A light homogenization method and system based on diffusion film

Abstract

The present invention relates to the technical field of diffusion films, and specifically to a method and system for light homogenization based on diffusion films, comprising the following steps: based on a high-speed camera, continuously capturing the light distribution image of the diffusion film, performing real-time decoding and transmittance analysis on the image to obtain preliminary image data, and performing scattering angle detection and light spot analysis on the preliminary image data to obtain the original image analysis record. The present invention significantly improves the ability to evaluate the light scattering characteristics through the capture and real-time decoding of the light distribution image, improves the measurement accuracy of light uniformity through in-depth analysis of the light spots, shadows and light density caused by the diffusion film, and realizes dynamic adjustment of the light source configuration based on measured data. This real-time feedback mechanism allows instant optimization of the light source brightness and material structure, achieving more refined light control, thereby improving the uniformity of light distribution. In addition, continuous monitoring of the adjustment effect ensures long-term light output quality.

Description

Technical Field

[0001] The present invention relates to the technical field of diffusion films, and in particular to a light homogenization method and system based on a diffusion film. Background Art

[0002] The field of diffusion film technology involves films made of specific materials to adjust the propagation characteristics of light and achieve uniform distribution of light. The film is usually made of polymers or other light-transmitting materials, and contains particles or structures on the surface or in the body for scattering light. The key function of the diffusion film is to improve the uniformity of the light source, reduce light spots and shadows, and make the light distribution smoother and wider. This technology is widely used in backlight units (BLU) and various lighting systems, especially in liquid crystal display (LCD) technology, photography, film and television production and medical lighting, where diffusion films play a vital role.

[0003] Among them, the main use of the light homogenization method based on the diffusion film is to improve the light quality in applications that require uniform lighting, such as in backlit displays, lighting devices and other equipment that requires fine light control. By precisely controlling the material composition and structural layout of the diffusion film, the light can be effectively dispersed, bright spots and shadows can be eliminated, thereby providing a softer and more uniform visual experience.

[0004] Traditional methods lack the ability to monitor and dynamically adjust the propagation characteristics of light in real time. In the design of diffusion films, static material composition and structural layout are used, resulting in the inability to maintain consistency in light quality under different ambient light conditions or usage scenarios. This static design limits its adaptability, especially in scenarios with extremely high requirements for light uniformity. Traditional methods cannot adjust the light source configuration to adapt to light changes, such as changes in ambient light conditions, resulting in reduced visual experience and energy efficiency. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a light homogenization method and system based on a diffusion film.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a light homogenization method based on a diffusion film, comprising the following steps:

[0007] S1: Based on a high-speed camera, continuously capture the light distribution image of the diffusion film, perform real-time decoding and transmittance analysis on the image to obtain preliminary image data, and perform scattering angle detection and spot analysis on the preliminary image data to obtain the original image analysis record;

[0008] S2: Based on the original image analysis record, a convolutional neural network is applied to perform feature recognition on the light distribution, analyze the light spot, shadow and light density caused by the diffusion film, obtain the light uniformity index, and adjust the light source brightness and material structure according to the index to obtain the light source optimization configuration;

[0009] S3: based on the optimized configuration of the light source, re-capturing the image, performing brightness contrast analysis and scattering angle detection on the newly captured diffusion film image, and comparing it with the original light uniformity index to obtain adjustment demand assessment information;

[0010] S4: Based on the adjustment demand assessment information, terminal testing is performed to evaluate the uniformity of light distribution, and performance data before and after adjustment is recorded. Through data comparison results, optimization and fine-tuning of the optical properties of the diffusion film are performed to obtain optical performance adjustment results.

[0011] The present invention is improved in that the step of obtaining the preliminary image data is specifically as follows:

[0012] S111: Based on a high-speed camera, a sequence of images of the light distribution of the diffuser film is continuously captured using the formula:

[0013] ;

[0014] Calculate the image average light intensity data ,in, For location The average light intensity value, For the Frame The light intensity value at the position, is the total number of frames, is the square root weighting parameter, is the normalization coefficient;

[0015] S112: Average light intensity data for the image To perform real-time decoding, apply the decoding function:

[0016] ;

[0017] Calculate the decoded image data ,in, and is the adjustment parameter in the decoding process, is the absolute value term weight parameter, is the average light intensity threshold;

[0018] S113: Calculate the transmittance of each pixel using a transmittance analysis method, using the formula:

[0019] ;

[0020] in, For location The light transmittance, is the decoded image data, for The maximum value of .

