A method and system for detecting the defect of uneven brightness of a liquid crystal panel

Abstract

The present invention relates to the technical field of liquid crystal panel defect detection, and specifically discloses a detection method and system for the brightness non-uniformity defect of a liquid crystal panel. A photometer is used to obtain an image of the liquid crystal panel in the display mode, and a high-precision sensor is used to capture the brightness information. An image segmentation algorithm is adopted, combining adaptive threshold segmentation, deep learning semantic segmentation, and region growing algorithm, to accurately identify and extract the problem areas of brightness non-uniformity, analyze the brightness difference degree, spatial distribution, and temporal stability respectively, and calculate the corresponding characteristic indices. By synthesizing these characteristic indices, a comprehensive score of each brightness non-uniform area is generated through standardization calculation, and the areas are divided into minor, medium, and severe defects. The present invention combines real-time monitoring, can dynamically adjust the detection time and parameter settings, performs intelligent detection for different types of brightness non-uniformity defects, optimizes the detection efficiency and accuracy, and avoids redundant detection.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid crystal panel defect detection, and particularly relates to a method and system for detecting the brightness non-uniformity defect of a liquid crystal panel. Background Art

[0002] Liquid crystal panels are widely used in devices such as televisions, monitors, and smartphones, and their display effects directly affect the user experience. During the production process of liquid crystal panels, brightness non-uniformity is one of the common defects. The brightness non-uniformity defect may be caused by problems with the quality of liquid crystal materials, unstable panel manufacturing processes, improper control circuit designs, etc. This defect not only affects the display effect but may also shorten the service life of the liquid crystal panel. To ensure the display quality and consistency of liquid crystal panels, accurate detection of brightness non-uniformity defects is crucial. Therefore, how to effectively identify and classify different types of brightness non-uniformity defects has become an important issue in the field of liquid crystal panel production and quality control.

[0003] The existing technologies have the following deficiencies:

[0004] Currently, most of the detection methods for brightness non-uniformity defects of liquid crystal panels are based on threshold judgments of static brightness differences, lacking a comprehensive analysis of multi-dimensional features of the brightness non-uniform regions. These methods only focus on brightness differences and ignore the spatial distribution of defects, the temporal stability of brightness changes, and the complexity of defect regions. In addition, existing detection systems usually use fixed detection times and parameter settings and cannot dynamically adjust the detection strategy according to different types of defects. Therefore, when encountering complex brightness non-uniformity problems, traditional methods cannot effectively cope, resulting in insufficient detection accuracy and common phenomena of misjudgment and missed judgment. Moreover, in the case of large-scale production and high-frequency detection, fixed detection parameters will lead to redundant detection processes, increasing the detection time and reducing the efficiency. These deficiencies seriously affect the accuracy of liquid crystal panel quality control and production efficiency, and there is an urgent need for an intelligent detection method that can dynamically adapt to different types of brightness non-uniformity defects. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for detecting the brightness non-uniformity defect of a liquid crystal panel to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for detecting the brightness non-uniformity defect of a liquid crystal panel, comprising the following steps:

[0008] S1: Use a photometer to obtain an image of the liquid crystal panel in the display mode, and preprocess the image to reduce noise and optimize the contrast;

[0009] S2: Identify and isolate the regions with brightness non-uniformity problems through an image segmentation algorithm, analyze the brightness difference degree, spatial distribution, and temporal stability of the regions with brightness non-uniformity problems respectively. According to the analysis results, calculate the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the regions with brightness non-uniformity problems;

[0010] S3: Conduct a comprehensive analysis on the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the extracted regions with brightness non-uniformity problems. According to the analysis results, generate a comprehensive score for each region with brightness non-uniformity;

[0011] S4: According to the comprehensive score of each region with brightness non-uniformity, divide each region with brightness non-uniformity into minor defects, medium defects, and severe defects;

[0012] S5: Combine real-time monitoring, dynamically adjust the detection time and parameter settings of different types of defects, and achieve intelligent detection for different types of brightness non-uniformity defects.

[0013] As a further solution of the present invention: The identifying and isolating the regions with brightness non-uniformity problems through an image segmentation algorithm specifically includes:

[0014] Through an adaptive threshold segmentation method, determine the brightness threshold based on the brightness distribution of the image, and extract the regions with brightness differences from the image;

[0015] Process the image through a semantic segmentation network based on deep learning, use a convolutional neural network to identify the regions with brightness non-uniformity, and the convolutional neural network is trained through a labeled data set to learn the brightness characteristics of the liquid crystal panel for region identification;

[0016] Adopt a region growing algorithm, dynamically expand the region boundary according to the brightness difference of adjacent pixels, and extract the regions with brightness non-uniformity problems.

[0017] As a further solution of the present invention: The analyzing the brightness difference degree, spatial distribution, and temporal stability of the regions with brightness non-uniformity problems respectively specifically includes:

[0018] Analyze the brightness difference degree of the regions with brightness non-uniformity problems, analyze the brightness difference degree of the regions with brightness non-uniformity problems, and calculate the brightness difference characteristic index according to the amplitude of the brightness fluctuation of the regions with brightness non-uniformity problems;

[0019] Analyze the spatial distribution of the regions with brightness non-uniformity problems, extract the geometric characteristics of the spatial distribution, including: the area, perimeter, length, and width of the regions with brightness non-uniformity problems, and calculate the spatial distribution characteristic index according to the abnormality degree of the spatial distribution geometric characteristics;

[0020] Analyze the temporal stability of the luminance non-uniformity problem area, analyze the stability of the luminance change in the luminance non-uniformity problem area, and calculate the temporal stability characteristic index according to the analysis result, so as to evaluate the temporal stability of the luminance non-uniformity problem area.

