Methods 0.000 claims abstract description 14
- 238000010276 construction Methods 0.000 claims abstract description 11
- 230000008859 change Effects 0.000 claims description 67
- 238000001228 spectrum Methods 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000011176 pooling Methods 0.000 claims description 18
- 238000009826 distribution Methods 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000012417 linear regression Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 6
- 238000010191 image analysis Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 21
- 238000012545 processing Methods 0.000 description 13
- 238000012544 monitoring process Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 238000013461 design Methods 0.000 description 7
- 230000017525 heat dissipation Effects 0.000 description 6
- 230000007613 environmental effect Effects 0.000 description 5
- 238000013021 overheating Methods 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005265 energy consumption Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000013178 mathematical model Methods 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000005297 material degradation process Methods 0.000 description 3
- 230000002035 prolonged effect Effects 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000001816 cooling Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000004643 material aging Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000020169 heat generation Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003707 image sharpening Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000004043 responsiveness Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
- Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection
Landscapes
- Testing Of Optical Devices Or Fibers (AREA)
- Radiation Pyrometers (AREA)
Abstract
The application relates to the technical field of LED detection and discloses an LED temperature optimization method and system based on multispectral detection. The method comprises the following steps: performing luminescence test and multispectral image acquisition on a target LED based on the first LED power output parameters to obtain a multiband spectral image and an infrared spectral image; respectively carrying out LED variation analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector; the luminous intensity and temperature relation vector is input into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output to obtain a second LED power output parameter corresponding to the target LED.
Description
Technical Field
The application relates to the technical field of LED detection, in particular to an LED temperature optimization method and system based on multispectral detection.
Background
LEDs are widely used due to their high efficiency, long life and environmental characteristics. With the progress of technology, the application scene of the LED is expanded from indoor and outdoor illumination to automobiles, advertisement display and high-end display equipment. However, the LED generates a lot of heat during the use process, and an excessively high temperature not only affects the light efficiency and the lifetime of the LED, but also may cause degradation of the light color quality.
Currently, temperature management and optimization of LEDs is mainly dependent on conventional heat dissipation designs, such as using physical methods of heat sinks, fans, etc. to reduce the operating temperature of the LEDs. While these methods are somewhat effective, they add bulk and cost to the product, and do not allow for precise control of the temperature state of each LED unit, which is particularly disadvantageous for applications requiring high precision light color control. In addition, the conventional heat dissipation technology cannot respond to the change of the operating condition of the LED in real time, so that the heat dissipation effect of the prior art is poor.
Disclosure of Invention
The application provides an LED temperature optimization method and system based on multispectral detection, which are used for analyzing and calculating the performance change in the LED operation process by adopting the multispectral detection technology so as to improve the control accuracy of the LED power output.
In a first aspect, the present application provides a method for optimizing LED temperature based on multispectral detection, the method for optimizing LED temperature based on multispectral detection comprising:
Performing luminescence test and multispectral image acquisition on a target LED based on the first LED power output parameters to obtain a multiband spectral image and an infrared spectral image;
Respectively carrying out LED variation analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector;
And inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output, so as to obtain a second LED power output parameter corresponding to the target LED.
In a second aspect, the present application provides an LED temperature optimization apparatus based on multispectral detection, the LED temperature optimization apparatus based on multispectral detection comprising:
the acquisition module is used for carrying out luminescence test and multispectral image acquisition on the target LED based on the first LED power output parameter to obtain a multiband spectral image and an infrared spectral image;
the construction module is used for respectively carrying out LED change analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector;
and the analysis module is used for inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output, so as to obtain a second LED power output parameter corresponding to the target LED.
A third aspect of the present application provides an LED temperature optimization apparatus based on multispectral detection, comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the multispectral detection-based LED temperature optimization apparatus to perform the multispectral detection-based LED temperature optimization method described above.
A fourth aspect of the application provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the above-described method of LED temperature optimization based on multi-spectral detection.
According to the technical scheme provided by the application, the temperature change of the LED is monitored in real time by using the infrared spectrogram image, so that the temperature abnormality in the LED light emitting process can be detected and responded in real time. This real-time monitoring capability ensures the accuracy of temperature management, thereby avoiding the drop in light efficiency and the reduction in lifetime caused by excessive temperatures. By analyzing the multiband spectral image and the infrared spectral image, the temperature is monitored, and the power output of the LED can be adjusted in real time. The dynamic adjustment can optimize power output according to real-time light emitting and temperature conditions, further improve energy efficiency and reduce energy consumption. Analysis of the multiband spectral image is helpful for monitoring the light color performance of the LED, and consistency and stability of light color output are ensured. The temperature optimization not only affects the service life and the energy efficiency, but also directly affects the color temperature and the light color purity, so that the light color quality of a product can be obviously improved by adjusting the light color output through optimizing the temperature parameter. As the temperature of the LED is effectively controlled, material degradation and light attenuation caused by overheating are avoided, and the service life of the LED is remarkably prolonged. The application adopts the multispectral detection technology to analyze and calculate the performance change in the LED operation process, thereby improving the control accuracy of the LED power output.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an embodiment of an LED temperature optimization method based on multispectral detection in an embodiment of the application;
FIG. 2 is a schematic diagram of an embodiment of an LED temperature optimization device based on multispectral detection in an embodiment of the application.
Detailed Description
The embodiment of the application provides an LED temperature optimization method and system based on multispectral detection. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present application is described below, referring to fig. 1, and an embodiment of an LED temperature optimization method based on multispectral detection in the embodiment of the present application includes:
step S101, carrying out luminescence test and multispectral image acquisition on a target LED based on a first LED power output parameter to obtain a multiband spectral image and an infrared spectral image;
it is to be understood that the implementation subject of the present application may be an LED temperature optimizing device based on multispectral detection, and may also be a terminal or a server, which is not limited herein. The embodiment of the application is described by taking a server as an execution main body as an example.
