A defect detection method and system for organic electroluminescent device

Abstract

The invention relates to the technical field of organic electroluminescent devices, and discloses a defect detection method and system for organic electroluminescent devices. The method comprises the following steps: photographing an organic electroluminescent device to be detected from multiple angles, processing the images from multiple angles, and obtaining a spliced ​​image to be detected from multiple angles; analyzing the spliced ​​image to be detected from multiple angles, and generating an analysis mark for the organic electroluminescent device to be detected from multiple angles; when a suspected defect mark is identified, dividing the spliced ​​image to be detected from multiple sub-defect detection areas, and calculating a first sub-defect detection coefficient and a second sub-defect detection coefficient of the spliced ​​image to be detected from multiple sub-defect detection areas; calculating a comprehensive defect detection coefficient based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, judging whether the organic electroluminescent device has a defect according to the comprehensive defect detection coefficient, realizing intelligent defect detection of the organic electroluminescent device, improving defect detection efficiency and accuracy, and further ensuring production efficiency and product quality.

Description

Technical Field

[0001] The present invention relates to the technical field of organic electroluminescent devices, and in particular to a defect detection method and system for organic electroluminescent devices. Background Art

[0002] Organic electroluminescent devices are a type of device that generates light through current drive and are widely used in the field of modern display technology. As an advanced display technology, organic electroluminescent devices are not only popular in the market due to their excellent display performance, but also face the challenges of continuous technological innovation and industrial upgrading. With the development of future technology and the expansion of applications, organic electroluminescent devices are expected to demonstrate their unique value in more fields.

[0003] Existing defect detection methods for organic electroluminescent devices mainly include visual inspection and electrical testing. Visual inspection relies on trained technicians to manually inspect the display screen using tools such as microscopes. Although some obvious defects can be found, it is inefficient and tends to miss minor defects. Electrical testing identifies defects by analyzing current and voltage characteristics, but this method usually requires special equipment and is difficult to achieve rapid and comprehensive inspection of the entire organic electroluminescent device, reducing defect detection efficiency. Summary of the invention

[0004] The embodiments of the present invention provide a defect detection method and system for an organic electroluminescent device, which realize intelligent defect detection of the organic electroluminescent device, improve defect detection efficiency and accuracy, and further ensure production efficiency and product quality.

[0005] In order to achieve the above object, the present invention provides a defect detection method for an organic electroluminescent device, comprising:

[0006] Determine an organic electroluminescent device to be detected for defects, photograph the organic electroluminescent device to be detected for defects at multiple angles to obtain multiple multi-angle images, and process the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected for defects;

[0007] Analyzing the spliced ​​image to be defect-detected, and generating analysis marks for the organic electroluminescent device to be defect-detected based on the analysis results, wherein the analysis marks include defect marks, qualified marks, and suspected defect marks;

[0008] When a suspected defect mark is identified, the defect detection spliced ​​image is divided into a plurality of sub-defect detection areas, and image analysis is performed on each sub-defect detection area, and a first sub-defect detection coefficient and a second sub-defect detection coefficient of the defect detection spliced ​​image are calculated based on the image analysis result;

[0009] A comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected is calculated based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and whether the organic electroluminescent device has a defect is determined according to the comprehensive defect detection coefficient.

[0010] Furthermore, when the multi-angle images are processed to obtain the spliced ​​image to be inspected for defects of the organic electroluminescent device to be inspected for defects, it includes:

[0011] Preprocessing the multi-angle images, wherein the preprocessing includes denoising, geometric correction and color balancing;

[0012] Extracting feature points from the preprocessed multi-angle images based on a first preset algorithm, matching the extracted feature points based on a second preset algorithm, and determining matching information between the multi-angle images, wherein the matching information includes relative positions and rotational relationships;

[0013] Based on the matching information, the multi-angle images are registered, and the registered multi-angle images are merged based on a third preset algorithm to obtain an initial defect detection spliced ​​image of the organic electroluminescent device to be detected;

[0014] The initial spliced ​​image to be defect-detected is optimized to obtain the spliced ​​image to be defect-detected of the organic electroluminescent device to be defect-detected, wherein the optimization includes removing splicing seams, adjusting image contrast, and sharpening image edges.

[0015] Furthermore, when analyzing the spliced ​​image to be inspected for defects and generating an analysis mark for the organic electroluminescent device to be inspected for defects based on the analysis result, it includes:

[0016] Acquire a qualified inspection stitching image corresponding to the stitching image to be inspected for defects;

[0017] Extracting all the pixels to be detected corresponding to the defect detection spliced ​​image, and extracting all the qualified pixels corresponding to the pixels to be detected from the qualified detection spliced ​​image;

[0018] Determine the pixel values ​​to be detected corresponding to all the pixels to be detected, and determine the qualified pixel values ​​corresponding to all the qualified pixels;

[0019] If the pixel values ​​to be detected are all equal to the qualified pixel values, a qualified mark is generated for the organic electroluminescent device to be defect-detected;

[0020] If the pixel values ​​to be detected are not equal to the qualified pixel values, a defect mark is generated for the organic electroluminescent device to be defect-detected;

[0021] If there are one or more pixel values ​​to be detected that are not equal to qualified pixel values, a suspected defect mark is generated for the organic electroluminescent device to be defect-detected.

