A feedback analysis-based LED light source control optimization method and system

Abstract

The application discloses an LED light source control optimization method and system based on feedback analysis. The method comprises the following steps: collecting working state parameters and a feedback image set of an LED light source in a working cycle; extracting global color features of images at each time node, calculating difference values of the global color features and required color features, and performing linear change evaluation to obtain global attenuation coefficients; selecting an optimal node, extracting feedback brightness and local matrix pixels, analyzing an RGB color matrix, and calculating local attenuation coefficients of three colors; combining stability analysis results of the working state parameters to analyze light source attenuation conditions, and realizing control optimization and equipment maintenance of the LED light source.

Description

Technical Field

[0001] This invention relates to the field of LED control and data analysis, and more specifically, to an LED light source control optimization method and system based on feedback analysis. Background Technology

[0002] LED light sources are electroluminescent devices based on semiconductor PN junctions. They generate high-purity, high-stability light through current excitation and possess technological advantages such as high photoelectric conversion efficiency, long lifespan (over 30,000 hours), fast response speed (microsecond level), and precise and controllable spectrum. They have been widely used in lighting, displays, security monitoring, industrial inspection, medical equipment, and many other fields, becoming the mainstream development direction in the current light source field.

[0003] However, in the practical application of LED light sources, their performance is constrained by various factors such as control strategies and application scenario adaptability, leading to increasingly prominent technical problems that seriously affect the application effect and promotion process of LED light sources. Regarding the evaluation of the lifespan and projection lighting effect of LED light sources, existing technologies lack scientific, accurate, and real-time evaluation methods. Most methods rely on fixed-period testing or experience-based judgment, which cannot dynamically reflect the color and lifespan degradation of the light source under different working states and complex environmental conditions. This results in the inability to predict the risk of light source failure in advance, and the inability to accurately test and detect the global and local color differences and degradation of the light source. Consequently, scientific optimization and control cannot be performed, which has a significant impact, especially in industrial machine vision inspection scenarios. Furthermore, adjustments can generally only be made based on human experience, making them impractical. Therefore, there is an urgent need to develop an LED light source control optimization method based on feedback analysis that can effectively solve the above problems. Summary of the Invention

[0004] This invention overcomes the shortcomings of the prior art and proposes an LED light source control optimization method and system based on feedback analysis.

[0005] The first aspect of this invention provides an LED light source control optimization method based on feedback analysis, comprising: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

[0006] In this scheme, S11 specifically refers to: Set the LED light source's working cycle and collect working status parameters and feedback image sets through the data acquisition unit; The data acquisition unit includes a light source sensor and a camera device; The Gaussian filtering algorithm is used to denoise the feedback image set. The size of the Gaussian filter convolution kernel is dynamically adjusted according to the image noise intensity. The kernel size is set within the range of 3×3-7×7. The image set is also preprocessed by standardization.

[0007] In this scheme, the operating status parameters include operating current, operating voltage, light source temperature, flicker frequency, and flicker duty cycle.

[0008] In this scheme, S12 specifically refers to: In the work cycle, multiple time nodes are set and images are extracted that are concentrated under each node. The color histogram analysis method is introduced to convert the image into the HSV color space and extract the color histograms of the H, S, and V channels respectively. The color histogram of each channel is normalized to obtain the normalized histogram of each channel. Multi-dimensional color features are generated based on the normalized histograms of multiple channels and used as global color features. The global color features and the required color features are vectorized and their differences are calculated to obtain multiple difference values. Linear regression is performed based on the difference values ​​at each time point, and the fitted linear coefficient is used as the global decay coefficient. The global decay coefficient is set based on the mean of the difference values.

[0009] In this scheme, S13 specifically refers to: Sort the differences in descending order, select a preset proportion of differences for marking, and mark the corresponding time nodes as preferred nodes. Using a light sensor, obtain the brightness feedback value of the LED light source within the preferred node; The image under the preferred node is extracted to obtain the preferred image. A local region based on N×N pixels is constructed. Multiple local regions are randomly selected in the preferred image to obtain multiple local matrix pixels. Color analysis is performed on the local matrix pixels using the RGB three dimensions to obtain three color matrices.

[0010] In this scheme, S13 also includes: Based on the lighting demand information during the work cycle, obtain the brightness demand value of the preferred node; The difference between the brightness feedback value of the preferred node and the brightness requirement value is calculated by equalization, and the local color adjustment weight is set according to the difference. Multiply the three color matrices with the local color adjustment weights; By introducing the Jordan standard form, a similarity judgment is made between the color matrix and the contrast matrix corresponding to each local matrix pixel, and the local attenuation coefficient is calculated based on the judgment result.

[0011] In this scheme, S14 specifically refers to: For each time node, the working status parameters are compared with the normal working parameters to analyze the degree of deviation of multiple parameters. Based on the weight settings of different parameters, the deviation is weighted to obtain the working stability coefficient of each node. The stability of the working state is evaluated by calculating the growth rate of the working stability coefficient based on multiple nodes. The correlation between operational stability and changes in global and local attenuation coefficients was analyzed to optimize the control and maintenance of LED light sources.

[0012] In this scheme, a working cycle includes multiple time nodes, and the selection of local matrix pixels includes multiple nodes.

[0013] A second aspect of the present invention also provides an LED light source control optimization system based on feedback analysis. The system includes a memory, a processor, and a data interface. The memory includes an LED light source control optimization program based on feedback analysis. When executed by the processor, the LED light source control optimization program based on feedback analysis performs the following steps: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

[0014] A third aspect of the present invention also provides a computer-readable storage medium comprising an LED light source control optimization program based on feedback analysis, wherein when the LED light source control optimization program based on feedback analysis is executed by a processor, it implements the steps of the LED light source control optimization method based on feedback analysis as described in any of the preceding claims.

[0015] This invention discloses an LED light source control optimization method and system based on feedback analysis. The method includes: acquiring LED light source operating status parameters and feedback image sets within a working cycle; extracting global color features from images at each time point, calculating the difference between the extracted and required color features, and performing linear change evaluation to obtain a global attenuation coefficient; selecting preferred nodes, extracting feedback brightness and local matrix pixels, parsing the RGB color matrix, and calculating the local attenuation coefficients for the three colors; combining the stability analysis of the operating status parameters with the light source attenuation results, and realizing LED light source control optimization and equipment maintenance. Attached Figure Description

[0016] Figure 1 A flowchart of an LED light source control optimization method based on feedback analysis according to the present invention is shown; Figure 2 The flowchart of the global attenuation coefficient analysis of the present invention is shown; Figure 3 A block diagram of an LED light source control optimization system based on feedback analysis according to the present invention is shown. Detailed Implementation

[0017] To better understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. In the embodiments, it will be understood that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0018] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0019] Figure 1 A flowchart of an LED light source control optimization method based on feedback analysis according to the present invention is shown.

[0020] like Figure 1 As shown, the first aspect of the present invention provides an LED light source control optimization method based on feedback analysis, comprising: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

[0021] It is worth mentioning that, through this embodiment, precise control of the LED light source is achieved, improving the lifespan, brightness, and color performance of the light source, enhancing the system's adaptability and scientific maintenance, and enabling precise adaptation to diverse lighting application scenarios. This invention comprehensively evaluates the control of the LED light source, and by analyzing the correlation between the attenuation coefficient and the operating state, it can further assess the lifespan of the LED light source, providing a scientific basis for intelligent control optimization, efficient light source testing, and high-precision stability and lifespan assessment of LED light sources.

[0022] According to an embodiment of the present invention, S11 specifically includes: Set the LED light source's working cycle and collect working status parameters and feedback image sets through the data acquisition unit; The data acquisition unit includes a light source sensor and a camera device; The Gaussian filtering algorithm is used to denoise the feedback image set. The size of the Gaussian filter convolution kernel is dynamically adjusted according to the image noise intensity. The kernel size is set within the range of 3×3-7×7. The image set is also preprocessed by standardization.

[0023] In this embodiment, it can be understood that the working cycle is specifically a preset cycle within a stable working time of the target LED light source. In a long-term working state, this includes multiple working cycles. Standardized preprocessing ensures consistency in image size, scale, color space, and resolution. The feedback image set is used to analyze the lighting effect presented by the LED light source projected onto a preset object.

[0024] According to an embodiment of the present invention, the operating status parameters include operating current, operating voltage, light source temperature, flicker frequency, and flicker duty cycle.

[0025] In the embodiments, it can be understood that the working status parameters are set by adding other light source parameters based on the analysis requirements. Here, environmental parameters can also be introduced to correct the attenuation of light source color feedback and the stability of the status, such as ambient light intensity, ambient temperature, ambient humidity, dust concentration and other parameters.

[0026] Figure 2 A flowchart of the global attenuation coefficient analysis of the present invention is shown.

[0027] According to an embodiment of the present invention, S12 specifically includes: In the work cycle, multiple time nodes are set and images are extracted that are concentrated under each node. The color histogram analysis method is introduced to convert the image into the HSV color space and extract the color histograms of the H, S, and V channels respectively. The color histogram of each channel is normalized to obtain the normalized histogram of each channel. Multi-dimensional color features are generated based on the normalized histograms of multiple channels and used as global color features. The global color features and the required color features are vectorized and their differences are calculated to obtain multiple difference values. Linear regression is performed based on the difference values ​​at each time point, and the fitted linear coefficient is used as the global decay coefficient. The global decay coefficient is set based on the mean of the difference values.

[0028] In this embodiment, it can be understood that the color histogram includes three channels: H (hue), S (saturation), and V (brightness). The required color features are specifically the color histogram features corresponding to the ideal feedback image in a specific scene, used for comparative analysis. The mean difference value can be directly set as the global attenuation coefficient; the larger the difference value, the more severe the attenuation. Vectorization and difference calculation can incorporate Mahalanobis distance calculation.

[0029] Linear regression fitting uses time as the independent variable and the difference value as the dependent variable, based on a univariate regression equation. When the global decay coefficient is positive, it indicates that the global color decay gradually increases over time; when the global decay coefficient is negative, it indicates that the global color decay gradually decreases over time; when the global decay coefficient is 0, it indicates that the global color decay remains stable.

[0030] According to an embodiment of the present invention, step S13 specifically includes: Sort the differences in descending order, select a preset proportion of differences for marking, and mark the corresponding time nodes as preferred nodes. Using a light sensor, obtain the brightness feedback value of the LED light source within the preferred node; The image under the preferred node is extracted to obtain the preferred image. A local region based on N×N pixels is constructed. Multiple local regions are randomly selected in the preferred image to obtain multiple local matrix pixels. Color analysis is performed on the local matrix pixels using the RGB three dimensions to obtain three color matrices.

[0031] In this embodiment, it is understood that each time point corresponds to a difference value. The preset ratio can be set to the first 20%-30% of the difference values, and the light source sensor includes a light sensor. Here, selecting the preferred node can reduce the impact of data noise caused by the instability of the light source at certain nodes, and effectively evaluate the local control accuracy and local color attenuation of the light source in the three dimensions of RGB. At the same time, the light source parameters can be reasonably adjusted according to the attenuation.

[0032] The color matrix is ​​represented as follows (taking 3×3 as an example):

[0033] The values ​​in the matrix represent color values. For example, in the matrix above, the value for red is [0, 255].

[0034] According to an embodiment of the present invention, step S13 further includes: Based on the lighting demand information during the work cycle, obtain the brightness demand value of the preferred node; The difference between the brightness feedback value of the preferred node and the brightness requirement value is calculated by equalization, and the local color adjustment weight is set according to the difference. Multiply the three color matrices with the local color adjustment weights; By introducing the Jordan standard form, a similarity judgment is made between the color matrix and the contrast matrix corresponding to each local matrix pixel, and the local attenuation coefficient is calculated based on the judgment result.

[0035] In the embodiments, it can be understood that the lighting demand information includes feedback requirements such as the lighting effect, brightness, and color of the light source.

[0036] The local attenuation coefficient includes attenuation coefficients in three dimensions (RGB light source). Taking red (R) as an example, multiple color matrices (R) are constructed based on multiple local matrix pixels. The proportion P of matrices with similarity is determined, and the corresponding attenuation coefficient is represented by (1-P). The larger the local attenuation coefficient, the greater the attenuation of the color LED feedback in the corresponding dimension, and the greater the deviation. Each local matrix pixel (a local pixel region) corresponds to three color matrices, and here we take red (R) as an example.

[0037] It is worth mentioning that this invention, by screening the global color attenuation coefficient and extracting the preferred nodes for multi-dimensional local color evaluation based on the attenuation differences of each node, can effectively reduce the feedback test error of LED light source caused by environmental factors. At the same time, by differentiating the color matrix according to environmental factors (brightness), the accuracy of the final attenuation analysis is improved. Finally, the color feedback attenuation is judged by the color matrix of each dimension, and the control status of the LED light source is comprehensively evaluated. Furthermore, by examining the correlation between the attenuation coefficient and the working status, the lifespan of the LED light source can be further evaluated, providing a scientific basis for intelligent control optimization, efficient light source testing, and high-precision stability and lifespan assessment of LED light sources.

[0038] The weighting of local color adjustments is calculated as follows: , Where Q is the adjustment weight and K is the preset correction coefficient. U represents the brightness feedback value of a preferred node, where U is the brightness requirement value and M is the number of preferred nodes.

[0039] According to an embodiment of the present invention, S14 specifically includes: For each time node, the working status parameters are compared with the normal working parameters to analyze the degree of deviation of multiple parameters. Based on the weight settings of different parameters, the deviation is weighted to obtain the working stability coefficient of each node. The stability of the working state is evaluated by calculating the growth rate of the working stability coefficient based on multiple nodes. The correlation between operational stability and changes in global and local attenuation coefficients was analyzed to optimize the control and maintenance of LED light sources.

[0040] In this embodiment, it is understood that in the deviation weighting, light source temperature, flicker frequency, and flicker duty cycle are generally assigned larger weights, while the rest are assigned default weights. The global attenuation coefficient reflects the overall color attenuation, while the local attenuation coefficient reflects the attenuation of the light source effect in the three color dimensions.

[0041] In S14, the correlation between changes in the operational stability of multi-dimensional parameters and the attenuation coefficient can be used to assess defects in the light source's lighting effect control and hardware. For example, if stability tends to decrease and the global attenuation coefficient increases, it indicates a defect in the light source's control and LED hardware, requiring inspection and adjustment of the LED bead angle, light source wavelength, brightness parameters, and color control parameters to meet the required scenario. If stability remains unchanged and the local attenuation coefficient increases, it indicates that the light source is working normally, but the multi-dimensional color control parameters need optimization. If stability tends to be normal and the attenuation coefficient tends to be stable or decreasing, it indicates normal operation and the lighting effect meets the requirements, with a longer lifespan in this state. If both are abnormal in certain areas, it indicates a defect, and continuous LED operation will affect the LED's lifespan, requiring inspection.

[0042] Here, by analyzing the relationship between stability and attenuation coefficient, the variation law of LED control state and working life can be analyzed, which can efficiently match maintenance, abnormal time nodes and screen out defective states, thereby improving the working efficiency and life of LED light source.

[0043] According to an embodiment of the present invention, a working cycle includes multiple time nodes, and the selection of local matrix pixels includes multiple nodes.

[0044] According to an embodiment of the present invention, selecting multiple local regions further includes: In the preferred image, an N×N pixel movement matrix is ​​set with a step size of N / 2. The movement matrix is ​​then moved vertically and horizontally in the preferred image. The preferred image is converted to grayscale, and the grayscale value is used as the pixel brightness value. During each movement, the deviation between the minimum and maximum image brightness values ​​within the movement matrix and the pixel requirement value is calculated. Based on minimizing the deviation, multiple movement matrices are selected and multiple local matrix pixel points are set.

[0045] Here, a preset step size is introduced to filter local matrix pixels in the optimized image. For images with high noise, the optimal image within the optimized image is selected, improving the accuracy of subsequent attenuation coefficient analysis. Pixel requirement values ​​can be inferred from the grayscale image based on brightness requirement values.

[0046] Figure 3 A block diagram of an LED light source control optimization system based on feedback analysis according to the present invention is shown.

[0047] A second aspect of the present invention also provides an LED light source control optimization system based on feedback analysis. The system includes a memory, a processor, and a data interface. The memory includes an LED light source control optimization program based on feedback analysis. When executed by the processor, the LED light source control optimization program based on feedback analysis performs the following steps: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

[0048] The data interface is used to send the analysis results to the user terminal. When the system is running, it can implement one or more steps of the LED light source control optimization method based on feedback analysis described above.

[0049] A third aspect of the present invention also provides a computer-readable storage medium comprising an LED light source control optimization program based on feedback analysis, wherein when the LED light source control optimization program based on feedback analysis is executed by a processor, it implements the steps of the LED light source control optimization method based on feedback analysis as described in any of the preceding claims.

[0050] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application can be generated, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic cable, data subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital universal optical disc), or a semiconductor medium (e.g., solid-state drive). In the various embodiments of this application, unless otherwise specified or logically conflicting, the terminology and / or descriptions between different embodiments are consistent and can be mutually referenced. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0051] In the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone, where A and B can be singular or plural. In the textual description of the embodiments of this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. In this application, "first," "second," and various numerical designations are only for the convenience of description and are not used to limit the scope of the embodiments of this application. For example, they are used to distinguish different messages, rather than to describe a specific order or sequence.

[0052] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

[0053] Finally, it should be noted that the above description is only a specific implementation of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the protection scope of this application.