Specifically, power output parameter initialization setting is conducted on the target LED, and corresponding first LED power output parameters are generated. The power output parameters of the LED are configured according to the characteristics and expected use conditions of the LED, and the parameters define the basic working state of the LED in the luminescence test, so that the LED can be operated under the premise of safety and efficiency. And carrying out luminescence test on the target LED according to the first LED power output parameter, monitoring the luminescence performance and stability of the LED in the test process, and ensuring that the LED meets the preset light output and efficiency requirements in actual operation. And acquiring multiband spectral images of the target LEDs through a plurality of preset visible light wave bands and special spectral cameras to obtain visible light initial spectral images of each visible light wave band. The visible light band typically includes the dominant wavelengths in the light spectrum of an LED, and the spectral image of each band reflects the spectral characteristics and luminous efficiency of that band. The initial spectral images acquired record the luminescence characteristics of these bands under standard operating conditions. And carrying out image resolution enhancement processing on the initial spectrum image of each visible light wave band. The enhanced image is clearer and the spectrum data is more accurate through an image processing algorithm, such as an interpolation algorithm and an image sharpening technology. And generating a multiband spectral image according to the visible light enhanced spectral image of each visible light band. And the interaction and influence among different wave bands are comprehensively considered, so that comprehensive data support is provided for subsequent luminous performance analysis and temperature control. In order to comprehensively evaluate the temperature characteristics of the LEDs, infrared spectrogram images of the target LEDs are acquired through a preset infrared light wave band and a special infrared camera. The infrared spectral image shows the thermal distribution and temperature variation of the LED during the light emission process, which is helpful for understanding and controlling the thermal management of the LED.
Step S102, respectively carrying out LED variation analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector;
Specifically, the multiband spectral image is analyzed, the change vector of the luminous intensity of the LED is calculated, and the key luminous intensity characteristic value set is extracted through curve characteristic calculation. Significant changes in the spectral image, which may represent an increase or decrease in the luminous performance of the LED, are determined by image processing techniques and data analysis methods, for example, using edge detection and peak recognition techniques. Meanwhile, the infrared spectrogram image is subjected to LED temperature change detection and curve characteristic calculation. By monitoring and calculating the temperature distribution in the infrared image, the temperature fluctuation of the LED in long-time operation is detected, and a characteristic value set of the temperature is calculated. These characteristic values are typically obtained by analyzing statistical data of the temperature distribution, such as average temperature, maximum temperature, temperature gradient, etc. The curve characteristic calculation of the temperature profile is used to help the system understand the thermal stability and thermal response characteristics of the LED at different power outputs. And extracting the characteristic value and creating the characteristic vector by the characteristic value set of the luminous intensity and the characteristic value set of the temperature. The most representative feature vector is extracted from a large number of feature values by a multivariate statistical analysis method such as principal component analysis or a machine learning algorithm such as a support vector machine. Finally, a comprehensive model is created according to the feature vector, the model can describe the dynamic relation between the luminous intensity and the temperature of the LED, and information is provided for temperature management of the LED.
And performing difference analysis on the multiband spectral images to obtain multiband difference images, wherein the multiband difference images highlight the spectral changes of the same LED at different time points or different power settings. The differential image analysis effectively captures minor changes due to LED aging, temperature changes, or other external factors, which may be ignored in conventional spectral image analysis. And (3) identifying the distribution condition of the luminous intensity in different spectral bands by carrying out change vector analysis on the multiband difference image. And the change vector analysis obtains the change trend of the LED performance by calculating the light intensity difference of each wave band image. And carrying out LED luminous intensity recognition on the multiband difference image according to the obtained luminous intensity distribution, and extracting luminous intensity data of the LEDs from the image data by using an image recognition and data extraction technology. And performing curve fitting on the LED luminous intensity data, and generating an LED luminous intensity change curve through a mathematical model such as polynomial fitting or exponential fitting and the like. And carrying out change rate integral calculation on the LED luminous intensity change curve by a difference algorithm, wherein the algorithm obtains a plurality of change rate integral values by calculating the change rate of each point on the curve and carrying out integral processing on the rates. The integrated value provides important information about how fast and how wide the LED luminous intensity changes and the range of variation, and is key data for evaluating LED performance stability and optimizing parameter settings. The plurality of change rate integrated values are subjected to time series correlation and set conversion, and these data are converted into a characteristic value set of luminous intensity by subjecting the integrated values to time series analysis and statistical processing.
And (3) extracting pixel temperature values of the infrared spectrogram images by analyzing the infrared radiation intensity of each pixel in the images to obtain a plurality of corresponding initial pixel temperatures, wherein the infrared radiation intensity can directly reflect the temperature of the corresponding pixel. And calibrating the LED area on the initial pixel temperature values to obtain target pixel temperature values of the LED area. Pixels in the infrared image representing the LED areas are identified and distinguished from the remaining background pixels. And carrying out average value operation on a plurality of target pixel temperature values of the LED area to obtain LED temperature data of the infrared spectrogram image. The average temperature data reflects the average heat performance of the LEDs in the current working state, and provides a basis for evaluating and comparing the temperature performance of different working conditions or different LEDs. Based on the averaged LED temperature data, a curve fitting operation is performed to construct an LED temperature variation curve. Curve fitting typically uses a mathematical model such as a polynomial or exponential function to model the trend of temperature data, exhibit temperature behavior over time or other variable changes, and can be used to predict future temperature changes. And identifying curve fluctuation characteristic points of the LED temperature change curve to obtain a plurality of temperature curve fluctuation characteristic points. These fluctuating feature points may represent critical turning points of temperature variation, such as rapid heating or cooling zones, which are typically associated with certain specific conditions or potential problems in the operation of the LED. Identifying these fluctuating feature points helps to understand the thermal behavior pattern of the LEDs and to formulate efficient temperature management strategies and improve LED design.
And calculating the mean value and standard deviation of the luminous intensity characteristic value set and the temperature characteristic value set. The mean provides the average performance for each feature set, while the standard deviation describes the degree of dispersion of the data points around the mean. From these statistics, a coefficient of variation, i.e., the ratio of standard deviation to mean, is calculated, providing a quantitative indicator of the stability and variability of each feature set. The variation coefficient of the luminous intensity and the variation coefficient of the temperature are important indexes for measuring the consistency of respective data sets, and a lower variation coefficient means that the data are more concentrated, and conversely, the data are more scattered. And setting a weight coefficient of the corresponding characteristic value set according to the variation coefficient, wherein the weight coefficient determines the importance and influence of each characteristic value in the subsequent analysis. For example, if the coefficient of variation of the luminous intensity is high, the weight may need to be reduced to avoid excessive influence of abnormal values on the overall analysis; conversely, if the coefficient of variation is low, the corresponding weight may be increased to emphasize the stability of the feature. And performing feature coding mapping and vector conversion on the luminous intensity feature value set and the temperature feature value set by using the set weight coefficients to obtain luminous intensity feature vectors and temperature feature vectors. The original eigenvalues are converted into a format more suitable for mathematical processing and machine learning models. Vector splicing is carried out on the luminous intensity characteristic vector and the temperature characteristic vector, and a luminous intensity and temperature relation vector is obtained. The vector contains the comprehensive information of the light output characteristic and the temperature characteristic of the LED, is a multi-dimensional data structure and can intuitively represent the correlation between the luminous intensity and the temperature.