[0022] Furthermore, when performing image analysis on each sub-defect detection area and calculating the first sub-defect detection coefficient and the second sub-defect detection coefficient of the spliced ​​image to be defect detected based on the image analysis result, it includes:

[0023] Analyze each sub-defect detection area, and if the pixel values ​​to be detected in the sub-defect detection area are all equal to the qualified pixel values, delete the corresponding sub-defect detection area;

[0024] Extract the remaining sub-defect detection areas and obtain the corresponding pixel values ​​to be analyzed;

[0025] Acquire a qualified pixel value to be analyzed corresponding to the pixel value to be analyzed, and if the pixel value to be analyzed is less than or equal to the qualified pixel value to be analyzed, calculate a first pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed, and divide the difference into a first pixel value set;

[0026] If the pixel value to be analyzed is greater than the qualified pixel value to be analyzed, a second pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed is calculated, and the pixel value is divided into a second pixel value set;

[0027] Performing curve fitting on the pixel values ​​to be analyzed in the first pixel value set to obtain a first pixel value curve;

[0028] Determine the slope corresponding to each first pixel difference on the first pixel value curve, determine the maximum slope a, and calculate the first calculation factor e according to the maximum slope a a ;

[0029] Normalize the first pixel value set, calculate the first variance of the first pixel value set after normalization, and calculate the first calculation factor e a and the first variance as the first defect value of the sub-defect detection area;

[0030] Performing curve fitting on the pixel values ​​to be analyzed in the second pixel value set to obtain a second pixel value curve;

[0031] Determine the slope corresponding to each second pixel difference on the second pixel value curve, determine the maximum slope b, and calculate the second calculation factor e according to the maximum slope b b ;

[0032] Normalize the second pixel value set, calculate the second variance of the normalized second pixel value set, and calculate the second calculation factor eb and the second variance as the second defect value of the sub-defect detection area;

[0033] Calculating a third defect value of the sub-defect detection area based on the first defect value and the second defect value;

[0034] The third defect values ​​of the remaining sub-defect detection areas are calculated, and the first sub-defect detection coefficient and the second sub-defect detection coefficient of the spliced ​​image to be defect-detected are calculated according to all the third defect values.

[0035] Further, when calculating the third defect value of the sub-defect detection area based on the first defect value and the second defect value, it includes:

[0036] The third defect value of the sub-defect detection area is calculated according to the following formula:

[0037] ;

[0038] Wherein, d is the third defect value of the sub-defect detection area, g1 is the first defect value, and g2 is the second defect value.

[0039] Further, when calculating the first sub-defect detection coefficient of the spliced ​​image to be defect-detected according to all the third defect values, it includes:

[0040] Extracting the same third defect value from all third defect values ​​and obtaining a plurality of defect value sequences;

[0041] Count the number of the first defect value sequence of the defect value sequence;

[0042] Extract a third defect value from each of all defect value sequences, and calculate the sum of the first defect values;

[0043] Obtaining a pre-calculated preset defect value, eliminating all defect value sequences that are less than the preset defect value, and counting the number of second defect value sequences of the remaining defect value sequences;

[0044] Extracting a defect value from each of the remaining defect value sequences and calculating a second defect value and value;

[0045] A first sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the number of the first defect value sequences, the number of the second defect value sequences, the first defect value sum, and the second defect value sum.

[0046] Further, when calculating the second sub-defect detection coefficient of the spliced ​​image to be defect-detected according to all the third defect values, it includes:

[0047] Calculate the average defect value of all third defect values, and generate a defect value series for all third defect values ​​that are less than or equal to the average defect value;

[0048] Generate a surplus defect value sequence for all third defect values ​​greater than the average defect value;

[0049] Calculate the difference between each third defect value in the defect value series and the average defect value, and generate a defect value difference series according to the principle of numerical value size;

[0050] Calculate the profit and defect value difference between each third defect value in the profit and defect value series and the average defect value, and generate a profit and defect value difference series according to the principle of numerical value size;

[0051] A second sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the defect value difference series and the excess defect value difference series.

[0052] Further, when calculating the second sub-defect detection coefficient of the spliced ​​image to be defect-detected according to the defect value difference series and the excess defect value difference series, it includes:

[0053] Determine the maximum defect value difference and the minimum defect value difference of the defect value difference sequence;

[0054] Calculating the defect difference between the maximum defect value difference and the minimum defect value difference;

[0055] Determine the maximum profit and defect value difference and the minimum profit and defect value difference of the profit and defect value difference series;

[0056] Calculating a margin value between the maximum margin-defect value difference and the minimum margin-defect value difference;

[0057] The second sub-defect detection coefficient of the defect detection spliced ​​image is calculated according to the following formula:

[0058] ;

[0059] Among them, k is the second sub-defect detection coefficient of the spliced ​​image to be defect detected, h1 is the deficit difference, h2 is the surplus difference, s1 is the weight corresponding to the deficit defect value difference series, n1 is the number of deficit defect value differences in the deficit defect value difference series, t i is the ith defect value difference in the defect value difference series, q i is the weight corresponding to the ith defect value difference, s2 is the weight corresponding to the surplus defect value difference series, n2 is the number of surplus defect value differences in the surplus defect value difference series, u j is the jth difference between the profit and defect values ​​in the difference series of profit and defect values, w jis the weight corresponding to the j-th profit and defect value difference.