[0021] As a further solution of the present invention: the process of obtaining the luminance difference characteristic index is as follows:

[0022] Apply discrete wavelet transform to the preprocessed image to decompose the image into a low-frequency part and a high-frequency part, specifically including:

[0023] Perform multi-level wavelet decomposition on the image to obtain the low-frequency component and high-frequency component of each level;

[0024] The calculation expression for wavelet decomposition of each level is: ; ;

[0025] Among them, is the low-frequency component, is the high-frequency component, and are the low-pass and high-pass filters, represents the original image, represents the pixel coordinates in the image;

[0026] Calculate the luminance difference of each high-frequency component, and the calculation expression is: ;

[0027] Among them, represents the pixel value of the high-frequency component, represents the mean value of the high-frequency component, represents the pixel in the high-frequency component, represents the total number of pixels in the high-frequency component;

[0028] Based on the calculated luminance difference, calculate the luminance difference characteristic index of each high-frequency component, and the calculation expression is: ;

[0029] Among them, is the total number of high-frequency subbands analyzed, represents the high-frequency subband, represents the th high-frequency subband's luminance difference, represents the luminance difference characteristic index of the high-frequency component;

[0030] Integrate the luminance difference characteristic indexes of the wavelet decomposition results of all levels to obtain the final luminance difference characteristic index, and the calculation expression is: ;

[0031] In the formula, Indicates the total number of wavelet decomposition layers, Indicates the wavelet decomposition layer, Indicates the brightness difference feature index.

[0032] As a further solution of the present invention: The acquisition process of the spatial distribution feature index is as follows:

[0033] Extract the brightness non-uniform area, regard each pixel as a node in the graph, and connect adjacent pixels with edges. The weight of the edge is represented by the pixel brightness difference, specifically including:

[0034] Each edge in the graph connects two adjacent pixels, and the weight of the edge Indicates the pixel and the pixel The brightness difference between them, and the calculation expression is: ;

[0035] Among them, Indicates the node The brightness value of the corresponding pixel, Indicates the node The brightness value of the corresponding pixel;

[0036] Extract the geometric features of the brightness non-uniform area, specifically including:

[0037] Calculate the area of the area through the number of nodes in the brightness non-uniform area in the graph. The calculation expression is: ;

[0038] Among them, Is the number of nodes in the graph, Indicates the area of the brightness non-uniform area;

[0039] Calculate the perimeter of the brightness non-uniform area. The calculation expression is: ;

[0040] Among them, Indicates the perimeter of the brightness non-uniform area, Indicates the edge set in the graph, Indicates the indicator function. If the two nodes connected by the edge belong to the inside and outside of the area respectively, then , otherwise ;

[0041] Calculate the aspect ratio of the brightness non-uniform area. The calculation expression is: ;

[0042] Among them, Indicates the length of the circumscribed rectangle, Indicates the width of the circumscribed rectangle, Indicates the aspect ratio of the brightness non-uniform area;

[0043] Based on the geometric features of the luminance non-uniformity region, comprehensively calculate the spatial distribution characteristic index, and the calculation expression is: ;

[0044] In the formula, represents the spatial distribution characteristic index.

[0045] As a further solution of the present invention: the acquisition process of the time stability characteristic index is as follows:

[0046] Obtain the luminance values of the luminance non-uniformity problem region at multiple time points to form a time series: ;

[0047] Among them, represents the acquisition time point, represents the length of the time series;

[0048] Convert the time series into a frequency-domain signal , and the calculation expression is: ;

[0049] Among them, represents the frequency-domain signal, represents the imaginary unit, represents the natural logarithm of the base, represents the frequency component;

[0050] Calculate the amplitude spectrum of the frequency-domain signal, and the calculation expression is: ;

[0051] Among them, represents the amplitude corresponding to the frequency component in the frequency domain;

[0052] Find the main frequency and the frequency energy of the luminance change by analyzing the amplitude spectrum;

[0053] Among them, the main frequency is the frequency component corresponding to the maximum amplitude in the amplitude spectrum, denoted as ;

[0054] Calculate the total energy of all frequency components in the amplitude spectrum to obtain the total frequency energy, and the calculation expression is: ;

[0055] Among them, represents the total frequency energy, represents the maximum frequency of the frequency component;

[0056] Calculate the time stability characteristic index according to the calculation process of the total frequency energy and the main frequency. The calculation expression is as follows: ;

[0057] Wherein, represents the time stability characteristic index, represents the time series of the maximum amplitude, represents the main frequency.

[0058] As a further solution of the present invention: comprehensively analyze the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index of the extracted brightness non-uniform problem area, and generate a comprehensive score for each brightness non-uniform area according to the analysis results, specifically including:

[0059] Obtain the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index of the brightness non-uniform problem area, perform normalization calculation processing on the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index, and calculate the comprehensive score of the brightness non-uniform area through the comprehensive score calculation expression;

[0060] Wherein, the comprehensive score calculation expression is as follows: ;

[0061] Wherein, represents the value after standardizing the spatial distribution characteristic index, represents the value after standardizing the brightness difference characteristic index, represents the value after standardizing the time stability characteristic index, represents the comprehensive score.

[0062] As a further solution of the present invention: divide each brightness non-uniform area into minor defects, medium defects and severe defects according to the comprehensive score of each brightness non-uniform area, specifically including:

[0063] Calculate the comprehensive score of each brightness non-uniform area respectively according to the comprehensive score calculation expression, compare the comprehensive score of each brightness non-uniform area with the first threshold. If the comprehensive score is greater than or equal to the first threshold, it means that the defect state of the corresponding area is minor, and it is recorded as a minor defect;

[0064] If the comprehensive score is less than the first threshold, then compare the comprehensive score with the second threshold. If the comprehensive score is less than or equal to the second threshold, it means that the defect state of the corresponding area is severe, and it is recorded as a severe defect;

[0065] If the comprehensive score is greater than the second threshold and less than the first threshold, it means that the defect state of the corresponding area is medium, and it is recorded as a medium defect;

[0066] Among them, the value of the first threshold is greater than the value of the second threshold.

[0067] As a further solution of the present invention: combining real-time monitoring, dynamically adjusting the detection time and parameter settings of different types of defects, and realizing intelligent detection for different types of brightness non-uniformity defects, specifically including:

[0068] By real-time monitoring the change of the brightness non-uniform area of the liquid crystal panel, classifying the defect types according to the brightness difference characteristics, spatial distribution characteristics and time stability characteristics;

[0069] When the system detects that the brightness non-uniformity defect of the liquid crystal panel in the area is a serious defect, automatically extend the detection time and increase the detection accuracy; for the brightness non-uniformity defect of the liquid crystal panel being a minor defect, the system automatically shortens the detection time and reduces redundant detection. When the system detects that the brightness non-uniformity defect of the liquid crystal panel in the area is a medium defect, the detection is carried out according to the original detection time.

[0070] A detection system for the brightness non-uniformity defect of a liquid crystal panel, including:

[0071] A data acquisition module, which uses a photometer to obtain an image of the liquid crystal panel in the display mode and preprocesses the image to reduce noise and optimize the contrast;

[0072] A region recognition module, which recognizes and separates the region containing the brightness non-uniformity problem through an image segmentation algorithm;

[0073] A region feature analysis module, which respectively analyzes the brightness difference degree, spatial distribution and time stability of the brightness non-uniformity problem region, and calculates the spatial distribution feature index, brightness difference feature index and time stability feature index of the brightness non-uniformity problem region according to the analysis results;

[0074] A comprehensive analysis module, which comprehensively analyzes the spatial distribution feature index, brightness difference feature index and time stability feature index of the extracted brightness non-uniformity problem region, and generates a comprehensive score for each brightness non-uniform region according to the analysis results;

[0075] A defect recognition and classification module, which classifies each brightness non-uniform region into minor defects, medium defects and serious defects according to the comprehensive score of each brightness non-uniform region;

[0076] A dynamic detection module, which combines real-time monitoring, dynamically adjusts the detection time and parameter settings of different types of defects, and realizes intelligent detection for different types of brightness non-uniformity defects.

[0077] The beneficial effects of the present invention:

[0078] (1) By comprehensively analyzing the multi-dimensional characteristics of the brightness non-uniformity defects of the liquid crystal panel (including the spatial distribution characteristic index, the brightness difference characteristic index, and the time stability characteristic index), the present invention realizes the precise identification, classification, and quantitative evaluation of the problem areas of brightness non-uniformity. Different from the traditional detection methods that only rely on a single brightness difference index, the present invention provides a comprehensive and dynamic analysis framework by combining multiple indexes, such as the geometric characteristics of spatial distribution (area, perimeter, aspect ratio) and the frequency characteristics of brightness fluctuations (such as the amplitude of brightness difference and time stability). This multi-dimensional evaluation method can not only capture the characteristics of various brightness non-uniformity defects more meticulously, but also conduct more accurate classification and scoring according to their time stability, spatial complexity, and brightness change degree. Through the comprehensive analysis of these indexes, the generated comprehensive score can clearly distinguish minor, medium, and severe defects, effectively improving the accuracy and robustness of the detection, and significantly reducing the false positives and false negatives caused by single-dimensional analysis. This method greatly improves the precision of the quality evaluation of liquid crystal panels and is particularly suitable for the automated quality control of large-scale production lines.

[0079] (2) By combining real-time monitoring with intelligent algorithms, the present invention can dynamically adjust the detection time and parameter settings according to the actual situation of the problem areas of brightness non-uniformity, and intelligently handle different types of brightness non-uniformity defects. Different from the traditional methods that rely on fixed detection cycles and parameter settings, the present invention automatically optimizes the detection strategy by real-time analyzing the spatial characteristics, brightness change amplitude, and time stability of the brightness non-uniform areas of the liquid crystal panel. When the system identifies relatively severe brightness non-uniformity defects, it automatically extends the detection time and improves the detection accuracy to ensure that such defects are fully identified; for relatively minor defects, the system automatically shortens the detection time and adjusts the detection sensitivity to avoid redundant detection, significantly improving the detection efficiency. Through this intelligent dynamic adjustment strategy, the present invention can not only optimize the detection process, making the detection work more flexible, fast, and accurate, but also effectively solve the problems of long detection time and low detection accuracy in traditional methods. Especially in the face of large-scale production and high-frequency detection, the solution of the present invention can significantly improve the production efficiency and the precision of quality control, providing a more efficient and intelligent quality detection solution in a rapidly changing manufacturing environment. Description of the Drawings

[0080] The present invention will be further described below with reference to the accompanying drawings.

[0081] Figure 1 is the specific step flow block diagram of a detection method for brightness non-uniformity defects of a liquid crystal panel according to the present invention;

[0082] Figure 2 is the flow block diagram of a detection system for brightness non-uniformity defects of a liquid crystal panel according to the present invention. Detailed implementation manners

[0083] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0084] Please refer to Figure 1 As shown, the present invention is a method for detecting the uneven brightness defect of a liquid crystal panel, including the following steps:

[0085] S1: Use a photometer to obtain an image of the liquid crystal panel in the display mode, and preprocess the image to reduce noise and optimize contrast;

[0086] S2: Identify and separate the regions with uneven brightness problems through an image segmentation algorithm, analyze the brightness difference degree, spatial distribution, and temporal stability of the regions with uneven brightness problems respectively. According to the analysis results, calculate the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the regions with uneven brightness problems;

[0087] S3: Comprehensively analyze the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the extracted regions with uneven brightness problems. According to the analysis results, generate a comprehensive score for each region with uneven brightness;

[0088] S4: According to the comprehensive score of each region with uneven brightness, divide each region with uneven brightness into minor defects, medium defects, and severe defects;

[0089] S5: Combine real-time monitoring, dynamically adjust the detection time and parameter settings for different types of defects, and realize intelligent detection for different types of uneven brightness defects.

[0090] In S1, use a photometer to obtain an image of the liquid crystal panel in the display mode, and preprocess the image to reduce noise and optimize contrast, which specifically includes:

[0091] First, use a photometer to obtain the image data of the liquid crystal panel in a specific display mode. The photometer captures the brightness information of the liquid crystal panel through a high-precision sensor to obtain the light intensity data of each pixel. To ensure the quality and accuracy of the obtained image, the photometer needs to set appropriate measurement parameters when performing image acquisition, including sampling frequency, exposure time, and image acquisition area. By adjusting these parameters, ensure that the captured image reflects the true brightness distribution of the liquid crystal panel in the display mode.

[0092] The collected images are processed using image preprocessing techniques to reduce noise and optimize the contrast of the images, ensuring more accurate subsequent analysis. The specific steps include:

[0093] Noise removal: In the images of liquid crystal panels, noise may originate from device interference or environmental factors. Therefore, first, a filtering method is used to suppress noise in the images, and the denoising technique uses Gaussian filtering. Gaussian filtering smooths the image by weighted averaging of surrounding pixels, reducing the impact of random noise on the image.

[0094] Contrast optimization: There are differences in the brightness distribution of liquid crystal panel images, making it difficult to identify details in some areas. Therefore, the contrast of the images is enhanced through histogram equalization technology, making the brightness value distribution more uniform, thereby enhancing the visibility of areas with uneven brightness in the images. Histogram equalization optimizes the contrast of the images by readjusting the brightness value range of the images, expanding the brightness difference, and ensuring that areas with uneven brightness can be clearly displayed.

[0095] In S2, the regions containing uneven brightness problems are identified and separated through an image segmentation algorithm. The degree of brightness difference, spatial distribution, and temporal stability of the regions with uneven brightness problems are analyzed respectively. According to the analysis results, the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the regions with uneven brightness problems are calculated, specifically including:

[0096] The identification and separation of the regions containing uneven brightness problems through the image segmentation algorithm specifically include:

[0097] Through the adaptive threshold segmentation method, the brightness threshold is determined based on the brightness distribution of the image, and the regions with brightness differences are extracted from the image;

[0098] The image is processed using a semantic segmentation network based on deep learning. A convolutional neural network is used to identify the regions with uneven brightness. The convolutional neural network is trained with a labeled dataset to learn the brightness characteristics of the liquid crystal panel for region identification;

[0099] The region growing algorithm is adopted to dynamically expand the region boundary according to the brightness difference of adjacent pixels, and the regions with uneven brightness problems are extracted.

[0100] The degree of brightness difference, spatial distribution, and temporal stability of the regions with uneven brightness problems are analyzed respectively, specifically including:

[0101] Analyze the degree of brightness difference of the regions with uneven brightness problems, calculate the brightness difference characteristic index according to the amplitude of the brightness fluctuation of the regions with uneven brightness problems;

[0102] Analyze the spatial distribution of the brightness non-uniformity problem areas, extract the geometric features of the spatial distribution, including: the area, perimeter, length, and width of the brightness non-uniformity problem areas, and calculate the spatial distribution characteristic index according to the abnormality degree of the geometric features of the spatial distribution;

[0103] Analyze the temporal stability of the brightness non-uniformity problem areas, analyze the stability of the brightness change of the brightness non-uniformity problem areas, and calculate the temporal stability characteristic index according to the analysis results, so as to evaluate the temporal stability of the brightness non-uniformity problem areas.

[0104] The acquisition process of the brightness difference characteristic index is as follows:

[0105] Apply discrete wavelet transform to the preprocessed image to decompose the image into a low-frequency part and a high-frequency part, specifically including:

[0106] Perform multi-level wavelet decomposition on the image to obtain the low-frequency component and the high-frequency component of each level;

[0107] The calculation expression of wavelet decomposition of each level is: ; ;

[0108] Among them, is the low-frequency component, is the high-frequency component, and are the low-pass and high-pass filters, represents the original image, represents the pixel coordinates in the image;

[0109] Calculate the brightness difference of each high-frequency component, and the calculation expression is: ;

[0110] Among them, represents the pixel value of the high-frequency component, represents the mean value of the high-frequency component, represents the pixel in the high-frequency component, represents the total number of pixels in the high-frequency component;

[0111] Based on the calculated brightness difference, calculate the brightness difference characteristic index of each high-frequency component, and the calculation expression is: ;

[0112] Among them, is the total number of high-frequency subbands analyzed, represents the high-frequency subband, represents the th brightness difference of the high-frequency subband, represents the brightness difference characteristic index of the high-frequency component;

[0113] The brightness difference feature indices of the wavelet decomposition results at all levels are synthesized to obtain the final brightness difference feature index, and the calculation expression is: ;

[0114] In the formula, represents the total number of wavelet decomposition levels, represents the wavelet decomposition level, represents the brightness difference feature index.

[0115] The process of obtaining the spatial distribution feature index is as follows:

[0116] Extract the brightness non-uniform area, regard each pixel as a node in the graph, and connect adjacent pixels with an edge. The weight of the edge is represented by the pixel brightness difference, specifically including:

[0117] Each edge in the graph connects two adjacent pixels, and the weight of the edge represents the brightness difference between pixel and pixel , and the calculation expression is: ;

[0118] Among them, represents the brightness value of the pixel corresponding to node , represents the brightness value of the pixel corresponding to node ;

[0119] Extract the geometric features of the brightness non-uniform area, specifically including:

[0120] Calculate the area of the area through the number of nodes in the brightness non-uniform area in the graph, and the calculation expression is: ;

[0121] Among them, is the number of nodes in the graph, represents the area of the brightness non-uniform area;

[0122] Calculate the perimeter of the brightness non-uniform area, and the calculation expression is: ;

[0123] Among them, represents the perimeter of the brightness non-uniform area, represents the edge set in the graph, represents the indicator function. If the two nodes connected by the edge belong to the inside and outside of the area respectively, then , otherwise ;

[0124] Calculate the aspect ratio of the brightness non-uniform area, and the calculation expression is: ;

[0125] Among them, represents the length of the circumscribed rectangle, represents the width of the circumscribed rectangle, represents the aspect ratio of the length and width of the luminance non-uniformity region;

[0126] According to the geometric characteristics of the luminance non-uniformity region, comprehensively calculate the spatial distribution characteristic index, and the calculation expression is: ;

[0127] In the formula, represents the spatial distribution characteristic index.

[0128] The acquisition process of the time stability characteristic index is as follows:

[0129] Obtain the luminance values of the luminance non-uniformity problem region at multiple time points to form a time series: ;

[0130] Among them, represents the acquisition time point, represents the length of the time series;

[0131] Transform the time series into a frequency-domain signal , and the calculation expression is: ;

[0132] Among them, represents the frequency-domain signal, represents the imaginary unit, represents the natural logarithm of the base, represents the frequency component;

[0133] Calculate the amplitude spectrum of the frequency-domain signal, and the calculation expression is: ;

[0134] Among them, represents the amplitude corresponding to the frequency component in the frequency domain;

[0135] Find the main frequency and the frequency energy of the luminance change by analyzing the amplitude spectrum;

[0136] Among them, the main frequency is the frequency component corresponding to the maximum amplitude in the amplitude spectrum, denoted as ;

[0137] Calculate the total energy of all frequency components in the amplitude spectrum to obtain the total frequency energy, and the calculation expression is: ;

[0138] Among them, represents the total frequency energy, represents the maximum frequency of the frequency component;

[0139] According to the calculation and processing of the total frequency energy and the main frequency, calculate the time stability characteristic index, and the calculation expression is: ;

[0140] where, represents the time stability characteristic index, represents the time series of the maximum amplitude, represents the main frequency.

[0141] In S3, comprehensively analyze the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index of the extracted brightness unevenness problem area. According to the analysis results, generate a comprehensive score for each brightness unevenness area, specifically including:

[0142] Obtain the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index of the brightness unevenness problem area, perform normalization calculation and processing on the spatial distribution characteristic index, brightness difference characteristic index and time stability characteristic index, and calculate the comprehensive score of the brightness unevenness area through the comprehensive score calculation expression;

[0143] where, the comprehensive score calculation expression is: ;

[0144] where, represents the value after standardizing the spatial distribution characteristic index, represents the value after standardizing the brightness difference characteristic index, represents the value after standardizing the time stability characteristic index, represents the comprehensive score.

[0145] According to the comprehensive score calculation expression, calculate the comprehensive score for each brightness unevenness problem area respectively, and perform comprehensive score marking.

[0146] It should be noted that: the comprehensive score reflects the degree of uneven defect of each brightness unevenness problem area, and the higher the value of the comprehensive score, the lower the degree of uneven defect of the corresponding brightness unevenness problem area.

[0147] In S4, according to the comprehensive score of each brightness unevenness area, divide each brightness unevenness area into minor defects, medium defects and serious defects, specifically including:

[0148] Calculate the comprehensive score of each brightness non-uniform area according to the comprehensive score calculation expression, and compare the comprehensive score of each brightness non-uniform area with the first threshold. If the comprehensive score is greater than or equal to the first threshold, it indicates that the defect state of the corresponding area is minor, and it is recorded as a minor defect;

[0149] If the comprehensive score is less than the first threshold, then compare the comprehensive score with the second threshold. If the comprehensive score is less than or equal to the second threshold, it indicates that the defect state of the corresponding area is serious, and it is recorded as a serious defect;

[0150] If the comprehensive score is greater than the second threshold and less than the first threshold, it indicates that the defect state of the corresponding area is medium, and it is recorded as a medium defect;

[0151] Among them, the value of the first threshold is greater than the value of the second threshold.

[0152] It should be noted that: the first standard threshold represents the lower limit of the area defect state, the second standard threshold represents the upper limit of the area defect state, and the area defect state between the first standard threshold and the second standard threshold is a medium defect. The selection of the first standard threshold and the second standard threshold is based on the statistical analysis of historical data and is set according to the actual needs of the system and the error tolerance.

[0153] In S5, combined with real-time monitoring, dynamically adjust the detection time and parameter settings of different types of defects to achieve intelligent detection of different types of brightness non-uniform defects, specifically including:

[0154] By real-time monitoring the change of the brightness non-uniform area of the liquid crystal panel, classify the defect types according to the brightness difference characteristics, spatial distribution characteristics and time stability characteristics.

[0155] For different types of brightness non-uniform defects, adjust the detection time and detection parameters in real time, and automatically optimize the detection strategy by analyzing the complexity and stability of the defect area to improve the detection efficiency and accuracy.

[0156] Dynamically adjust the time window and sensitivity parameters during the detection process according to the characteristics of each brightness non-uniform defect, specifically including: when the system detects that the brightness non-uniform defect of the liquid crystal panel in the area is a serious defect, automatically extend the detection time and increase the detection accuracy; for the brightness non-uniform defect of the liquid crystal panel being a minor defect, the system automatically shortens the detection time and reduces redundant detection. When the system detects that the brightness non-uniform defect of the liquid crystal panel in the area is a medium defect, the detection is carried out according to the original detection time.

[0157] Please refer to Figure 2 As shown, a detection system for brightness non-uniform defects of a liquid crystal panel includes:

[0158] A data acquisition module, which uses a photometer to obtain an image of the liquid crystal panel in the display mode and preprocesses the image to reduce noise and optimize contrast;

[0159] A region recognition module, which recognizes and separates the regions with brightness non-uniformity problems through an image segmentation algorithm;

[0160] A region feature analysis module, which analyzes the brightness difference degree, spatial distribution and temporal stability of the regions with brightness non-uniformity problems respectively. According to the analysis results, it calculates the spatial distribution feature index, brightness difference feature index and temporal stability feature index of the regions with brightness non-uniformity problems;

[0161] A comprehensive analysis module, which comprehensively analyzes the spatial distribution feature index, brightness difference feature index and temporal stability feature index of the extracted regions with brightness non-uniformity problems. According to the analysis results, it generates a comprehensive score for each region with brightness non-uniformity;

[0162] A defect identification and classification module, which classifies each region with brightness non-uniformity into minor defects, medium defects and severe defects according to the comprehensive score of each region with brightness non-uniformity;

[0163] A dynamic detection module, which combines real-time monitoring to dynamically adjust the detection time and parameter settings of different types of defects, and realizes intelligent detection for different types of brightness non-uniformity defects.

[0164] The working principle of the present invention: Image data is obtained in the display mode of the liquid crystal panel using a photometer, and the panel brightness information is captured through a high-precision sensor. Noise removal techniques (such as Gaussian filtering) and contrast optimization (such as histogram equalization) are used to preprocess the image to ensure the improvement of image quality and the clear display of regions with brightness non-uniformity. Through image segmentation algorithms (such as adaptive threshold segmentation, deep learning semantic segmentation and region growing algorithm), the regions with brightness non-uniformity problems are recognized and extracted, and their brightness differences, spatial distributions and temporal stabilities are analyzed respectively, and the spatial distribution feature index, brightness difference feature index and temporal stability feature index are calculated. Through the comprehensive analysis of these indices, a comprehensive score for each region with brightness non-uniformity is generated, and the regions are classified into minor defects, medium defects and severe defects according to the score.

[0165] The present invention also combines real-time monitoring means, capable of dynamically adjusting the detection time and parameter settings, and performing intelligent detection on different types of brightness non-uniformity defects. Specifically, the system classifies the defect types in real time according to the brightness difference characteristics, spatial distribution characteristics, and time stability characteristics. For severe defects, the detection time is automatically extended and the detection accuracy is improved; for minor defects, the detection time is automatically shortened to avoid redundant detection and improve the detection efficiency. Medium defects are processed according to the original detection time, thereby optimizing the detection process and enhancing the flexibility and intelligence level of the detection. Through this method, efficient detection and accurate classification of the brightness non-uniformity problem of the liquid crystal panel can be achieved, and the intelligent degree of product quality monitoring can be improved.

[0166] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0167] It should be understood that the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. Among them, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0168] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0169] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application.

Claims

1. A method for detecting the brightness non-uniformity defect of a liquid crystal panel, characterized in that, Including the following steps: S1: Use a photometer to obtain an image of the liquid crystal panel in the display mode, and preprocess the image to reduce noise and optimize contrast; S2: Identify and separate the regions with brightness non-uniformity problems through an image segmentation algorithm, analyze the brightness difference degree, spatial distribution, and temporal stability of the regions with brightness non-uniformity problems respectively. According to the analysis results, calculate the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the regions with brightness non-uniformity problems; The process of obtaining the brightness difference characteristic index is as follows: Apply discrete wavelet transform to the preprocessed image to decompose the image into a low-frequency part and a high-frequency part, specifically including: Perform multi-level wavelet decomposition on the image to obtain the low-frequency component and high-frequency component of each level; The calculation expression for each layer of wavelet decomposition is as follows: ; ; Among them, is the low-frequency component, is the high-frequency component, and are the low-pass and high-pass filters, represents the original image, represents the pixel coordinates in the image; Calculate the brightness difference of each high-frequency component, and the calculation expression is as follows: ; Among them, represents the pixel value of the high-frequency component, represents the mean value of the high-frequency component, represents the pixels in the high-frequency component, represents the total number of pixels in the high-frequency component; Based on the calculated brightness difference, calculate the brightness difference characteristic index of each high-frequency component, and the calculation expression is: ； Among them, is the total number of high-frequency subbands for analysis, represents a high-frequency subband, represents the luminance difference of the th high-frequency subband, and represents the luminance difference characteristic index of the high-frequency component; The luminance difference feature indices of the wavelet decomposition results at all levels are synthesized to obtain the final luminance difference feature index, and the calculation expression is as follows: ; In the formula, represents the total number of wavelet decomposition layers, represents the wavelet decomposition layer, represents the brightness difference feature index; S3: Conduct a comprehensive analysis of the spatial distribution characteristic index, brightness difference characteristic index, and temporal stability characteristic index of the extracted regions with brightness non-uniformity problems. According to the analysis results, generate a comprehensive score for each region with brightness non-uniformity; S4: According to the comprehensive score of each region with brightness non-uniformity, divide each region with brightness non-uniformity into minor defects, medium defects, and severe defects; S5: Combine real-time monitoring, dynamically adjust the detection time and parameter settings for different types of defects, and achieve intelligent detection for different types of brightness non-uniformity defects.

2. The detection method for the uneven brightness defect of a liquid crystal panel according to claim 1, wherein The specific process of identifying and separating the regions with brightness non-uniformity problems through the image segmentation algorithm includes: Determine the brightness threshold based on the brightness distribution of the image through the adaptive threshold segmentation method, and extract the regions with brightness differences from the image; Process the image through a semantic segmentation network based on deep learning, use a convolutional neural network to identify the regions with brightness non-uniformity, and the convolutional neural network is trained through a labeled data set to learn the brightness characteristics of the liquid crystal panel for region identification; Adopt the region growing algorithm, dynamically expand the region boundary according to the brightness difference of adjacent pixels, and extract the regions with brightness non-uniformity problems.

3. The detection method for the brightness non-uniformity defect of a liquid crystal panel according to claim 1, characterized in that The specific process of analyzing the brightness difference degree, spatial distribution, and temporal stability of the regions with brightness non-uniformity problems respectively includes: Analyze the brightness difference degree of the regions with brightness non-uniformity problems, analyze the brightness difference degree of the regions with brightness non-uniformity problems, and calculate the brightness difference characteristic index according to the amplitude of the brightness fluctuation of the regions with brightness non-uniformity problems; Analyze the spatial distribution of the regions with brightness non-uniformity problems, extract the geometric characteristics of the spatial distribution, including: the area, perimeter, length, and width of the regions with brightness non-uniformity problems, and calculate the spatial distribution characteristic index according to the abnormality degree of the spatial distribution geometric characteristics; Analyze the temporal stability of the regions with brightness non-uniformity problems, analyze the stability of the brightness change of the regions with brightness non-uniformity problems, and calculate the temporal stability characteristic index according to the analysis results, so as to evaluate the temporal stability of the regions with brightness non-uniformity problems.

4. The detection method for the uneven brightness defect of a liquid crystal panel according to claim 1, characterized in that, The process of obtaining the spatial distribution characteristic index is as follows: Extract the brightness non-uniformity region. Treat each pixel as a node in a graph, and connect adjacent pixels with edges. The weight of the edge is represented by the pixel brightness difference, specifically including: Each edge in the figure connects two adjacent pixels, and the weight of the edge represents the pixel and the pixel The brightness difference between them is calculated by the expression: ; Among them, represents a node corresponding to the brightness value of the pixel, represents a node corresponding to the brightness value of the pixel; Extract the geometric features of the brightness non-uniformity region, specifically including: Calculate the area of the region by the number of nodes in the uneven brightness region in the figure. The calculation formula is: ; Among them, is the number of nodes in the figure, represents the area of the luminance non-uniformity region; Calculate the perimeter of the luminance non-uniformity region, and the calculation expression is: ; Among them, represents the perimeter of the luminance non-uniformity region, represents the edge set in the figure, represents the indicator function. If the two nodes connected by the edge respectively belong to the inside and outside of the region, then , otherwise ; Calculate the aspect ratio of the luminance non-uniform area, and the calculation expression is as follows: ; Among them, represents the length of the circumscribed rectangle, represents the width of the circumscribed rectangle, represents the aspect ratio of the length and width of the luminance non-uniformity region; According to the geometric features of the brightness non-uniformity region, comprehensively calculate the spatial distribution feature index. The calculation expression is: ； In the formula, represents the spatial distribution characteristic index.

5. The detection method for the uneven brightness defect of a liquid crystal panel according to claim 4, characterized in that The acquisition process of the time stability feature index is as follows: Obtain the brightness values of the brightness non-uniformity problem region at multiple time points to form a time series: ； Among them, represents the acquisition time point, represents the length of the time series; Convert the time series through Fourier transform into a frequency-domain signal , and the calculation formula is: ； Among them, represents the frequency-domain signal, represents the imaginary unit, represents the natural logarithm, represents the frequency component; Calculate the amplitude spectrum of the frequency-domain signal, and the calculation expression is as follows: ; Among them, represents the amplitude corresponding to the frequency component in the frequency domain; Find the main frequency and the frequency energy of the brightness change by analyzing the amplitude spectrum; Among them, the main frequency is the frequency component corresponding to the maximum amplitude in the amplitude spectrum, denoted as ; Calculate the total energy of all frequency components in the amplitude spectrum to obtain the total frequency energy. The calculation expression is: ； Among them, represents the total frequency energy, represents the maximum frequency of the frequency component; According to the calculation and processing of the total frequency energy and the main frequency, calculate the time stability feature index. The calculation expression is: ； Among them, represents the time stability characteristic index, represents the time series with the maximum amplitude, represents the main frequency.

6. The detection method for the uneven brightness defect of a liquid crystal panel according to claim 1, characterized in that Comprehensively analyze the spatial distribution feature index, brightness difference feature index, and time stability feature index of the extracted brightness non-uniformity problem region. According to the analysis results, generate a comprehensive score for each brightness non-uniformity region, specifically including: Obtain the spatial distribution feature index, brightness difference feature index, and time stability feature index of the brightness non-uniformity problem region. Perform normalization calculation and processing on the spatial distribution feature index, brightness difference feature index, and time stability feature index, and calculate the comprehensive score of the brightness non-uniformity region through the comprehensive score calculation expression; Among them, the comprehensive score calculation expression is: ； Among them, represents the value after standardizing the spatial distribution feature index, represents the value after standardizing the brightness difference feature index, represents the value after standardizing the time stability feature index, represents the comprehensive score.

7. A method for detecting the defect of uneven brightness of a liquid crystal panel according to claim 1, characterized in that, According to the comprehensive score of each brightness non-uniformity region, divide each brightness non-uniformity region into minor defects, medium defects, and severe defects, specifically including: Calculate the comprehensive score of each brightness non-uniformity region according to the comprehensive score calculation expression, and compare the comprehensive score of each brightness non-uniformity region with the first threshold. If the comprehensive score is greater than or equal to the first threshold, it means that the defect state of the corresponding region is minor, and it is recorded as a minor defect; If the comprehensive score is less than the first threshold, then compare the comprehensive score with the second threshold. If the comprehensive score is less than or equal to the second threshold, it means that the defect state of the corresponding region is severe, and it is recorded as a severe defect; If the comprehensive score is greater than the second threshold and less than the first threshold, it means that the defect state of the corresponding region is medium, and it is recorded as a medium defect; Among them, the value of the first threshold is greater than the value of the second threshold.

8. The detection method for the defect of uneven brightness of a liquid crystal panel according to claim 1, characterized in that Combined with real-time monitoring, dynamically adjust the detection time and parameter settings of different types of defects to achieve intelligent detection of different types of brightness non-uniformity defects, specifically including: Through real-time monitoring of the change of the brightness non-uniformity region of the liquid crystal panel, classify the defect types according to the brightness difference feature, spatial distribution feature, and time stability feature; When the system detects that the brightness non-uniformity defect of the liquid crystal panel in the region is a severe defect, automatically extend the detection time and increase the detection accuracy; for the brightness non-uniformity defect of the liquid crystal panel that is a minor defect, the system automatically shortens the detection time and reduces redundant detection. When the system detects that the brightness non-uniformity defect of the liquid crystal panel in the region is a medium defect, then detect according to the original detection time.

9. A detection system for the defect of uneven brightness of a liquid crystal panel, characterized in that, For a method for detecting brightness non-uniformity defects of a liquid crystal panel as described in any one of claims 1-8, including: Data acquisition module, which uses a photometer to obtain images of the liquid crystal panel in the display mode and preprocesses the images to reduce noise and optimize contrast; Region recognition module, which identifies and separates the regions with brightness non-uniformity problems through an image segmentation algorithm; Region feature analysis module, which analyzes the brightness difference degree, spatial distribution, and temporal stability of the regions with brightness non-uniformity problems respectively. According to the analysis results, it calculates the spatial distribution feature index, brightness difference feature index, and temporal stability feature index of the regions with brightness non-uniformity problems; Comprehensive analysis module, which comprehensively analyzes the spatial distribution feature index, brightness difference feature index, and temporal stability feature index of the extracted regions with brightness non-uniformity problems. According to the analysis results, it generates a comprehensive score for each region with brightness non-uniformity; Defect identification and classification module, which classifies each region with brightness non-uniformity into minor defects, medium defects, and severe defects according to the comprehensive score of each region with brightness non-uniformity; Dynamic detection module, which combines real-time monitoring and dynamically adjusts the detection time and parameter settings for different types of defects to achieve intelligent detection for different types of brightness non-uniformity defects.

Images