And step S103, inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output, so as to obtain a second LED power output parameter corresponding to the target LED.
Specifically, the relation vector of the luminous intensity and the temperature is input into a preset LED luminous performance detection model. The model employs a deep learning framework that includes multiple convolutional pooling layers, an attention mechanism layer, and a linear regression function that work together to predict and optimize the power output of the LED. And carrying out convolution characteristic operation on the luminous intensity and temperature relation vector through a plurality of convolution pooling layers, extracting important characteristics of input data through a plurality of filters by the convolution layer, and capturing different characteristics of the data by each filter to generate a plurality of characteristic diagrams. The pooling layer downsamples the multiple feature maps to reduce data dimensionality and improve processing efficiency while preserving the most critical information. The convolution and pooling operations can deeply mine complex relationships between luminous intensity and temperature data and convert these relationships into a plurality of output convolution feature vectors. And inputting the output convolution feature vector of each convolution pooling layer into an attention mechanism layer for attention mechanism weighted fusion. The attention mechanism generates attention fusion feature vectors by distributing different weights to different features and weighting and fusing the feature vectors, so that the model can adaptively and mainly process the features which are most critical to the light emitting performance prediction, and the prediction accuracy and the model responsiveness are improved. And inputting the attention fusion feature vector into a linear regression function to calculate the LED luminous performance value. By linear regression analysis, the model can estimate the optimal luminous performance of the LED according to the processed feature vector, and output an LED luminous performance predicted value which represents the ideal power output of the LED under the current temperature and luminous intensity conditions. And carrying out dynamic compensation analysis according to the predicted value of the luminous performance of the LED. And adjusting the original first LED power output parameter to generate a new and optimized second LED power output parameter. Dynamic compensation takes into account the actual luminous efficiency and expected lifetime of the LED, ensuring that the LED operates within optimal energy consumption efficiency and temperature control ranges without sacrificing performance.
According to the embodiment of the application, the temperature change of the LED is monitored in real time by using the infrared spectrogram image, so that the temperature abnormality in the LED light emitting process can be detected and responded in real time. This real-time monitoring capability ensures the accuracy of temperature management, thereby avoiding the drop in light efficiency and the reduction in lifetime caused by excessive temperatures. By analyzing the multiband spectral image and the infrared spectral image, the temperature is monitored, and the power output of the LED can be adjusted in real time. The dynamic adjustment can optimize power output according to real-time light emitting and temperature conditions, further improve energy efficiency and reduce energy consumption. Analysis of the multiband spectral image is helpful for monitoring the light color performance of the LED, and consistency and stability of light color output are ensured. The temperature optimization not only affects the service life and the energy efficiency, but also directly affects the color temperature and the light color purity, so that the light color quality of a product can be obviously improved by adjusting the light color output through optimizing the temperature parameter. As the temperature of the LED is effectively controlled, material degradation and light attenuation caused by overheating are avoided, and the service life of the LED is remarkably prolonged. The application adopts the multispectral detection technology to analyze and calculate the performance change in the LED operation process, thereby improving the control accuracy of the LED power output.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Initializing and setting power output parameters of a target LED to generate corresponding first LED power output parameters;
(2) Performing a lighting test on the target LED according to the first LED power output parameter;
(3) The method comprises the steps that multiband spectrum image acquisition is conducted on a target LED through a plurality of preset visible light wave bands and spectrum cameras, and a visible light initial spectrum image of each visible light wave band is obtained;
(4) The method comprises the steps of carrying out image resolution enhancement on a visible light initial spectrum image of each visible light wave band to obtain a visible light enhanced spectrum image of each visible light wave band, and generating a multiband spectrum image according to the visible light enhanced spectrum image of each visible light wave band;
(5) And acquiring an infrared spectrogram image of the target LED through a preset infrared light wave band and an infrared camera to obtain an infrared spectrogram image of the infrared light wave band.
Specifically, power output parameter initialization setting is conducted on the target LED, and corresponding first LED power output parameters are generated. The initial power output parameters of an LED are set according to its design specifications and intended application, for example, assuming that an LED is designed to be manufactured for high efficiency illumination, its initial power setting may be set at a medium level to balance luminous efficiency and heat generation. And carrying out luminescence test on the target LED according to the first LED power output parameter, and monitoring the luminescence behavior of the LED under the set power and any possible abnormality by using special equipment. And acquiring spectrum images of the target LEDs through a plurality of preset visible light wave bands and special spectrum cameras. The camera is tuned to capture spectral data at different wavelengths from the ultraviolet to the near infrared. For example, for a multi-color LED lamp, multiple filters may be used to capture spectral images of blue, green, yellow, and red light, respectively. A visible light initial spectral image is recorded for each band, which images provide preliminary visual information about the fundamental emission characteristics of the LED at the respective visible light band. In order to improve the quality and usability of the spectrum data, the image resolution enhancement processing is carried out on the initial spectrum image of each visible light wave band, so as to obtain the visible light enhancement spectrum image of each visible light wave band. Digital image processing techniques such as interpolation, sharpening, and denoising are used to improve image quality so that spectral lines in the image are more sharp and accurate. And generating a multiband spectrum image according to the visible light enhanced spectrum image of each visible light wave band, wherein the image fuses the data of all the key wave bands. And acquiring an infrared spectrogram image of the target LED through a preset infrared light wave band and an infrared camera to obtain an infrared spectrogram image of the infrared light wave band. The infrared spectrogram image is mainly used for capturing the thermal characteristics of the LED when the LED emits light, and is beneficial to monitoring and managing the thermal efficiency of the LED. For example, in one test, if the LED is found to exhibit an abnormal rise in temperature in the infrared image after continued operation, this may indicate that the heat dissipation design of the LED or its power output parameters need to be adjusted. The infrared spectral image provides visual information about the thermal distribution of the LED in operation, helping to optimize the temperature management strategy of the LED and improve its overall performance.