[0060] Furthermore, when judging whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient, it includes:

[0061] Determining whether the organic electroluminescent device has a defect according to the relationship between the comprehensive defect detection coefficient and a preset comprehensive defect detection coefficient;

[0062] When the comprehensive defect detection coefficient is less than the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has no defects;

[0063] When the comprehensive defect detection coefficient is greater than or equal to the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has a defect.

[0064] In order to achieve the above object, the present invention also provides a defect detection system for an organic electroluminescent device, comprising:

[0065] An image processing module, used for determining an organic electroluminescent device to be detected for defects, photographing the organic electroluminescent device to be detected for defects at multiple angles to obtain multiple multi-angle images, and processing the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected for defects;

[0066] A mark generation module, used for analyzing the spliced ​​image to be defect-detected, and generating analysis marks for the organic electroluminescent device to be defect-detected based on the analysis results, wherein the analysis marks include defect marks, qualified marks and suspected defect marks;

[0067] a coefficient calculation module, for dividing the stitched image to be defect detected into a plurality of sub-defect detection areas when a suspected defect mark is identified, performing image analysis on each sub-defect detection area, and calculating a first sub-defect detection coefficient and a second sub-defect detection coefficient of the stitched image to be defect detected based on the image analysis result;

[0068] The defect detection module is used to calculate the comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and determine whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The present invention discloses a defect detection method and system for an organic electroluminescent device. The method comprises the following steps: photographing an organic electroluminescent device to be detected from multiple angles, processing the multiple angle images, and obtaining a spliced ​​image to be detected from multiple angles; analyzing the spliced ​​image to be detected from multiple angles, and generating an analysis mark for the organic electroluminescent device to be detected from multiple angles; when a suspected defect mark is identified, dividing the spliced ​​image to be detected from multiple defect sub-areas, and calculating a first sub-defect detection coefficient and a second sub-defect detection coefficient of the spliced ​​image to be detected from multiple defect sub-areas; calculating a comprehensive defect detection coefficient based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and judging whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient, thereby realizing intelligent defect detection of the organic electroluminescent device, improving defect detection efficiency and accuracy, and further ensuring production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0072] Figure 1 A schematic diagram showing a flow chart of a defect detection method for an organic electroluminescent device in an embodiment of the present invention is shown;

[0073] Figure 2 The schematic diagram of the structure of the defect detection system of the organic electroluminescent device in the embodiment of the present invention is shown. DETAILED DESCRIPTION

[0074] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0075] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0076] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0077] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0078] The following is a description of preferred embodiments of the present invention with reference to the accompanying drawings.

[0079] like Figure 1 As shown, an embodiment of the present invention discloses a defect detection method for an organic electroluminescent device, comprising:

[0080] S110: determining an organic electroluminescent device to be inspected for defects, photographing the organic electroluminescent device to be inspected for defects at multiple angles to obtain multiple multi-angle images, and processing the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be inspected for defects;

[0081] In some embodiments of the present application, when processing multi-angle images to obtain a spliced ​​image to be inspected for defects of the organic electroluminescent device to be inspected for defects, the process includes:

[0082] Preprocessing the multi-angle images, wherein the preprocessing includes denoising, geometric correction and color balancing;

[0083] Extracting feature points from the preprocessed multi-angle images based on a first preset algorithm, matching the extracted feature points based on a second preset algorithm, and determining matching information between the multi-angle images, wherein the matching information includes relative positions and rotational relationships;

[0084] Based on the matching information, the multi-angle images are registered, and the registered multi-angle images are merged based on a third preset algorithm to obtain an initial defect detection spliced ​​image of the organic electroluminescent device to be detected;

[0085] The initial spliced ​​image to be defect-detected is optimized to obtain the spliced ​​image to be defect-detected of the organic electroluminescent device to be defect-detected, wherein the optimization includes removing splicing seams, adjusting image contrast, and sharpening image edges.

[0086] In this embodiment, denoising is usually achieved by using methods such as Gaussian filtering or median filtering. Geometric correction is used to correct distortion in the image and ensure alignment between images. Color balancing adjusts the exposure and color distribution of the image to make them visually consistent, thereby eliminating unnecessary information that affects subsequent processing and improving image quality.

[0087] In this embodiment, the first preset algorithm includes SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features), and one of them can be used according to actual conditions. The second preset algorithm includes the RANSAC algorithm, which can determine the relative position and rotation relationship between images.