[0021] The present invention is improved in that the steps of obtaining the original image analysis record are specifically as follows:

[0022] S121: Using the preliminary image data, perform light spot positioning and calculate the geometric center of each light spot using the formula:

[0023] ;

[0024] Get the spot center data, where is the center coordinate of the light spot, is the light intensity value of the pixel point in the spot area, are pixel coordinates, and is the offset from the center of the light spot, is the calculation window, is the weight coefficient within the window;

[0025] S122: Use the spot center data to calculate the scattering angle using the formula:

[0026] ;

[0027] Generate scattering angle data, where represents the scattering angle, is the distance from the center of the spot to the center of the image, is the reference distance, It is spot weight;

[0028] S123: combining the scattering angle data and the light spot center data, analyzing the shape, size and scattering characteristics of the light spot, and obtaining an original image analysis record.

[0029] The present invention is improved in that the step of obtaining the light uniformity index is specifically as follows:

[0030] S211: Based on the original image analysis record, apply the convolutional neural network to extract light feature data, including spot size, shadow area and light density, and calculate the average value of the extracted features and standard deviation , using the formula:

[0031] and ;

[0032] Generates statistical summary data, where For the The measured value of a light characteristic, is the number of features, is the feature mean, is the characteristic standard deviation;

[0033] S212: Using the mean and standard deviation, combined with the shape of the light distribution, the formula is:

[0034] ;

[0035] in, is the light uniformity index, and are the standard deviation and mean value calculated from the light feature data, is the skewness of the light feature, and is the adjustment parameter.

[0036] The present invention is improved in that the step of obtaining the optimal configuration of the light source is specifically as follows:

[0037] S221: judging whether the demand index needs to be adjusted according to the light uniformity index, and judging whether the demand index needs to be adjusted = ( < )like < If yes, then yes, otherwise no;

[0038] S222: Calculate new light source brightness and material structure adjustment parameters according to adjustment requirements, using the formula:

[0039] and ;

[0040] Get the adjusted light source brightness and material structure data ,in, and is the reference light source brightness and material structure setting, 、 and is the adjustment coefficient, which is used to determine the adjustment sensitivity and response mode. is the threshold value, is the light uniformity index;

[0041] S223: Using the obtained light source brightness and material structure data , adjust the light source and obtain the optimal configuration of the light source.

[0042] The present invention is improved in that the steps of brightness contrast analysis and scattering angle detection are specifically as follows:

[0043] S311: Based on the optimized configuration of the light source, recapture the diffusion film image and perform brightness and contrast analysis using the formula:

[0044] and ;

[0045] Calculate the average brightness of an image and contrast index , generates brightness and contrast analysis results, where The first The brightness value of a pixel, It is with The weight associated with each pixel, is the total number of pixels, is a small constant;

[0046] S312: Detect the scattering angle of the newly captured image using the formula:

[0047] ;

[0048] Calculate the average deviation of the weighted scatter angle , generate the scattering angle detection result, where It is The scattering angle of each detection point is It is with The weight associated with each pixel, is the weighted average of the scattering angles.

[0049] The present invention is improved in that the step of comparing the original light uniformity index is specifically as follows:

[0050] S321: Summarize the detection result data of brightness, contrast and scattering angle, and create a data analysis record including brightness, contrast and scattering angle;

[0051] S322: Determine the degree of improvement of the characteristics of the new and old images based on the data analysis records, using the formula:

[0052] ;

[0053] Obtain adjustment needs assessment information, including: and is the brightness value of the new and old images, and is the contrast index between the new and old images, and is the scattering angle deviation between the new and old images, , and is the weight coefficient.