In a specific embodiment, the process of executing step S102 may specifically include the following steps:
(1) Carrying out LED luminous intensity change vector analysis and curve characteristic calculation on the multiband spectral image to obtain a luminous intensity characteristic value set;
(2) Carrying out LED temperature change detection and curve characteristic calculation on the infrared spectrogram image to obtain a temperature characteristic value set;
(3) And extracting characteristic values and creating characteristic vectors for the characteristic value set of the luminous intensity and the characteristic value set of the temperature to obtain a relation vector of the luminous intensity and the temperature.
Specifically, the multi-band spectrum image is subjected to LED luminous intensity change vector analysis, and the light intensity change of each band is calculated by comparing spectrum images of the same wavelength at different time points or under different environmental conditions. And calculating curve characteristics according to the change vector, and converting the data of the light intensity change into a specific mathematical expression through mathematical modeling such as polynomial fitting or Fourier transformation to describe the dynamic change of the LED light output along with time. For example, if an LED shows a tendency for the blue light intensity to gradually decay during continuous use, the change may be detailed by curve fitting, and further analyzing the possible cause of the tendency, such as material aging or thermal effects. And detecting the LED temperature change of the infrared spectrogram image. The infrared image can provide detailed information of the heat radiation of the LED during operation. The temperature change of the LEDs in each operation stage is monitored by detecting the temperature change of the images, and the temperature change is calculated by curve characteristics, so that the temperature data are converted into a form which can be quantitatively analyzed, the overheat problem of the LEDs can be identified, a system is helped to determine whether the heat dissipation design needs to be improved or the power setting needs to be adjusted so as to avoid overheat, and the stability and the efficiency of the LEDs are ensured. And extracting the characteristic values of the luminous intensity characteristic value set and the temperature characteristic value set. Key feature values are extracted from the respective data sets by statistical analysis and machine learning techniques, which feature values describe the core characteristics of the luminescence intensity and temperature variation. For example, the characteristic values of the luminous intensity may include the maximum light intensity, the average value of the light intensity, and the rate of change of the light intensity with time, and the characteristic values of the temperature may include the maximum temperature, the average temperature, and the fluctuation range of the temperature. And converting the characteristic value set into characteristic vectors, and combining the characteristic vectors of the luminous intensity and the temperature through a data fusion technology to form a luminous intensity and temperature relation vector. The vector contains the integrated information between the two important performance indicators.
In a specific embodiment, the performing step performs LED luminous intensity variation vector analysis and curve feature calculation on the multiband spectral image, and the process of obtaining the luminous intensity feature value set may specifically include the following steps:
(1) Performing difference image analysis on the multiband spectral image to obtain a multiband difference image;
(2) Carrying out change vector analysis on the multiband difference image to obtain luminous intensity distribution;
(3) Carrying out LED luminous intensity recognition on the multiband difference image according to luminous intensity distribution to obtain LED luminous intensity data;
(4) Performing curve fitting on the LED luminous intensity data to obtain an LED luminous intensity change curve;
(5) Carrying out change rate integral calculation on the LED luminous intensity change curve through a difference algorithm to obtain a plurality of change rate integral values;
(6) And performing time sequence association and integrated conversion on the plurality of change rate integral values to obtain a luminous intensity characteristic value set.
Specifically, the multi-band spectrum image is subjected to differential image analysis, and the spectrum changes of the LEDs at different time points or under different environmental conditions are captured. Spectral images of the same LED in different states are compared, and a multi-band difference image that clearly shows LED changes, which spectral regions show significant changes, which may be caused by LED temperature increases, material aging, or other environmental factors, is obtained by image processing techniques such as background subtraction and image alignment. And carrying out change vector analysis on the multiband difference image so as to quantify the change of the luminous intensity of each band. And analyzing each pixel point in the difference image, and calculating the variation amplitude and direction of the light intensity of the pixel points to obtain an overall luminous intensity distribution diagram. And analyzing the multiband difference image according to the luminous intensity distribution data to identify the specific luminous intensity of the LED. By setting a specific threshold and using pixel clustering techniques, key luminous intensity data is extracted from the luminous intensity distribution. And performing curve fitting on the luminous intensity data to form an LED luminous intensity change curve. The luminous data of the scattered points are converted into continuous curves by using a mathematical model such as polynomial fitting or exponential smoothing and the like, so that the change of the luminous characteristics of the LEDs with time is clearly described. And carrying out change rate integral calculation on the LED luminous intensity change curve by utilizing a differential algorithm. The difference algorithm obtains the integral change trend by calculating the change speed between adjacent data points on the curve and then integrating the speed. These integral values are important parameters that measure the change in the luminous performance of the LED over time and can indicate the stability or decay rate of the LED performance. And performing time sequence association and aggregation conversion on the change rate integral value to obtain a final luminous intensity characteristic value set. The integrated value is converted into a set of characteristic values describing the luminous characteristics of the LED by a statistical analysis method such as time series analysis and a characteristic extraction technology, and the characteristic values can be directly used for performance evaluation and optimization decision of the LED.
In a specific embodiment, the performing step performs LED temperature change detection and curve feature calculation on the infrared spectrogram image, and the process of obtaining the temperature feature value set may specifically include the following steps:
(1) Extracting pixel temperature values of the infrared spectrogram images to obtain a plurality of corresponding initial pixel temperature values;
(2) Calibrating the LED area on the initial pixel temperature values to obtain target pixel temperature values of the LED area;
(3) Performing average value operation on a plurality of target pixel temperature values of the LED area to obtain LED temperature data of an infrared spectrogram image;
(4) Performing curve fitting on the LED temperature data to obtain an LED temperature change curve;
(5) Carrying out curve fluctuation feature point identification on the LED temperature change curve to obtain a plurality of temperature curve fluctuation feature points;
(6) And generating a temperature characteristic value set corresponding to the infrared spectrogram according to the plurality of temperature curve fluctuation characteristic points.