[0088] In this embodiment, multiple images are registered based on affine transformation, perspective transformation or nonlinear transformation model to eliminate factors such as translation, rotation, deformation and the like caused by different shooting angles between images, so as to achieve precise alignment between images.

[0089] In this embodiment, the third preset algorithm includes weighted mean, multi-band fusion, etc., and one of them can be adopted according to the actual situation to merge the aligned multi-angle images. During the merging process, adjustments are made based on the image exposure, color and other information to ensure color consistency and brightness balance.

[0090] In this embodiment, the initial defect detection stitching image is optimized to improve the overall visual effect and stitching quality of the image.

[0091] The beneficial effect of the above technical solution is that the present invention processes multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected, which can lay a reliable foundation for defect detection of the organic electroluminescent device to be detected.

[0092] S120: analyzing the spliced ​​image to be defect-detected, and generating an analysis mark for the organic electroluminescent device to be defect-detected based on the analysis result, wherein the analysis mark includes a defect mark, a qualified mark, and a suspected defect mark;

[0093] In some embodiments of the present application, when analyzing the spliced ​​image to be detected for defects and generating an analysis mark for the organic electroluminescent device to be detected for defects based on the analysis result, the method includes:

[0094] Acquire a qualified inspection stitching image corresponding to the stitching image to be inspected for defects;

[0095] Extracting all the pixels to be detected corresponding to the defect detection spliced ​​image, and extracting all the qualified pixels corresponding to the pixels to be detected from the qualified detection spliced ​​image;

[0096] Determine the pixel values ​​to be detected corresponding to all the pixels to be detected, and determine the qualified pixel values ​​corresponding to all the qualified pixels;

[0097] If the pixel values ​​to be detected are all equal to the qualified pixel values, a qualified mark is generated for the organic electroluminescent device to be defect-detected;

[0098] If the pixel values ​​to be detected are not equal to the qualified pixel values, a defect mark is generated for the organic electroluminescent device to be defect-detected;

[0099] If there are one or more pixel values ​​to be detected that are not equal to qualified pixel values, a suspected defect mark is generated for the organic electroluminescent device to be defect-detected.

[0100] In this embodiment, the qualified inspection spliced ​​image is obtained based on the qualified organic electroluminescent device.

[0101] In this embodiment, the pixels to be detected correspond to the qualified pixels one by one, thereby ensuring that the pixel values ​​to be detected correspond to the qualified pixel values ​​one by one.

[0102] The beneficial effect of the above technical solution is that the present invention can generate defect marks, qualified marks or suspected defect marks for the organic electroluminescent device to be detected according to the pixel value to be detected and the qualified pixel value. When a defect mark is generated, an alarm reminder is directly issued. When a qualified mark is generated, the next step of production is carried out. When a suspected defect mark is generated, further defect detection and analysis is performed to avoid the occurrence of defective organic electroluminescent devices.

[0103] S130: when a suspected defect mark is identified, the stitched image to be defect detected is divided into a plurality of sub-defect detection areas, image analysis is performed on each sub-defect detection area, and a first sub-defect detection coefficient and a second sub-defect detection coefficient of the stitched image to be defect detected are calculated based on the image analysis result;

[0104] In this embodiment, the stitched image to be detected is divided according to the size of the stitched image to be detected and the number of sub-defect detection areas that actually need to be divided to ensure the uniformity of each sub-defect detection area. Here, the number of sub-defect detection areas is preferably 50, which can be adjusted according to actual conditions.

[0105] In some embodiments of the present application, when performing image analysis on each sub-defect detection area and calculating the first sub-defect detection coefficient and the second sub-defect detection coefficient of the spliced ​​image to be defect detected based on the image analysis result, it includes:

[0106] Analyze each sub-defect detection area, and if the pixel values ​​to be detected in the sub-defect detection area are all equal to the qualified pixel values, delete the corresponding sub-defect detection area;

[0107] Extract the remaining sub-defect detection areas and obtain the corresponding pixel values ​​to be analyzed;

[0108] Acquire a qualified pixel value to be analyzed corresponding to the pixel value to be analyzed, and if the pixel value to be analyzed is less than or equal to the qualified pixel value to be analyzed, calculate a first pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed, and divide the difference into a first pixel value set;

[0109] If the pixel value to be analyzed is greater than the qualified pixel value to be analyzed, a second pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed is calculated, and the pixel value is divided into a second pixel value set;

[0110] Performing curve fitting on the pixel values ​​to be analyzed in the first pixel value set to obtain a first pixel value curve;

[0111] Determine the slope corresponding to each first pixel difference on the first pixel value curve, determine the maximum slope a, and calculate the first calculation factor e according to the maximum slope a a ;

[0112] Normalize the first pixel value set, calculate the first variance of the first pixel value set after normalization, and calculate the first calculation factor e a and the first variance as the first defect value of the sub-defect detection area;

[0113] Performing curve fitting on the pixel values ​​to be analyzed in the second pixel value set to obtain a second pixel value curve;

[0114] Determine the slope corresponding to each second pixel difference on the second pixel value curve, determine the maximum slope b, and calculate the second calculation factor e according to the maximum slope b b ;

[0115] Normalize the second pixel value set, calculate the second variance of the normalized second pixel value set, and calculate the second calculation factor e b and the second variance as the second defect value of the sub-defect detection area;

[0116] Calculating a third defect value of the sub-defect detection area based on the first defect value and the second defect value;

[0117] The third defect values ​​of the remaining sub-defect detection areas are calculated, and the first sub-defect detection coefficient and the second sub-defect detection coefficient of the spliced ​​image to be defect-detected are calculated according to all the third defect values.