Claims

1. A method for optimizing LED light source control based on feedback analysis, characterized in that, include: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

2. The LED light source control optimization method based on feedback analysis according to claim 1, characterized in that, Specifically, S11 is as follows: Set the LED light source's working cycle and collect working status parameters and feedback image sets through the data acquisition unit; The data acquisition unit includes a light source sensor and a camera device; The Gaussian filtering algorithm is used to denoise the feedback image set. The size of the Gaussian filter convolution kernel is dynamically adjusted according to the image noise intensity. The kernel size is set within the range of 3×3-7×7. The image set is also preprocessed by standardization.

3. The LED light source control optimization method based on feedback analysis according to claim 2, characterized in that, The operating status parameters include operating current, operating voltage, light source temperature, strobe frequency, and strobe duty cycle.

4. The LED light source control optimization method based on feedback analysis according to claim 3, characterized in that, Specifically, S12 is: In the work cycle, multiple time nodes are set and images are extracted that are concentrated under each node. The color histogram analysis method is introduced to convert the image into the HSV color space and extract the color histograms of the H, S, and V channels respectively. The color histogram of each channel is normalized to obtain the normalized histogram of each channel. Multi-dimensional color features are generated based on the normalized histograms of multiple channels and used as global color features. The global color features and the required color features are vectorized and their differences are calculated to obtain multiple difference values. Linear regression is performed based on the difference values ​​at each time point, and the fitted linear coefficient is used as the global decay coefficient. The global decay coefficient is set based on the mean of the difference values.

5. The LED light source control optimization method based on feedback analysis according to claim 4, characterized in that, Specifically, S13 is as follows: Sort the differences in descending order, select a preset proportion of differences for marking, and mark the corresponding time nodes as preferred nodes. Using a light sensor, obtain the brightness feedback value of the LED light source within the preferred node; The image under the preferred node is extracted to obtain the preferred image. A local region based on N×N pixels is constructed. Multiple local regions are randomly selected in the preferred image to obtain multiple local matrix pixels. Color analysis is performed on the local matrix pixels using the RGB three dimensions to obtain three color matrices.

6. The LED light source control optimization method based on feedback analysis according to claim 5, characterized in that, S13 further includes: Based on the lighting demand information during the work cycle, obtain the brightness demand value of the preferred node; The difference between the brightness feedback value of the preferred node and the brightness requirement value is calculated by equalization, and the local color adjustment weight is set according to the difference. Multiply the three color matrices with the local color adjustment weights; By introducing the Jordan standard form, a similarity judgment is made between the color matrix and the contrast matrix corresponding to each local matrix pixel, and the local attenuation coefficient is calculated based on the judgment result.

7. The LED light source control optimization method based on feedback analysis according to claim 6, characterized in that, Specifically, S14 is: For each time node, the working status parameters are compared with the normal working parameters to analyze the degree of deviation of multiple parameters. Based on the weight settings of different parameters, the deviation is weighted to obtain the working stability coefficient of each node. The stability of the working state is evaluated by calculating the growth rate of the working stability coefficient based on multiple nodes. The correlation between operational stability and changes in global and local attenuation coefficients was analyzed to optimize the control and maintenance of LED light sources.

8. The LED light source control optimization method based on feedback analysis according to claim 7, characterized in that, A work cycle includes multiple time nodes, and the selection of local matrix pixels includes multiple nodes.

9. An LED light source control optimization system based on feedback analysis, characterized in that, The system includes: a memory, a processor, and a data interface. The memory includes an LED light source control optimization program based on feedback analysis. When the processor executes the LED light source control optimization program based on feedback analysis, it performs the following steps: S11. Within one working cycle, the working status parameters of the LED light source and the feedback image set are collected through the data acquisition unit, and the image set is preprocessed. S12. In the feedback image set, based on each time node, extract the corresponding image and introduce a color histogram to extract color features, obtain global color features, calculate the difference between the global color features and the required color features, evaluate the linear change based on the difference value at each time node, and use the linear coefficient as the global attenuation coefficient. S13. Select time nodes where the difference value is within a preset range and mark them to obtain the preferred nodes. Extract the feedback brightness and the corresponding local matrix pixels of the image under the preferred nodes, and perform color analysis on the local matrix pixels based on the three dimensions of RGB to obtain three color matrices. Dynamically adjust the color matrix values ​​based on the deviation between the feedback brightness and the expected value. Determine the similarity between the color matrix and the contrast matrix based on the Jordan standard form, and calculate the local attenuation coefficient. S14. Analyze the working stability of the LED light source through working status parameters, and optimize the control of the LED light source and maintain the equipment by combining the global attenuation coefficient and the local attenuation coefficient.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes an LED light source control optimization program based on feedback analysis, which, when executed by a processor, implements the steps of the LED light source control optimization method based on feedback analysis as described in any one of claims 1 to 8.

Images