Specifically, extracting pixel temperature values of infrared spectrogram images to obtain a plurality of corresponding initial pixel temperature values. The temperature value of each pixel point reflects the heat radiation intensity of the point, and a detailed temperature distribution map is obtained by converting the data. And (5) calibrating the LED area on the initial pixel temperature value. Areas directly related to the LEDs are identified from the entire image, while data for background and other non-LED areas are ignored. By applying image segmentation techniques, such as thresholding or edge detection, the LED body and surrounding environment can be effectively distinguished, and only the temperature values of the LED areas are extracted. And carrying out average value operation on a plurality of target pixel temperature values of the LED area to obtain a single value representing the average temperature of the whole LED area. By means of the mean value operation, the influence of abnormal values of single pixels is reduced, and a more stable and reliable temperature reading is obtained. And performing curve fitting on the temperature data to generate an LED temperature change curve. The discrete temperature data is converted into a smooth curve using statistical or mathematical models, such as polynomial fits or exponential smoothing, to facilitate observation and analysis of trends in temperature over time or other conditions such as current. And identifying curve fluctuation characteristic points of the LED temperature change curve to obtain a plurality of temperature curve fluctuation characteristic points. Including peaks, valleys and any unusual fluctuations in the look-up curve, these characteristic points may indicate problems with the thermal performance of the LED, such as overheating or insufficient cooling. Based on the identified plurality of temperature curve fluctuation feature points, a corresponding set of temperature feature values is generated. This set includes statistical indicators of temperature maxima, minima, averages, and fluctuation amplitudes, which provide a quantified reference for thermal management of the LEDs, contributing to performance optimization and fault prevention. For example, assume that an infrared camera is used to capture temperature images of a streetlamp during the night and daytime, respectively. Through the infrared image obtained by analysis, the LED is seen to have a temperature distribution generally higher than that at night in daytime due to a higher ambient temperature. Through data processing and curve fitting, it was found that the temperature rose sharply over some specified period of time, which may be due to overheating of some components in the LED driving circuit. By identifying these temperature fluctuation feature points, the heat dissipation system design is investigated and optimized to ensure that the LED street lamp maintains good operating performance, avoiding possible product failure.
In a specific embodiment, the performing step performs feature value extraction and feature vector creation on the set of luminous intensity feature values and the set of temperature feature values, and the process of obtaining the luminous intensity and temperature relation vector may specifically include the following steps:
(1) Calculating the mean value and standard deviation of the luminous intensity characteristic value set to obtain a luminous intensity characteristic mean value and a luminous intensity characteristic standard deviation, and calculating the mean value and standard deviation of the temperature characteristic value set to obtain a temperature characteristic mean value and a temperature characteristic standard deviation;
(2) Calculating a corresponding luminous intensity variation coefficient according to the luminous intensity characteristic mean value and the luminous intensity characteristic standard deviation, and calculating a corresponding temperature variation coefficient according to the temperature characteristic mean value and the temperature characteristic standard deviation;
(3) Setting a first weight coefficient of a luminous intensity characteristic value set according to the luminous intensity variation coefficient, and setting a second weight coefficient of a temperature characteristic value set according to the temperature variation coefficient;
(4) Performing feature code mapping and vector conversion on the luminous intensity feature value set according to the first weight coefficient to obtain a luminous intensity feature vector, and performing feature code mapping and vector conversion on the temperature feature value set according to the second weight coefficient to obtain a temperature feature vector;
(5) Vector splicing is carried out on the luminous intensity characteristic vector and the temperature characteristic vector, and a luminous intensity and temperature relation vector is obtained.
Specifically, the luminous intensity characteristic value set and the temperature characteristic value set are subjected to statistical analysis, and the mean value and the standard deviation of the luminous intensity characteristic value set and the temperature characteristic value set are calculated. The mean gives the average level of LED performance, while the standard deviation provides a measure of the variation in performance, showing the degree of dispersion of the data points around the mean. The coefficient of variation, which is a statistic characterizing the relative variability, is calculated by calculating the mean and standard deviation of the luminous intensity and temperature, and the calculation formula is the standard deviation divided by the mean. Higher coefficients of variation may indicate that the performance of the LED is unstable under certain operating conditions, requiring further investigation and optimization. And setting a weight coefficient according to the variation coefficient. The weight coefficient is set based on the assumption that: the higher the coefficient of variation, the greater the uncertainty of the corresponding feature, and therefore the lower the weight should be given when performing the final analysis. By the method, the contribution of each feature in the performance analysis is effectively balanced, and analysis result deviation caused by certain extreme values is avoided. And performing feature coding mapping and vector conversion on the luminous intensity feature value set and the temperature feature value set by using the set weight coefficients. The data is pre-processed, including normalization and normalization, to make it suitable for subsequent mathematical processing and model analysis. Feature encoding and vector conversion are the conversion of raw data into a format suitable for machine learning and data analysis, by which the data is further abstracted into mathematical objects for complex computation and pattern recognition. Finally, the method includes the steps of. Vector splicing is carried out on the luminous intensity characteristic vector and the temperature characteristic vector, and a comprehensive vector containing all key information is created. The luminous intensity and temperature relationship vectors are used to train predictive models, or to analyze the behavior characteristics of the LEDs in depth, thereby helping the system optimize product design and improve performance. For example, suppose a newly developed LED is evaluated for outdoor lighting. Tests were performed under a number of different environmental conditions including high temperature, low temperature, high humidity and dry conditions. By collecting and analyzing the luminous intensity and temperature data under different conditions, calculating a characteristic value set of the luminous intensity and the temperature, and then carrying out statistics and data processing, a luminous intensity and temperature relation vector is obtained, and the vector shows the performance of the LED under various environments.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model, wherein the LED luminous performance detection model comprises: a plurality of convolution pooling layers, an attention mechanism layer and a linear regression function;
(2) Carrying out convolution characteristic operation on the luminous intensity and temperature relation vector through a plurality of convolution pooling layers to obtain an output convolution characteristic vector of each convolution pooling layer;
(3) Inputting the output convolution feature vector of each convolution pooling layer into an attention mechanism layer for attention mechanism weighted fusion to obtain an attention fusion feature vector;
(4) Inputting the attention fusion feature vector into a linear regression function to calculate the LED luminous performance value, so as to obtain an LED luminous performance predicted value;
(5) And carrying out dynamic compensation analysis on the first LED power output parameter according to the LED luminous performance predicted value to obtain a second LED power output parameter corresponding to the target LED.