[0118] In this embodiment, the pixel values ​​corresponding to the remaining sub-defect detection areas are used as the pixel values ​​to be analyzed.

[0119] In this embodiment, the curve fitting method and the method for determining the slope corresponding to each pixel difference are complicated and mature, and will not be introduced in detail here.

[0120] In this embodiment, the first calculation factor e a It means that the maximum slope a is the exponential power with e as the base, where e is equal to the base of the natural logarithm, and here it is preferably 2.71828.

[0121] In this embodiment, the second calculation factor e b It means that e is the base and the maximum slope b is the exponential power.

[0122] The beneficial effect of the above technical solution is: the present invention obtains the first pixel value curve and the second pixel value curve by fitting, and then determines the first defect value and the second defect value, and calculates the third defect value of the sub-defect detection area, which lays a foundation for the calculation of the first sub-defect detection coefficient and the second sub-defect detection coefficient, and ensures the calculation accuracy of the first sub-defect detection coefficient and the second sub-defect detection coefficient.

[0123] In some embodiments of the present application, when calculating the third defect value of the sub-defect detection area based on the first defect value and the second defect value, it includes:

[0124] The third defect value of the sub-defect detection area is calculated according to the following formula:

[0125] ;

[0126] Wherein, d is the third defect value of the sub-defect detection area, g1 is the first defect value, and g2 is the second defect value.

[0127] In some embodiments of the present application, when calculating the first sub-defect detection coefficient of the stitched image to be defect-detected according to all the third defect values, it includes:

[0128] Extracting the same third defect value from all third defect values ​​and obtaining a plurality of defect value sequences;

[0129] Count the number of the first defect value sequence of the defect value sequence;

[0130] Extract a third defect value from each of all defect value sequences, and calculate the sum of the first defect values;

[0131] Obtaining a pre-calculated preset defect value, eliminating all defect value sequences that are less than the preset defect value, and counting the number of second defect value sequences of the remaining defect value sequences;

[0132] Extracting a defect value from each of the remaining defect value sequences and calculating a second defect value and value;

[0133] A first sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the number of the first defect value sequences, the number of the second defect value sequences, the first defect value sum, and the second defect value sum.

[0134] In this embodiment, each defect value sequence includes the same third defect value, such as {6, 6, 6}, {8, 8}, {11, 11, 11}, etc.

[0135] In this embodiment, a third defect value is extracted from all defect value sequences respectively, that is, 6, 8, and 11 mentioned above.

[0136] In this embodiment, the preset defect value refers to the average value of all third defect values.

[0137] In this embodiment, if the preset defect value is 7, the remaining defect value sequences are {8, 8}, {11, 11, 11}, and one defect value, 8 and 11, is extracted from the remaining defect value sequences respectively.

[0138] In this embodiment, the above is shown by way of example, and the specific details shall be subject to the actual situation and are not specifically limited here.

[0139] In this embodiment, the first sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the following formula:

[0140] ;

[0141] Among them, c is the first sub-defect detection coefficient of the spliced ​​image to be defect detected, y1 is the first defect value and value, y2 is the second defect value and value, r2 is the number of second defect value sequences, and r1 is the number of first defect value sequences.

[0142] The beneficial effect of the above technical solution is: the present invention calculates the first sub-defect detection coefficient of the spliced ​​image to be defect detected according to the number of first defect value sequences, the number of second defect value sequences, the sum of first defect values ​​and the sum of second defect values, thereby ensuring the calculation accuracy of the first sub-defect detection coefficient, and can also provide data support for the defect detection efficiency and defect detection accuracy of organic electroluminescent devices, thereby improving work efficiency.

[0143] In some embodiments of the present application, when calculating the second sub-defect detection coefficient of the stitched image to be defect-detected according to all the third defect values, it includes:

[0144] Calculate the average defect value of all third defect values, and generate a defect value series for all third defect values ​​that are less than or equal to the average defect value;

[0145] Generate a surplus defect value sequence for all third defect values ​​greater than the average defect value;

[0146] Calculate the difference between each third defect value in the defect value series and the average defect value, and generate a defect value difference series according to the principle of numerical value size;

[0147] Calculate the profit and defect value difference between each third defect value in the profit and defect value series and the average defect value, and generate a profit and defect value difference series according to the principle of numerical value size;

[0148] A second sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the defect value difference series and the excess defect value difference series.

[0149] In this embodiment, the difference of the defect value of the head defect in the defect value difference series is the largest, and the difference of the defect value of the tail defect is the smallest, and the difference of the defect value of the head defect in the surplus defect value difference series is the largest, and the difference of the defect value of the tail surplus is the smallest.

[0150] The beneficial effect of the above technical solution is that the present invention calculates the second sub-defect detection coefficient of the spliced ​​image to be defect detected according to the deficiency defect value difference series and the surplus defect value difference series, which effectively ensures the calculation accuracy of the second sub-defect detection coefficient, thereby indirectly reducing the possibility of misjudgment and missed judgment of the organic electroluminescent device.

[0151] In some embodiments of the present application, when calculating the second sub-defect detection coefficient of the spliced ​​image to be defect-detected according to the defect value difference series and the surplus defect value difference series, it includes:

[0152] Determine the maximum defect value difference and the minimum defect value difference of the defect value difference sequence;

[0153] Calculating the defect difference between the maximum defect value difference and the minimum defect value difference;

[0154] Determine the maximum profit and defect value difference and the minimum profit and defect value difference of the profit and defect value difference series;

[0155] Calculating a margin value between the maximum margin-defect value difference and the minimum margin-defect value difference;

[0156] The second sub-defect detection coefficient of the defect detection spliced ​​image is calculated according to the following formula:

[0157] ;

[0158] Among them, k is the second sub-defect detection coefficient of the spliced ​​image to be defect detected, h1 is the deficit difference, h2 is the surplus difference, s1 is the weight corresponding to the deficit defect value difference series, n1 is the number of deficit defect value differences in the deficit defect value difference series, t i is the ith defect value difference in the defect value difference series, q i is the weight corresponding to the ith defect value difference, s2 is the weight corresponding to the surplus defect value difference series, n2 is the number of surplus defect value differences in the surplus defect value difference series, u j is the jth difference between the profit and defect values ​​in the difference series of profit and defect values, w j is the weight corresponding to the j-th profit and defect value difference.

[0159] In this embodiment, max{h1, h2} refers to the larger value of the deficit value and the surplus value. If the two are equal, one of the deficit value and the surplus value is randomly selected.

[0160] S140: calculating a comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and judging whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient.

[0161] In this embodiment, when calculating the comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, the corresponding calculation coefficient is configured for the first sub-defect detection coefficient, which is preferably 0.6 here, and the corresponding calculation coefficient is configured for the second sub-defect detection coefficient, which is preferably 0.4 here. The comprehensive defect detection coefficient = 0.6 × first sub-defect detection coefficient + 0.4 × second sub-defect detection coefficient, where × is the multiplication sign in mathematical operations.

[0162] In some embodiments of the present application, when judging whether the organic electroluminescent device has a defect according to the comprehensive defect detection coefficient, the method includes:

[0163] Determining whether the organic electroluminescent device has a defect according to the relationship between the comprehensive defect detection coefficient and a preset comprehensive defect detection coefficient;

[0164] When the comprehensive defect detection coefficient is less than the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has no defects;

[0165] When the comprehensive defect detection coefficient is greater than or equal to the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has a defect.

[0166] In this embodiment, the preset comprehensive defect detection coefficient is used to distinguish whether the organic electroluminescent device has defects, and is preferably 12 here, and can be adjusted according to actual conditions.

[0167] The beneficial effects of the above technical solution are: the present invention realizes intelligent defect detection of organic electroluminescent devices, improves defect detection efficiency and accuracy, further ensures production efficiency and product quality, eliminates the need for staff to participate in defect detection, reduces long-term labor costs, avoids missed judgments and misjudgments, reduces the generation of defective finished products, and reduces costs.

[0168] In order to further explain the technical idea of ​​the present invention, the technical solution of the present invention is now described in combination with specific application scenarios.

[0169] Correspondingly, such as Figure 2 As shown, the present application also provides a defect detection system for an organic electroluminescent device, comprising:

[0170] An image processing module, used for determining an organic electroluminescent device to be detected for defects, photographing the organic electroluminescent device to be detected for defects at multiple angles to obtain multiple multi-angle images, and processing the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected for defects;

[0171] A mark generation module, used for analyzing the spliced ​​image to be defect-detected, and generating analysis marks for the organic electroluminescent device to be defect-detected based on the analysis results, wherein the analysis marks include defect marks, qualified marks and suspected defect marks;

[0172] a coefficient calculation module, for dividing the stitched image to be defect detected into a plurality of sub-defect detection areas when a suspected defect mark is identified, performing image analysis on each sub-defect detection area, and calculating a first sub-defect detection coefficient and a second sub-defect detection coefficient of the stitched image to be defect detected based on the image analysis result;

[0173] The defect detection module is used to calculate the comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and determine whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient.

[0174] In the description of the above embodiments, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0175] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed by the present invention may be used in combination with each other in any manner, and the fact that these combinations are not fully described in this specification is only for the sake of omitting space and saving resources.

[0176] Those skilled in the art can understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A defect detection method for an organic electroluminescent device, characterized in that: include: Determine an organic electroluminescent device to be detected for defects, photograph the organic electroluminescent device to be detected for defects at multiple angles to obtain multiple multi-angle images, and process the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected for defects; Analyzing the spliced ​​image to be defect-detected, and generating analysis marks for the organic electroluminescent device to be defect-detected based on the analysis results, wherein the analysis marks include defect marks, qualified marks, and suspected defect marks; When a suspected defect mark is identified, the defect detection spliced ​​image is divided into a plurality of sub-defect detection areas, and image analysis is performed on each sub-defect detection area, and a first sub-defect detection coefficient and a second sub-defect detection coefficient of the defect detection spliced ​​image are calculated based on the image analysis result; Calculating a comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and judging whether the organic electroluminescent device has a defect according to the comprehensive defect detection coefficient; When analyzing the spliced ​​image to be defect-detected and generating an analysis mark for the organic electroluminescent device to be defect-detected based on the analysis result, the method includes: Acquire a qualified inspection stitching image corresponding to the stitching image to be inspected for defects; Extracting all the pixels to be detected corresponding to the defect detection spliced ​​image, and extracting all the qualified pixels corresponding to the pixels to be detected from the qualified detection spliced ​​image; Determine the pixel values ​​to be detected corresponding to all the pixels to be detected, and determine the qualified pixel values ​​corresponding to all the qualified pixels; If the pixel values ​​to be detected are all equal to the qualified pixel values, a qualified mark is generated for the organic electroluminescent device to be defect-detected; If the pixel values ​​to be detected are not equal to the qualified pixel values, a defect mark is generated for the organic electroluminescent device to be defect-detected; If there are one or more pixel values ​​to be detected that are not equal to qualified pixel values, generating a suspected defect mark for the organic electroluminescent device to be defect-detected; When performing image analysis on each sub-defect detection area and calculating the first sub-defect detection coefficient and the second sub-defect detection coefficient of the spliced ​​image to be defect detected based on the image analysis result, it includes: Analyze each sub-defect detection area, and if the pixel values ​​to be detected in the sub-defect detection area are all equal to the qualified pixel values, delete the corresponding sub-defect detection area; Extract the remaining sub-defect detection areas and obtain the corresponding pixel values ​​to be analyzed; Acquire a qualified pixel value to be analyzed corresponding to the pixel value to be analyzed, and if the pixel value to be analyzed is less than or equal to the qualified pixel value to be analyzed, calculate a first pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed, and divide the difference into a first pixel value set; If the pixel value to be analyzed is greater than the qualified pixel value to be analyzed, a second pixel difference between the pixel value to be analyzed and the qualified pixel value to be analyzed is calculated, and the pixel value is divided into a second pixel value set; Performing curve fitting on the pixel values ​​to be analyzed in the first pixel value set to obtain a first pixel value curve; Determine the slope corresponding to each first pixel difference on the first pixel value curve, and determine the maximum slope a, and calculate the first calculation factor e according to the maximum slope a a ; Normalize the first pixel value set, calculate the first variance of the first pixel value set after the normalization, and calculate the first calculation factor e a and the first variance as the first defect value of the sub-defect detection area; Performing curve fitting on the pixel values ​​to be analyzed in the second pixel value set to obtain a second pixel value curve; Determine the slope corresponding to each second pixel difference on the second pixel value curve, determine the maximum slope b, and calculate the second calculation factor e according to the maximum slope b b ; Normalize the second pixel value set, calculate the second variance of the normalized second pixel value set, and calculate the second calculation factor e b and the second variance as the second defect value of the sub-defect detection area; Calculating a third defect value of the sub-defect detection area based on the first defect value and the second defect value; The third defect values ​​of the remaining sub-defect detection areas are calculated, and the first sub-defect detection coefficient and the second sub-defect detection coefficient of the stitched image to be defect-detected are calculated according to all the third defect values.

2. The defect detection method of an organic electroluminescent device according to claim 1, characterized in that: When the multi-angle images are processed to obtain the spliced ​​image to be inspected for defects of the organic electroluminescent device to be inspected for defects, the method includes: Preprocessing the multi-angle images, wherein the preprocessing includes denoising, geometric correction and color balancing; Extracting feature points from the preprocessed multi-angle images based on a first preset algorithm, matching the extracted feature points based on a second preset algorithm, and determining matching information between the multi-angle images, wherein the matching information includes relative positions and rotational relationships; Based on the matching information, the multi-angle images are registered, and the registered multi-angle images are merged based on a third preset algorithm to obtain an initial defect detection spliced ​​image of the organic electroluminescent device to be detected; The initial spliced ​​image to be defect-detected is optimized to obtain the spliced ​​image to be defect-detected of the organic electroluminescent device to be defect-detected, wherein the optimization includes removing splicing seams, adjusting image contrast, and sharpening image edges.

3. The defect detection method of an organic electroluminescent device according to claim 1, characterized in that: When calculating a third defect value of the sub-defect detection area based on the first defect value and the second defect value, the method includes: The third defect value of the sub-defect detection area is calculated according to the following formula: ； Wherein, d is the third defect value of the sub-defect detection area, g1 is the first defect value, and g2 is the second defect value.

4. The defect detection method of an organic electroluminescent device according to claim 1, characterized in that: When calculating the first sub-defect detection coefficient of the spliced ​​image to be defect-detected according to all the third defect values, it includes: Extracting the same third defect value from all the third defect values ​​and obtaining a plurality of defect value sequences; Count the number of the first defect value sequence of the defect value sequence; Extract a third defect value from each of all defect value sequences, and calculate the sum of the first defect values; Obtaining a pre-calculated preset defect value, eliminating all defect value sequences that are less than the preset defect value, and counting the number of second defect value sequences of the remaining defect value sequences; Extracting a defect value from each of the remaining defect value sequences and calculating a second defect value and value; A first sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the number of the first defect value sequences, the number of the second defect value sequences, the first defect value sum, and the second defect value sum.

5. The defect detection method of an organic electroluminescent device according to claim 1, characterized in that: When calculating the second sub-defect detection coefficient of the spliced ​​image to be defect-detected according to all the third defect values, it includes: Calculate the average defect value of all third defect values, and generate a defect value series for all third defect values ​​that are less than or equal to the average defect value; Generate a surplus defect value sequence for all third defect values ​​greater than the average defect value; Calculate the difference between each third defect value in the defect value series and the average defect value, and generate a defect value difference series according to the principle of numerical value size; Calculate the profit and defect value difference between each third defect value in the profit and defect value series and the average defect value, and generate a profit and defect value difference series according to the principle of numerical value size; A second sub-defect detection coefficient of the spliced ​​image to be defect-detected is calculated according to the defect value difference series and the excess defect value difference series.

6. The defect detection method of an organic electroluminescent device according to claim 5, characterized in that: When calculating the second sub-defect detection coefficient of the spliced ​​image to be defect-detected according to the defect value difference series and the excess defect value difference series, it includes: Determine the maximum defect value difference and the minimum defect value difference of the defect value difference sequence; Calculating the defect difference between the maximum defect value difference and the minimum defect value difference; Determine the maximum profit and defect value difference and the minimum profit and defect value difference of the profit and defect value difference series; Calculating a margin value between the maximum margin-defect value difference and the minimum margin-defect value difference; The second sub-defect detection coefficient of the defect detection spliced ​​image is calculated according to the following formula: ； Among them, k is the second sub-defect detection coefficient of the spliced ​​image to be defect detected, h1 is the deficit difference, h2 is the surplus difference, s1 is the weight corresponding to the deficit defect value difference series, n1 is the number of deficit defect value differences in the deficit defect value difference series, t i is the ith defect value difference in the defect value difference series, q i is the weight corresponding to the ith defect value difference, s2 is the weight corresponding to the surplus defect value difference series, n2 is the number of surplus defect value differences in the surplus defect value difference series, u j is the jth difference between the profit and defect values ​​in the difference series of profit and defect values, w j is the weight corresponding to the j-th profit and defect value difference.

7. The defect detection method of an organic electroluminescent device according to claim 1, characterized in that: When judging whether the organic electroluminescent device has a defect according to the comprehensive defect detection coefficient, the method includes: Determining whether the organic electroluminescent device has a defect according to the relationship between the comprehensive defect detection coefficient and a preset comprehensive defect detection coefficient; When the comprehensive defect detection coefficient is less than the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has no defects; When the comprehensive defect detection coefficient is greater than or equal to the preset comprehensive defect detection coefficient, it is determined that the organic electroluminescent device has a defect.

8. A defect detection system for an organic electroluminescent device, applied to the defect detection method for an organic electroluminescent device as claimed in any one of claims 1 to 7, characterized in that: include: An image processing module, used for determining an organic electroluminescent device to be detected for defects, photographing the organic electroluminescent device to be detected for defects at multiple angles to obtain multiple multi-angle images, and processing the multi-angle images to obtain a spliced ​​image of the organic electroluminescent device to be detected for defects; A mark generation module, used for analyzing the spliced ​​image to be defect-detected, and generating analysis marks for the organic electroluminescent device to be defect-detected based on the analysis results, wherein the analysis marks include defect marks, qualified marks and suspected defect marks; a coefficient calculation module, for dividing the stitched image to be defect detected into a plurality of sub-defect detection areas when a suspected defect mark is identified, performing image analysis on each sub-defect detection area, and calculating a first sub-defect detection coefficient and a second sub-defect detection coefficient of the stitched image to be defect detected based on the image analysis result; The defect detection module is used to calculate the comprehensive defect detection coefficient of the organic electroluminescent device to be defect-detected based on the first sub-defect detection coefficient and the second sub-defect detection coefficient, and determine whether the organic electroluminescent device has defects according to the comprehensive defect detection coefficient.

Images