[0054] The present invention is improved in that the step of obtaining the optical performance adjustment result is specifically:

[0055] S411: Based on the adjustment demand assessment information, a baseline optical performance test is performed, the light distribution uniformity, brightness and scattering angle associated with the test are recorded, and the uniformity of the initial light distribution is calculated using the formula:

[0056] ;

[0057] Get initial optical performance evaluation results ,in, is an indicator of the uniformity of light distribution. It is The light intensity at each measuring point, is the maximum light intensity in the measurement point, is the average value of the normalized light intensity, is the total number of measurement points;

[0058] S412: Based on the initial optical performance evaluation results , determine the optical characteristic area that needs to be improved, adjust the light source brightness and the diffusion film structure, and obtain the adjusted information of the light source and the diffusion film;

[0059] S413: Re-perform the optical performance test, collect the adjusted light distribution data, and calculate the uniformity of the light distribution. The formula is:

[0060] ;

[0061] Get the adjusted optical performance evaluation results , compare the data before and after adjustment, determine whether the adjustment has achieved the expected goal, and obtain the optical performance adjustment results.

[0062] A light homogenization system based on a diffusion film, the system comprising:

[0063] The image capture module uses a high-speed camera to continuously capture the light distribution image of the diffusion film, transmit the image data in real time, and synchronize the image capture and data transmission to generate an image data queue;

[0064] The image decoding and analysis module extracts images from the image data queue for decoding, analyzes the light transmittance in the image, performs scattering angle and light spot feature detection, and integrates image feature data to generate a preliminary image analysis record;

[0065] The feature recognition and adjustment module applies a convolutional neural network to analyze light distribution, including light spots, shadows, and light density, based on the preliminary image analysis records, calculates a light uniformity index according to the analysis results, and adjusts the light source brightness and material structure according to the index to generate an optimized light source configuration;

[0066] The performance test module recaptures and analyzes the diffusion film image based on the light source optimization configuration, performs brightness and contrast analysis, re-detects the scattering angle, compares the updated light uniformity index with the old index, and generates adjustment demand assessment information;

[0067] The optical performance fine-tuning module performs terminal testing of optical characteristics according to the adjustment demand evaluation information, evaluates the uniformity of light distribution, records the performance data before and after the adjustment, compares the data, and generates an optical performance adjustment result.

[0068] Compared with the prior art, the advantages and positive effects of the present invention are:

[0069] In the present invention, the ability to evaluate light scattering characteristics is significantly improved by capturing and decoding light distribution images in real time. The measurement accuracy of light uniformity is improved by deeply analyzing the light spots, shadows and light density caused by the diffusion film, and dynamic adjustment of the light source configuration based on measured data is realized. This real-time feedback mechanism allows instant optimization of light source brightness and material structure, achieving more precise light control, thereby improving the uniformity of light distribution. In addition, continuous monitoring of adjustment effects ensures long-term light output quality and provides a more optimized visual experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A flow chart of a light homogenization method based on a diffusion film is proposed for the present invention;

[0071] Figure 2 A schematic diagram of a process for obtaining preliminary image data in a light homogenization method based on a diffusion film is provided for the present invention;

[0072] Figure 3 A schematic diagram of the acquisition process of the original image analysis record in a light homogenization method based on a diffusion film is proposed in the present invention;

[0073] Figure 4 A schematic diagram of a process for obtaining a light uniformity index in a light uniformity method based on a diffusion film is provided in the present invention;

[0074] Figure 5 A schematic diagram of a process for obtaining an optimal configuration of a light source in a light homogenization method based on a diffusion film is provided in the present invention;

[0075] Figure 6A schematic flow chart of brightness contrast analysis and scattering angle detection in a light homogenization method based on a diffusion film is proposed for the present invention;

[0076] Figure 7 A schematic diagram of a comparison process of a light uniformity method based on a diffusion film and an original light uniformity index in the present invention;

[0077] Figure 8 The present invention provides a schematic diagram of a process for obtaining optical performance adjustment results in a light homogenization method based on a diffusion film. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0079] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0080] Example: See Figure 1 The present invention provides a technical solution: a light homogenization method based on a diffusion film, comprising the following steps:

[0081] S1: Based on a high-speed camera, continuously capture the light distribution image of the diffusion film, perform real-time decoding and transmittance analysis on the image to obtain preliminary image data, and perform scattering angle detection and spot analysis on the preliminary image data to obtain the original image analysis record;

[0082] S2: Based on the original image analysis records, a convolutional neural network is applied to perform feature recognition on the light distribution, analyze the light spots, shadows and light density caused by the diffusion film, obtain the light uniformity index, and adjust the light source brightness and material structure according to the index to obtain the optimal configuration of the light source;

[0083] S3: Based on the optimized configuration of the light source, recapture the image, perform brightness contrast analysis and scattering angle detection on the newly captured diffusion film image, and compare it with the original light uniformity index to obtain adjustment demand assessment information;

[0084] S4: Based on the adjustment demand assessment information, perform terminal testing, evaluate the uniformity of light distribution, record the performance data before and after adjustment, and compare the data to perform long-term optimization and fine-tune the optical properties of the diffusion film to obtain the optical performance adjustment results.

[0085] The original image analysis records include light scattering indicators, spot size, and scattering angle data. The light source optimization configuration includes brightness adjustment parameters, light source position settings, and optical material characteristics. The adjustment needs evaluation information includes adjustment indicators, performance comparison data, and optimization needs list. The optical performance adjustment results specifically refer to the uniformity improvement index, performance stability, and optimized transmittance.

[0086] See also Figure 2 , the steps for obtaining preliminary image data are as follows:

[0087] S111: Based on a high-speed camera, a sequence of images of the light distribution of the diffuser film is continuously captured using the formula:

[0088] ;

[0089] Calculate the image average light intensity data ,in, For location The average light intensity value, For the Frame The light intensity value at the position, is the total number of frames, is the square root weighting parameter, which improves the dynamic range of the data. is the normalization coefficient, which maintains data stability;

[0090] S112: Average light intensity data for the image To perform real-time decoding, apply the decoding function:

[0091] ;

[0092] Calculate the decoded image data ,in, and is the adjustment parameter in the decoding process, is the absolute value weighting parameter, which is used to enhance the sensitivity to sudden changes in light intensity. is the average light intensity threshold, used to adjust the decoding sensitivity;

[0093] S113: Calculate the transmittance of each pixel using a transmittance analysis method, using the formula:

[0094] ;

[0095] in, For location The light transmittance, is the decoded image data, for The maximum value of .

[0096] Assume the following data:

[0097] Assume there are 3 frames of data The light intensity at a certain pixel is ;

[0098] set up .

[0099] Apply the formula to each frame of data:

[0100] ;

[0101] ;

[0102] ;

[0103] Calculate the average:

[0104] ;

[0105] Assume the following data:

[0106] From the above ;

[0107] set up ;

[0108] Application formula:

[0109] ;

[0110] ;

[0111] ;

[0112] Assume the following data:

[0113] From the above , assuming .

[0114] Application formula:

[0115] .

[0116] See also Figure 3 , the steps for obtaining the original image analysis record are as follows:

[0117] S121: Use the preliminary image data to locate the light spot and calculate the geometric center of each light spot using the formula:

[0118] ;

[0119] Get the spot center data, where is the center coordinate of the light spot, is the light intensity value of the pixel point in the spot area, are pixel coordinates, and is the offset relative to the center of the spot, is the calculation window, is the weight coefficient within the window;

[0120] S122: Use the spot center data to calculate the scattering angle using the formula:

[0121] ;

[0122] Generate scattering angle data, where represents the scattering angle, is the distance from the center of the spot to the center of the image, is the reference distance, such as the image radius, It is spot weights to enhance the impact of important spots;

[0123] S123: combining the scattering angle data and the light spot center data, analyzing the shape, size and scattering characteristics of the light spot, and obtaining an original image analysis record.

[0124] Assume the following data:

[0125] Assume a simple 3x3 window with each weight being 1 and the center of the window being the center of the spot:

[0126] Weight Matrix for:

[0127] ;

[0128] The light intensity value matrix is:

[0129] ;

[0130] Calculate the center spot intensity:

[0131] ;

[0132] Maximum This is what we want, determine the center of the light spot.

[0133] Assume the following data:

[0134] Assume there are three light spots, with distances of 10, 15, and 20 pixels respectively, and weights They are 1.0, 0.8, 0.6 respectively, reference distance Pixel.

[0135] Calculate the scattering angle:

[0136] ;

[0137] ;

[0138] .

[0139] See also Figure 4 , the steps to obtain the light uniformity index are as follows:

[0140] S211: Based on the original image analysis record, apply convolutional neural network to extract light feature data, including spot size, shadow area and light density, and calculate the average value of the extracted features and standard deviation , using the formula:

[0141] and ;

[0142] Generates statistical summary data, where For the The measured value of a light characteristic, is the number of features, is the feature mean, is the characteristic standard deviation;

[0143] S212: Using the mean and standard deviation, combined with the shape of the light distribution, the formula is:

[0144] ;

[0145] in, is the light uniformity index, and are the standard deviation and mean value calculated from the light feature data, is the skewness of the light feature, providing additional shape information of the light distribution, and It is to adjust the parameters to optimize the model's sensitivity to the shape of the data distribution.

[0146] Assume the following data:

[0147] Data Collection:

[0148] Assume that n=5 light intensity measurements are extracted from the original image, which are .

[0149] Calculate the average ::

[0150] ;

[0151] Calculate standard deviation :

[0152] First calculate each Square the difference from the mean, then sum and take the square root

[0153] ;

[0154] ;

[0155] Assume the following data:

[0156] Setting adjustment parameters and , assuming that the skewness of the light distribution , in a real scenario, the measured .

[0157] Calculate the exponential part:

[0158] ;

[0159] Use approximation ,when When I was very young, .

[0160] Calculate the denominator:

[0161] ;

[0162] ;

[0163] Final calculation:

[0164] .

[0165] See also Figure 5 , the specific steps for obtaining the light source optimization configuration are:

[0166] S221: Based on the light uniformity index, determine whether the demand index needs to be adjusted, and determine whether adjustment is needed = ( < )like < If yes, then yes, otherwise no;

[0167] S222: Calculate new light source brightness and material structure adjustment parameters according to adjustment requirements, using the formula:

[0168] and ;

[0169] Get the adjusted light source brightness and material structure data ,in, and is the reference light source brightness and material structure setting, 、 and is the adjustment coefficient, which is used to determine the adjustment sensitivity and response mode. is the threshold value, is the light uniformity index;

[0170] S223: Using the obtained light source brightness and material structure data , adjust the light source and obtain the optimal configuration of the light source.

[0171] Assume the following data:

[0172] set up (Threshold) is 80;

[0173] Assume that the current (Light uniformity index) is 70;

[0174] set up (Current light source brightness benchmark) is 1000lumen;

[0175] set up (Current material structure benchmark) is 500 units;

[0176] Adjustment factor =0.05, =0.03;

[0177] Assumptions =2 is the nonlinear response index.

[0178] Determine whether to adjust:

[0179] because (Threshold) is 80, (Light Uniformity Index) is 70, which is less than 80, so the result is "Yes", so light source optimization adjustment is required.

[0180] Light source brightness adjustment:

[0181] ;

[0182] Substituting in the assumed values:

[0183] ;

[0184] calculate:

[0185] ;

[0186] ;

[0187] Material structure adjustment:

[0188] ;

[0189] Insert hypothetical values:

[0190] ;

[0191] calculate:

[0192] ;

[0193] .

[0194] See also Figure 6 , the steps of brightness contrast analysis and scattering angle detection are as follows:

[0195] S311: Based on the optimized light source configuration, recapture the diffuser image and perform brightness and contrast analysis using the formula:

[0196] and ;

[0197] Calculate the average brightness of an image and contrast index , generates brightness and contrast analysis results, where The first The brightness value of a pixel, It is with The weight associated with each pixel, is the total number of pixels, is a small constant used to prevent division by zero;

[0198] S312: Detect the scattering angle of the newly captured image using the formula:

[0199] ;

[0200] Calculate the average deviation of the weighted scatter angle , generate the scattering angle detection result, where It is The scattering angle of each detection point is It is with The weight associated with each pixel is assigned according to the importance of the detection point. is the weighted average of the scattering angles.

[0201] Assume the following data:

[0202] Assume there are 5 pixels with brightness values ;

[0203] The corresponding weight ;

[0204] is a small constant used to avoid zero denominator.

[0205] brightness Calculation:

[0206] Weighted brightness sum:

[0207] ;

[0208] ;

[0209] Sum of weights:

[0210] ;

[0211] Weighted average brightness :

[0212] ;

[0213] Contrast Calculation:

[0214] Maximum brightness ;

[0215] Minimum brightness min .

[0216] Contrast :

[0217] ;

[0218] Assume data:

[0219] Scattering Angle ;

[0220] Use the same weight .

[0221] Weighted average scattering angle :

[0222] Weighted sum of scatter angles:

[0223] ;

[0224] ;

[0225] Sum of weights:

[0226] ;

[0227] Weighted average scattering angle :

[0228] ;

[0229] Weighted average scattering angle deviation :

[0230] Sum of scattering angle deviations:

[0231] ;

[0232] ;

[0233] ;

[0234] Weighted average scattering angle deviation :

[0235] .

[0236] See also Figure 7 , and the comparison steps with the original light uniformity index are as follows:

[0237] S321: Summarize the detection result data of brightness, contrast and scattering angle, and create a data analysis record including brightness, contrast and scattering angle;

[0238] S322: Determine the degree of improvement of the characteristics of the new and old images based on the data analysis records, using the formula:

[0239] ;

[0240] Obtain adjustment needs assessment information, including: and is the brightness value of the new and old images, and is the contrast index between the new and old images, and is the scattering angle deviation between the new and old images, , and are weight coefficients that affect the importance of brightness, contrast, and scattering angle in the overall evaluation.

[0241] Assume the following data:

[0242] =220 and =200;

[0243] =1.1 and =1.0;

[0244] =5° and =6°;

[0245] =0.3, =0.5, = 0.2, indicating the importance of brightness, contrast, and scattering angle in the overall evaluation.

[0246] Calculate the brightness ratio:

[0247] ;

[0248] Calculate the contrast ratio:

[0249] ;

[0250] Calculate the scattering angle ratio:

[0251] ;

[0252] Integration ratio and weight:

[0253] Weighted sum = + + :

[0254] ;

[0255] Calculate the final comparison result:

[0256] .

[0257] See also Figure 8 , the steps for obtaining the optical performance adjustment result are as follows:

[0258] S411: Based on the adjustment demand assessment information, perform a baseline optical performance test, record the light distribution uniformity, brightness, and scattering angle associated with the test, and calculate the uniformity of the initial light distribution using the formula:

[0259] ;

[0260] Get initial optical performance evaluation results ,in, is an indicator of the uniformity of light distribution. It is The light intensity at each measurement point is measured by an optical sensor. is the maximum light intensity at the measurement point, which is used to normalize the light intensity at each measurement point. is the average value of the normalized light intensity, is the total number of measurement points;

[0261] S412: Based on the initial optical performance evaluation results , determine the optical characteristic area that needs to be improved, adjust the light source brightness and the diffusion film structure, and obtain the adjusted information of the light source and the diffusion film;

[0262] S413: Re-perform the optical performance test, collect the adjusted light distribution data, and recalculate the uniformity of the light distribution. The formula is:

[0263] ;

[0264] Get the adjusted optical performance evaluation results , compare the data before and after adjustment, determine whether the adjustment has achieved the expected goal, and obtain the optical performance adjustment results.

[0265] Assume the following data:

[0266] Consider the following light intensity data points (in lumens):

[0267] Measuring point light intensity ;

[0268] Calculate the maximum light intensity .

[0269] Normalized light intensity calculation:

[0270] Normalized light intensity ;

[0271] Normalized value

[0272] Calculate the mean of the normalized light intensity:

[0273] Average light intensity ;

[0274] Calculate the uniformity index using the formula

[0275] Uniformity index calculation:

[0276] ;

[0277] ;

[0278] A light homogenization system based on a diffusion film, the system comprising:

[0279] The image capture module uses a high-speed camera to continuously capture the light distribution image of the diffusion film, transmit the image data in real time, and synchronize the image capture and data transmission to generate an image data queue;

[0280] The image decoding and analysis module extracts images from the image data queue for decoding, analyzes the light transmittance in the image, performs scattering angle and light spot feature detection, and integrates the image feature data to generate a preliminary image analysis record;

[0281] The feature recognition and adjustment module uses a convolutional neural network to analyze light distribution, including light spots, shadows, and light density, based on the preliminary image analysis records. It calculates the light uniformity index based on the analysis results, adjusts the light source brightness and material structure based on the index, and generates an optimized light source configuration.

[0282] The performance test module recaptures and analyzes the diffuser image based on the light source optimization configuration, performs brightness and contrast analysis, re-detects the scattering angle, compares the updated light uniformity index with the old index, and generates adjustment needs assessment information;

[0283] The optical performance fine-tuning module performs terminal testing of optical characteristics based on the adjustment demand assessment information, evaluates the uniformity of light distribution, records the performance data before and after adjustment, and compares the data to generate the optical performance adjustment results.

[0284] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A light homogenization method based on a diffusion film, characterized in that: The following steps are involved: Based on a high-speed camera, continuously capture the light distribution image of the diffusion film, perform real-time decoding and transmittance analysis on the image to obtain preliminary image data, and perform scattering angle detection and spot analysis on the preliminary image data to obtain the original image analysis record; Based on the original image analysis record, a convolutional neural network is applied to perform feature recognition on the light distribution, analyze the light spots, shadows and light density caused by the diffusion film, obtain the light uniformity index, and adjust the light source brightness and material structure according to the index to obtain the light source optimization configuration; Based on the light source optimization configuration, recapture the image, perform brightness contrast analysis and scattering angle detection on the newly captured diffusion film image, and compare it with the original light uniformity index to obtain adjustment demand assessment information; Based on the adjustment demand assessment information, terminal testing is performed to evaluate the uniformity of light distribution, and the performance data before and after the adjustment is recorded. Through data comparison results, long-term optimization and fine-tuning of the optical properties of the diffusion film are performed to obtain optical performance adjustment results.

2. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps for obtaining the preliminary image data are specifically as follows: Based on a high-speed camera, the image sequence of the light distribution of the diffusion film is continuously captured, using the formula: ； Calculate the image average light intensity data ,in, For location The average light intensity value, For the Frame The light intensity value at the position, is the total number of frames, is the square root weighting parameter, is the normalization coefficient; Average light intensity data for the image To perform real-time decoding, apply the decoding function: ； Calculate the decoded image data ,in, and is the adjustment parameter in the decoding process, is the absolute value term weight parameter, is the average light intensity threshold; Use the transmittance analysis method to calculate the transmittance of each pixel. The formula is: ； in, For location The light transmittance, is the decoded image data, for The maximum value of .

3. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps for obtaining the original image analysis record are specifically as follows: The preliminary image data is used to locate the light spot and calculate the geometric center of each light spot using the formula: ； Get the spot center data, where is the center coordinate of the light spot, is the light intensity value of the pixel point in the spot area, are pixel coordinates, and is the offset from the center of the light spot, is the calculation window, is the weight coefficient within the window; Using the spot center data, calculate the scattering angle using the formula: ； Generate scattering angle data, where represents the scattering angle, is the distance from the center of the spot to the center of the image, is the reference distance, It is spot weight; Combine the scattering angle data and the spot center data to analyze the shape, size and scattering characteristics of the spot and obtain the original image analysis record.

4. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps of obtaining the light uniformity index are specifically as follows: Based on the original image analysis record, a convolutional neural network is applied to extract light feature data, including spot size, shadow area and light density, and the average value of the extracted features is calculated and standard deviation , using the formula: ； Generates statistical summary data, where For the The measured value of a light characteristic, is the number of features, is the feature mean, is the characteristic standard deviation; Using the mean and standard deviation, combined with the shape of the light distribution, the formula is used: ； in, is the light uniformity index, and are the standard deviation and mean value calculated from the light characteristic data, is the skewness of the light feature, and is the adjustment parameter.

5. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps for obtaining the optimal configuration of the light source are specifically as follows: According to the light uniformity index, it is determined whether the demand index needs to be adjusted. If < If yes, then adjustment is needed; otherwise, no adjustment is needed. According to the adjustment requirements, calculate the new light source brightness and material structure adjustment parameters using the formula: ； Get the adjusted light source brightness and material structure data ,in, and is the reference light source brightness and material structure setting, , and is the adjustment coefficient, which is used to determine the adjustment sensitivity and response mode. is the threshold value, is the light uniformity index; Using the brightness of the light source and material structure data , adjust the light source and obtain the optimal configuration of the light source.

6. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps of brightness contrast analysis and scattering angle detection are specifically as follows: Based on the optimized light source configuration, the diffuser image was recaptured and analyzed for brightness and contrast using the formula: ； Calculate the average brightness of an image and contrast index , generates brightness and contrast analysis results, where The first The brightness value of a pixel, It is with The weight associated with each pixel, is the total number of pixels, is a small constant; The scattering angle detection is performed on the newly captured image using the formula: ； Calculate the average deviation of the weighted scatter angle , generate the scattering angle detection result, where It is The scattering angle of each detection point is It is with The weight associated with each pixel, is the weighted average of the scattering angles.

7. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The step of comparing the original light uniformity index is specifically as follows: Summarize the test result data of brightness, contrast and scattering angle, and establish data analysis records including brightness, contrast and scattering angle; Based on the data analysis records, the degree of improvement of the characteristics of the new and old images is determined using the formula: ； Obtain adjustment needs assessment information, including: and is the brightness value of the new and old images, and is the contrast index between the new and old images, and is the scattering angle deviation between the new and old images, , and is the weight coefficient.

8. The light homogenization method based on a diffusion film according to claim 1, characterized in that: The steps of obtaining the optical performance adjustment result are specifically as follows: Based on the adjustment requirement assessment information, a baseline optical performance test is performed, and the light distribution uniformity, brightness, and scattering angle associated with the test are recorded. The uniformity of the initial light distribution is calculated using the formula: ； Get initial optical performance evaluation results ,in, is an indicator of the uniformity of light distribution. It is The light intensity at each measuring point, is the maximum light intensity in the measurement point, is the average value of the normalized light intensity, is the total number of measurement points; Based on the initial optical performance evaluation results , determine the optical characteristic area that needs to be improved, adjust the light source brightness and the diffusion film structure, and obtain the adjusted information of the light source and the diffusion film; Re-test the optical performance, collect the adjusted light distribution data, and calculate the uniformity of the light distribution. The formula is: ； Get the adjusted optical performance evaluation results , compare the data before and after adjustment, determine whether the adjustment has achieved the expected goal, and obtain the optical performance adjustment results.

9. A light homogenization system based on a diffusion film, characterized in that: According to any one of claims 1 to 8, the light homogenization method based on a diffusion film is performed, and the system comprises: The image capture module uses a high-speed camera to continuously capture the light distribution image of the diffusion film, transmit the image data in real time, and synchronize the image capture and data transmission to generate an image data queue; The image decoding and analysis module extracts images from the image data queue for decoding, analyzes the light transmittance in the image, performs scattering angle and light spot feature detection, and integrates image feature data to generate a preliminary image analysis record; The feature recognition and adjustment module applies a convolutional neural network to analyze light distribution, including light spots, shadows, and light density, based on the preliminary image analysis records, calculates a light uniformity index according to the analysis results, and adjusts the light source brightness and material structure according to the index to generate an optimized light source configuration; The performance test module recaptures and analyzes the diffusion film image based on the light source optimization configuration, performs brightness and contrast analysis, re-detects the scattering angle, compares the updated light uniformity index with the old index, and generates adjustment demand assessment information; The optical performance fine-tuning module performs terminal testing of optical characteristics according to the adjustment demand evaluation information, evaluates the uniformity of light distribution, records the performance data before and after the adjustment, compares the data, and generates an optical performance adjustment result.

Images