Specifically, the relation vector of the luminous intensity and the temperature is input into a preset LED luminous performance detection model. The vector contains key performance indicators for the LED under different conditions. And carrying out convolution characteristic operation on the luminous intensity and temperature relation vector through a plurality of convolution pooling layers, extracting local characteristics through convolution operation, wherein the pooling layers are used for reducing the space dimension of the characteristics, enhancing the abstract capacity of the model and reducing the calculated amount. Each convolution and pooling operation captures key information of LED performance and converts the data into a form that is easier to analyze, resulting in an output convolution feature vector for each convolution pooling layer. And inputting the output convolution feature vector of each convolution pooling layer into an attention mechanism layer for attention mechanism weighted fusion, and optimizing the accuracy of performance prediction by a model through learning how to distribute weights of different features. Attention mechanisms enhance the predictive power of models by focusing on those features that are most helpful in prediction, and the models can automatically identify and focus on those features that are most critical to LED lighting performance. The obtained attention fusion feature vector is a result obtained by comprehensively considering all important features and fusing the importance weights through learning. And inputting the attention fusion feature vector into a linear regression function to calculate the LED luminous performance value. The function calculates a predicted value from the input eigenvector, which represents an estimate of the luminous performance of the LED by the model. Linear regression is a predictive tool that can provide a straightforward, easily understood output, namely the expected performance index of an LED. And carrying out dynamic compensation analysis on the first LED power output parameter according to the LED luminous performance predicted value to obtain a second LED power output parameter corresponding to the target LED. And adjusting the power output parameters of the LEDs according to the performance predicted by the model, so as to optimize the operation efficiency and prolong the service life of the LEDs. For example, if the prediction indicates that the luminous efficiency of the LED decreases at higher temperatures, the power output parameters may be adjusted accordingly to reduce power consumption and control the temperature increase. Dynamic adjustment can help achieve more energy efficient and durable LED operation performance.
The method for optimizing LED temperature based on multispectral detection in the embodiment of the present application is described above, and the device for optimizing LED temperature based on multispectral detection in the embodiment of the present application is described below, referring to fig. 2, an embodiment of the device for optimizing LED temperature based on multispectral detection in the embodiment of the present application includes:
The acquisition module 201 is configured to perform luminescence test and multispectral image acquisition on the target LED based on the first LED power output parameter, so as to obtain a multiband spectral image and an infrared spectral image;
The construction module 202 is configured to perform LED variation analysis and feature vector construction on the multiband spectral image and the infrared spectral image respectively, so as to obtain a relation vector between luminous intensity and temperature;
The analysis module 203 is configured to input the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to perform LED luminous performance calculation and power output dynamic compensation analysis, so as to obtain a second LED power output parameter corresponding to the target LED.
Through the cooperation of the components, the temperature change of the LED can be monitored in real time by using infrared spectrogram images, so that the temperature abnormality in the LED light emitting process can be detected and responded in real time. This real-time monitoring capability ensures the accuracy of temperature management, thereby avoiding the drop in light efficiency and the reduction in lifetime caused by excessive temperatures. By analyzing the multiband spectral image and the infrared spectral image, the temperature is monitored, and the power output of the LED can be adjusted in real time. The dynamic adjustment can optimize power output according to real-time light emitting and temperature conditions, further improve energy efficiency and reduce energy consumption. Analysis of the multiband spectral image is helpful for monitoring the light color performance of the LED, and consistency and stability of light color output are ensured. The temperature optimization not only affects the service life and the energy efficiency, but also directly affects the color temperature and the light color purity, so that the light color quality of a product can be obviously improved by adjusting the light color output through optimizing the temperature parameter. As the temperature of the LED is effectively controlled, material degradation and light attenuation caused by overheating are avoided, and the service life of the LED is remarkably prolonged. The application adopts the multispectral detection technology to analyze and calculate the performance change in the LED operation process, thereby improving the control accuracy of the LED power output.
The application also provides a multi-spectrum detection-based LED temperature optimization device, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the multi-spectrum detection-based LED temperature optimization method in the embodiments.
The present application also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, which when executed on a computer, cause the computer to perform the steps of the LED temperature optimization method based on multispectral detection.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random acceS memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. An LED temperature optimization method based on multispectral detection, the method comprising:
Performing luminescence test and multispectral image acquisition on a target LED based on the first LED power output parameters to obtain a multiband spectral image and an infrared spectral image;
Respectively carrying out LED variation analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector;
And inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output, so as to obtain a second LED power output parameter corresponding to the target LED.
2. The method for optimizing LED temperature based on multispectral detection according to claim 1, wherein the performing the light emission test and the multispectral image acquisition on the target LED based on the first LED power output parameter to obtain the multiband spectral image and the infrared spectral image comprises:
initializing and setting power output parameters of a target LED to generate corresponding first LED power output parameters;
Performing a lighting test on the target LED according to the first LED power output parameter;
The method comprises the steps that multiband spectrum image acquisition is conducted on a target LED through a plurality of preset visible light wave bands and spectrum cameras, and a visible light initial spectrum image of each visible light wave band is obtained;
The method comprises the steps of carrying out image resolution enhancement on a visible light initial spectrum image of each visible light wave band to obtain a visible light enhanced spectrum image of each visible light wave band, and generating a multiband spectrum image according to the visible light enhanced spectrum image of each visible light wave band;
And acquiring an infrared spectrogram image of the target LED through a preset infrared light wave band and an infrared camera to obtain the infrared spectrogram image of the infrared light wave band.
3. The method for optimizing LED temperature based on multi-spectral detection according to claim 2, wherein the performing LED variation analysis and feature vector construction on the multi-band spectral image and the infrared spectral image, respectively, to obtain a luminous intensity and temperature relation vector, comprises:
Carrying out LED luminous intensity change vector analysis and curve characteristic calculation on the multiband spectral image to obtain a luminous intensity characteristic value set;
carrying out LED temperature change detection and curve characteristic calculation on the infrared spectrogram image to obtain a temperature characteristic value set;
And extracting the characteristic value of the luminous intensity characteristic value set and the characteristic value of the temperature characteristic value set and creating a characteristic vector to obtain a luminous intensity and temperature relation vector.
4. The method for optimizing LED temperature based on multi-spectral detection of claim 3, wherein said performing LED luminous intensity variation vector analysis and curve feature calculation on said multi-band spectral image to obtain a set of luminous intensity feature values comprises:
performing difference image analysis on the multiband spectral image to obtain a multiband difference image;
performing change vector analysis on the multiband difference image to obtain luminous intensity distribution;
Carrying out LED luminous intensity recognition on the multiband difference image according to the luminous intensity distribution to obtain LED luminous intensity data;
Performing curve fitting on the LED luminous intensity data to obtain an LED luminous intensity change curve;
performing change rate integral calculation on the LED luminous intensity change curve through a difference algorithm to obtain a plurality of change rate integral values;
And performing time sequence association and integrated conversion on the plurality of change rate integral values to obtain a luminous intensity characteristic value set.
5. The method for optimizing LED temperature based on multispectral detection according to claim 3, wherein said performing LED temperature change detection and curve feature calculation on the infrared spectrogram image to obtain a temperature feature value set comprises:
extracting pixel temperature values of the infrared spectrogram images to obtain a plurality of corresponding initial pixel temperature values;
calibrating the LED area to the initial pixel temperature values to obtain target pixel temperature values of the LED area;
Performing average value operation on a plurality of target pixel temperature values of the LED area to obtain LED temperature data of the infrared spectrogram image;
Performing curve fitting on the LED temperature data to obtain an LED temperature change curve;
Performing curve fluctuation feature point identification on the LED temperature change curve to obtain a plurality of temperature curve fluctuation feature points;
and generating a temperature characteristic value set corresponding to the infrared spectrogram according to the plurality of temperature curve fluctuation characteristic points.
6. The method for optimizing LED temperature based on multispectral detection of claim 3, wherein said performing feature value extraction and feature vector creation on said set of luminous intensity feature values and said set of temperature feature values to obtain luminous intensity and temperature relationship vectors comprises:
Calculating the mean value and standard deviation of the luminous intensity characteristic value set to obtain a luminous intensity characteristic mean value and a luminous intensity characteristic standard deviation, and calculating the mean value and standard deviation of the temperature characteristic value set to obtain a temperature characteristic mean value and a temperature characteristic standard deviation;
Calculating a corresponding luminous intensity variation coefficient according to the luminous intensity characteristic mean value and the luminous intensity characteristic standard deviation, and calculating a corresponding temperature variation coefficient according to the temperature characteristic mean value and the temperature characteristic standard deviation;
Setting a first weight coefficient of the luminous intensity characteristic value set according to the luminous intensity variation coefficient, and setting a second weight coefficient of the temperature characteristic value set according to the temperature variation coefficient;
Performing feature code mapping and vector conversion on the luminous intensity feature value set according to the first weight coefficient to obtain a luminous intensity feature vector, and performing feature code mapping and vector conversion on the temperature feature value set according to the second weight coefficient to obtain a temperature feature vector;
And vector splicing is carried out on the luminous intensity characteristic vector and the temperature characteristic vector to obtain a luminous intensity and temperature relation vector.
7. The method for optimizing LED temperature based on multispectral detection according to claim 1, wherein the step of inputting the luminous intensity and temperature relation vector into a preset LED luminous performance detection model to perform LED luminous performance calculation and power output dynamic compensation analysis, to obtain a second LED power output parameter corresponding to the target LED, comprises:
Inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model, wherein the LED luminous performance detection model comprises the following components: a plurality of convolution pooling layers, an attention mechanism layer and a linear regression function;
performing convolution characteristic operation on the luminous intensity and temperature relation vector through the plurality of convolution pooling layers to obtain an output convolution characteristic vector of each convolution pooling layer;
Inputting the output convolution feature vector of each convolution pooling layer into the attention mechanism layer to perform attention mechanism weighted fusion to obtain an attention fusion feature vector;
Inputting the attention fusion feature vector into the linear regression function to calculate the LED luminous performance value, so as to obtain an LED luminous performance predicted value;
And carrying out dynamic compensation analysis on the first LED power output parameter according to the LED luminous performance predicted value to obtain a second LED power output parameter corresponding to the target LED.
8. An LED temperature optimization device based on multispectral detection, wherein the LED temperature optimization device based on multispectral detection comprises:
the acquisition module is used for carrying out luminescence test and multispectral image acquisition on the target LED based on the first LED power output parameter to obtain a multiband spectral image and an infrared spectral image;
the construction module is used for respectively carrying out LED change analysis and feature vector construction on the multiband spectral image and the infrared spectral image to obtain a luminous intensity and temperature relation vector;
and the analysis module is used for inputting the relation vector of the luminous intensity and the temperature into a preset LED luminous performance detection model to calculate the luminous performance of the LED and dynamically compensate and analyze the power output, so as to obtain a second LED power output parameter corresponding to the target LED.
9. An LED temperature optimization apparatus based on multispectral detection, characterized in that the LED temperature optimization apparatus based on multispectral detection comprises: a memory and at least one processor, the memory having instructions stored therein;
The at least one processor invokes the instructions in the memory to cause the multispectral detection-based LED temperature optimization apparatus to perform the multispectral detection-based LED temperature optimization method of any one of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement the LED temperature optimization method based on multispectral detection of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410533460.4A CN118130996B (en) | 2024-04-30 | 2024-04-30 | LED temperature optimization method and system based on multispectral detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410533460.4A CN118130996B (en) | 2024-04-30 | 2024-04-30 | LED temperature optimization method and system based on multispectral detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118130996A true CN118130996A (en) | 2024-06-04 |
| CN118130996B CN118130996B (en) | 2024-07-26 |
Family
ID=91245930
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410533460.4A Active CN118130996B (en) | 2024-04-30 | 2024-04-30 | LED temperature optimization method and system based on multispectral detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118130996B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130082622A1 (en) * | 2011-09-29 | 2013-04-04 | National Chiao Tung University | Method for mixing light of led cluster |
| US20160161294A1 (en) * | 2014-12-01 | 2016-06-09 | Henry Ip | Automated test systems and methods for light-emitting arrays |
| CN106979832A (en) * | 2017-03-22 | 2017-07-25 | 河南北方红阳机电有限公司 | A kind of optical fibre light splitting temp measuring system and its temp measuring method |
| CN208224466U (en) * | 2018-06-08 | 2018-12-11 | 临沂大学 | A kind of quick LED lamp bead photoelectric parameter integrated test system |
| CN109885890A (en) * | 2019-01-23 | 2019-06-14 | 中国地质大学（武汉） | A kind of light source light spectrum optimization method and system based on color entropy |
| CN111487037A (en) * | 2020-04-16 | 2020-08-04 | 中国科学院微电子研究所 | System and method for detecting uniformity of light source |
| CN114705698A (en) * | 2022-06-02 | 2022-07-05 | 季华实验室 | Defect detection method, device, system and storage medium |
| CN116734220A (en) * | 2023-08-11 | 2023-09-12 | 广州市升龙灯光设备有限公司 | Stage lamp calibration method and system |
| CN117479373A (en) * | 2023-12-28 | 2024-01-30 | 深圳市兴邦维科科技有限公司 | Method and system for adjusting brightness of light-emitting diode |
| CN117495512A (en) * | 2023-12-29 | 2024-02-02 | 干霸干燥剂(深圳)有限公司 | Order data management method, device, equipment and storage medium |
| CN117575672A (en) * | 2023-11-15 | 2024-02-20 | 国网吉林省电力有限公司营销服务中心 | Industry electric quantity prediction method and device based on space-time feature transfer learning |
| CN117881047A (en) * | 2023-12-26 | 2024-04-12 | 季华实验室 | Integrating sphere spectrum adjusting method and device and computer readable storage medium |
-
2024
- 2024-04-30 CN CN202410533460.4A patent/CN118130996B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130082622A1 (en) * | 2011-09-29 | 2013-04-04 | National Chiao Tung University | Method for mixing light of led cluster |
| US20160161294A1 (en) * | 2014-12-01 | 2016-06-09 | Henry Ip | Automated test systems and methods for light-emitting arrays |
| CN106979832A (en) * | 2017-03-22 | 2017-07-25 | 河南北方红阳机电有限公司 | A kind of optical fibre light splitting temp measuring system and its temp measuring method |
| CN208224466U (en) * | 2018-06-08 | 2018-12-11 | 临沂大学 | A kind of quick LED lamp bead photoelectric parameter integrated test system |
| CN109885890A (en) * | 2019-01-23 | 2019-06-14 | 中国地质大学（武汉） | A kind of light source light spectrum optimization method and system based on color entropy |
| CN111487037A (en) * | 2020-04-16 | 2020-08-04 | 中国科学院微电子研究所 | System and method for detecting uniformity of light source |
| CN114705698A (en) * | 2022-06-02 | 2022-07-05 | 季华实验室 | Defect detection method, device, system and storage medium |
| CN116734220A (en) * | 2023-08-11 | 2023-09-12 | 广州市升龙灯光设备有限公司 | Stage lamp calibration method and system |
| CN117575672A (en) * | 2023-11-15 | 2024-02-20 | 国网吉林省电力有限公司营销服务中心 | Industry electric quantity prediction method and device based on space-time feature transfer learning |
| CN117881047A (en) * | 2023-12-26 | 2024-04-12 | 季华实验室 | Integrating sphere spectrum adjusting method and device and computer readable storage medium |
| CN117479373A (en) * | 2023-12-28 | 2024-01-30 | 深圳市兴邦维科科技有限公司 | Method and system for adjusting brightness of light-emitting diode |
| CN117495512A (en) * | 2023-12-29 | 2024-02-02 | 干霸干燥剂(深圳)有限公司 | Order data management method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 褚金金;崔桂华;柳耀武;徐婷;阮秀凯;蔡启博;谈燕花;: "一种基于光谱可调LED光源和多光谱成像技术的物体表面颜色测量方法", 光学学报, no. 08, 28 March 2018 (2018-03-28) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118130996B (en) | 2024-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116780316B (en) | Heat radiation system of high-power solid laser | |
| CN112032034A (en) | Sea water pump health state assessment method, model, device and storage medium | |
| CN113609901A (en) | Power transmission and transformation equipment fault monitoring method and system | |
| CN117320236B (en) | Lighting control method and system of unmanned aerial vehicle | |
| CN118013427B (en) | Liquid cooling heat dissipation automobile charging pile for liquid leakage early warning and early warning method | |
| CN115661017A (en) | Infrared thermal imaging super-pixel segmentation and fusion method, system, medium and computer | |
| KR20160022540A (en) | Method for estimating customer baseline load using data maning and apparatus thereof | |
| CN111768402A (en) | MU-SVM-based method for evaluating freshness of iced pomfret | |
| CN111444178B (en) | Light pollution evaluation method and device | |
| CN118094347B (en) | Self-adaptive adjustment optimization method and device for liquid cooling energy storage module and storage medium | |
| CN117479373A (en) | Method and system for adjusting brightness of light-emitting diode | |
| CN117851817B (en) | Intelligent filtering analysis method, system and storage medium for humid air | |
| CN118130996B (en) | LED temperature optimization method and system based on multispectral detection | |
| CN117475806B (en) | Display screen self-adaptive response method and device based on multidimensional sensing data feedback | |
| CN117222069A (en) | Intelligent control method of illumination driving power supply | |
| Zohra et al. | A linear regression model for estimating facial image quality | |
| CN117931925A (en) | Remote information management method and system for energy-saving equipment operation | |
| CN116539285A (en) | Light source detection method, device, equipment and storage medium based on artificial intelligence | |
| CN116668859A (en) | GIS equipment computer lab environment monitoring device | |
| CN111091243A (en) | PCA-GM-based power load prediction method, system, computer-readable storage medium, and computing device | |
| CN115935254A (en) | LED early abnormity diagnosis method and system based on spectral power distribution | |
| CN117294824B (en) | Image optimization method, device and equipment of laser projection optical machine and storage medium | |
| CN117974658B (en) | Cable anomaly identification method and system based on image processing | |
| CN117613822B (en) | Method, device, equipment and storage medium for fast shutdown of frequency converter | |
| CN118249479B (en) | Intelligent temperature management super-capacity energy storage method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |