Defect detection method, system and apparatus for micro-display device

Abstract

The present application discloses a defect detection method, system and apparatus for a micro-display device, for use in improving the defect detection precision of the micro-display device. The defect detection method comprises: on the basis of a first initial image and a second initial image of a micro-display device, acquiring first initial coordinate parameters and second initial coordinate parameters corresponding to a plurality of lamp beads; and on the basis of standard coordinate parameters, acquiring corresponding first physical coordinate parameters and second physical coordinate parameters to calculate a coordinate offset of any lamp bead.

Description

Micro-display defect detection method, system and device TECHNICAL FIELD

[0001] The present application relates to the technical field of display devices, and in particular to a micro-display defect detection method, system and device. BACKGROUND

[0002] A micro-display is a micro flat display panel using micro light emitting diode (Micro-LED) display technology or sub-millimeter light emitting diode (Mini-LED) display technology. Both Micro-LED chips and Mini-LED chips have the characteristics of small size and high density, but this can also cause the surface to be easily contaminated, worn or otherwise physically damaged. Since any small defect can affect the performance and reliability of the Micro-LED chip or Mini-LED chip, it is particularly important to improve the accuracy of defect detection in the production process of micro display devices.

[0003] Currently, in order to meet the tight rhythm requirements of defect detection, a high-speed camera is used to capture images in a very short time by using a flying shot mode, and a quick and accurate analysis is performed to detect and record any defects or abnormalities.

[0004] However, in the actual detection process, due to the high-speed movement of the detection platform or the camera, the collected images can have a large offset, making it impossible to accurately locate multiple lamp beads on the image, so that accurate defect analysis cannot be performed according to the collected images, and thus the defect detection accuracy is reduced. TECHNICAL PROBLEM

[0005] The present application provides a micro-display defect detection method, system and device, which can improve the defect detection accuracy of the micro-display. TECHNICAL SOLUTION

[0006] The first aspect of the present application provides a micro-display defect detection method, comprising:

[0007] obtaining a first initial image and a second initial image of the micro-display, the first initial image being an image collected by a camera in a flying shot mode, and the second initial image being an image collected by a positioning assembly;

[0008] obtaining a plurality of first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and obtaining a plurality of second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image;

[0009] According to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image, first effective coordinate parameters and abnormal coordinate parameters in the first initial coordinate parameters are obtained;

[0010] According to the first effective coordinate parameters, second effective coordinate parameters corresponding to the abnormal coordinate parameters are obtained;

[0011] According to the first effective coordinate parameters and the second effective coordinate parameters, first physical coordinate parameters of any of the lamp beads are obtained;

[0012] According to the second initial coordinate parameters and the standard coordinate parameters, second physical coordinate parameters of any of the lamp beads are obtained;

[0013] According to the first physical coordinate parameters and the second physical coordinate parameters, coordinate offset of any of the lamp beads is calculated.

[0014] Optionally, the step of obtaining the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters according to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image comprises:

[0015] The standard image of the micro display is obtained;

[0016] According to the standard image, the standard coordinate parameters corresponding to the plurality of lamp beads are obtained;

[0017] The correlation coefficient of the standard coordinate parameters and the first initial coordinate parameters is obtained;

[0018] It is judged whether the correlation coefficient is in a preset range;

[0019] If the correlation coefficient is in the preset range, it is judged whether the first initial coordinate parameters of each of the lamp beads match the corresponding standard coordinate parameters;

[0020] If the first initial coordinate parameters of the lamp beads match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp beads are determined as first effective coordinate parameters; if the first initial coordinate parameters of the lamp beads do not match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp beads are determined as abnormal coordinate parameters.

[0021] Optionally, the step of obtaining the second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters comprises:

[0022] The first effective coordinate parameters corresponding to two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameters are obtained;

[0023] The second effective coordinate parameter corresponding to the abnormal coordinate parameter is calculated by using an interpolation method.

[0024] Optionally, the step of obtaining the first physical coordinate parameter of any of the lamp beads according to the first effective coordinate parameter and the second effective coordinate parameter comprises:

[0025] obtaining a minimum circumscribed rectangle corresponding to any of the lamp beads according to the first effective coordinate parameter and the second effective coordinate parameter;

[0026] obtaining a peripheral rectangle located at the periphery of the minimum circumscribed rectangle according to a preset interval;

[0027] determining a transition area where the boundary of the lamp bead is located from the direction of the peripheral rectangle to the minimum circumscribed rectangle;

[0028] performing threshold segmentation on the transition area and screening boundary pixel points meeting a preset condition;

[0029] fitting a target rectangle corresponding to any of the lamp beads according to the obtained plurality of boundary pixel points;

[0030] obtaining the coordinate value of a center pixel point located in the target rectangle;

[0031] obtaining the first physical coordinate parameter of any of the lamp beads according to the coordinate value of the center pixel point and a preset calibration coefficient.

[0032] Optionally, the step of obtaining the second physical coordinate parameter of any of the lamp beads according to the second initial coordinate parameter and the standard coordinate parameter comprises:

[0033] calculating the difference between the standard coordinate parameter of any of the lamp beads and the corresponding second initial coordinate parameter;

[0034] obtaining the second physical coordinate parameter of any of the lamp beads according to the second initial coordinate parameter, the difference value and the frame value corresponding to the obtained second initial image.

[0035] Optionally, the defect detection method further comprises:

[0036] dividing each of the lamp beads on the first initial image into a sub-region according to the first effective coordinate parameter and the second effective coordinate parameter;

[0037] obtaining feature information of the lamp bead in any of the sub-regions;

[0038] calculating a restored coordinate parameter of the lamp bead according to the feature information;

[0039] obtaining a restored image according to the restored coordinate parameter.

[0040] Optionally, after the step of obtaining the restoration image according to the restoration coordinate parameter, the defect detection method further comprises:

[0041] differencing the restoration image and the standard image to obtain a difference image;

[0042] performing image preprocessing on the difference image;

[0043] performing binaryzation processing on the difference image after image preprocessing;

[0044] performing connected region marking on the difference image after binaryzation processing;

[0045] obtaining feature information of each connected region;

[0046] judging whether the lamp bead located in any connected region has a defect according to the feature information.

[0047] The second aspect of the present application provides a defect detection system of a micro display, comprising:

[0048] an obtaining unit, configured to obtain a first initial image and a second initial image of the micro display, the first initial image being an image collected by a camera in a flying shutter mode, the second initial image being an image collected by a positioning component, and configured to obtain first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and obtain second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image;

[0049] a calculating unit, configured to obtain first effective coordinate parameters and abnormal coordinate parameters in the first initial coordinate parameters according to standard coordinate parameters corresponding to the plurality of lamp beads in a preset standard image, and obtain second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters, and configured to obtain first physical coordinate parameters of any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters, and obtain second physical coordinate parameters of any lamp bead according to the second initial coordinate parameters and the standard coordinate parameters, and configured to calculate a coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters.

[0050] Optionally, the defect detection system further comprises a matching module, the matching module being configured to divide each lamp bead on the first initial image into a sub-region according to the first effective coordinate parameters and the second effective coordinate parameters;

[0051] the matching module is further configured to obtain feature information of the lamp bead in any sub-region;

[0052] The matching module is also used to calculate the restored coordinate parameters of the LED bead based on the feature information;

[0053] The matching module is also used to obtain the restored image based on the restored coordinate parameters.

[0054] Optionally, the defect detection system further includes a detection module, which is used to perform difference analysis between the restored image and the standard image to obtain a difference image;

[0055] The detection module is also used to perform image preprocessing on the differential image;

[0056] The detection module is also used to perform binarization processing on the differential image after image preprocessing;

[0057] The detection module is also used to mark connected regions in the binarized difference image;

[0058] The detection module is also used to acquire feature information of each of the connected regions;

[0059] The detection module is also used to determine whether the LED bead located in any of the connected regions has a defect based on the feature information.

[0060] A third aspect of this application provides a defect detection device for a microdisplay, comprising a defect detection system for the microdisplay and a worktable for placing the microdisplay to be tested. The defect detection system for the microdisplay includes:

[0061] The acquisition unit is configured to acquire a first initial image and a second initial image of the microdisplay, wherein the first initial image is an image captured by the camera in flight mode and the second initial image is an image captured by the positioning component, and is configured to acquire first initial coordinate parameters corresponding to multiple LED beads based on the first initial image, and acquire second initial coordinate parameters corresponding to multiple LED beads based on the second initial image.

[0062] The calculation unit is configured to obtain a first valid coordinate parameter and an abnormal coordinate parameter from the first initial coordinate parameters based on the standard coordinate parameters corresponding to the multiple LEDs in the preset standard image, and to obtain a second valid coordinate parameter corresponding to the abnormal coordinate parameter based on the first valid coordinate parameter. The calculation unit is also configured to obtain a first physical coordinate parameter of any LED based on the first valid coordinate parameter and the second valid coordinate parameter, and to obtain a second physical coordinate parameter of any LED based on the second initial coordinate parameter and the standard coordinate parameters, and to calculate the coordinate offset of any LED based on the first physical coordinate parameter and the second physical coordinate parameter. Beneficial effects

[0063] The defect detection method for a microdisplay provided in this application first acquires a first initial image and a second initial image of the microdisplay. The first initial image is an image captured by a camera in flight mode, and the second initial image is an image captured by a positioning component. Then, based on the first initial image, first initial coordinate parameters corresponding to multiple LEDs are acquired. The first initial coordinate parameters include first valid coordinate parameters and abnormal coordinate parameters. Second initial coordinate parameters corresponding to multiple LEDs are also acquired based on the second initial image. Next, based on the standard coordinate parameters corresponding to multiple LEDs in a preset standard image, first valid coordinate parameters and abnormal coordinate parameters are acquired from the first initial coordinate parameters. Second valid coordinate parameters corresponding to the abnormal coordinate parameters are acquired based on the first valid coordinate parameters. Further, first physical coordinate parameters of any LED are acquired based on the first and second valid coordinate parameters. Further, second physical coordinate parameters of any LED are acquired based on the second initial coordinate parameters and standard coordinate parameters. Finally, the coordinate offset of any LED is calculated based on the first and second physical coordinate parameters. Through the above steps, the initial image acquired by the microdisplay in aerial photography mode can be preliminarily located. Based on the preliminary location results, the first initial coordinate parameters corresponding to multiple LEDs are obtained. Then, using a preset standard image, the first valid coordinate parameters and the second valid coordinate parameters corresponding to the abnormal coordinate parameters in the first initial coordinate parameters are determined. Next, based on the first and second valid coordinate parameters, the first initial image is further located to obtain the first physical coordinate parameters for accurate positioning of multiple LEDs. Then, based on the second initial coordinate parameters corresponding to the second initial image acquired by the positioning component and the standard coordinate parameters corresponding to the standard image, the reference points of multiple LEDs are located to obtain the second physical coordinate parameters corresponding to multiple LEDs. Finally, based on these first and second physical coordinate parameters, the coordinate offset of multiple LEDs is determined. Therefore, by performing multiple positioning operations, the inaccurate positioning of LEDs due to image offset can be reduced, thereby improving the defect detection accuracy of the microdisplay. Attached Figure Description

[0064] Figure 1 is a schematic diagram of an embodiment of a defect detection method for a microdisplay according to this application;

[0065] Figures 2-1, 2-2, 2-3 and 2-4 are schematic diagrams of another embodiment of a defect detection method for a microdisplay according to this application;

[0066] Figure 3 is a schematic diagram of an embodiment of a defect detection system for a microdisplay according to this application;

[0067] Figure 4 is a schematic diagram of another embodiment of a defect detection system for a microdisplay according to this application;

[0068] FIG. 5 is a schematic diagram of one embodiment of a defect detection device for a micro display. Embodiments of the present application

[0069] The present application provides a defect detection method, system and device for a micro display, which can improve the defect detection accuracy of the micro display.

[0070] The defect detection method for a micro display described in the present application can be executed on a server, a system, a terminal or other device with logical processing capability, which is not limited in the present application. The embodiments of the present application are described by taking a system with image analysis and processing capability as an example. Referring to FIG. 1, one embodiment of the defect detection method for a micro display in the present application includes the following steps:

[0071] 101. The system acquires a first initial image and a second initial image of the micro display, wherein the first initial image is an image captured by a camera in a high-speed flying shot mode, and the second initial image is an image captured by a positioning component.

[0072] In the embodiment, first, the aperture focal length of the camera lens is fixed under the condition of meeting the high-speed flying shot mode, then the positional relationship between the detection platform and the camera is adjusted, the point light source is fixed on one side of the camera lens, and the camera parameters are saved to complete the preliminary work. When the high-speed flying shot mode is started, the camera shooting action is triggered to complete the image acquisition of the micro display to generate the first initial image of the micro display, and the first initial image is transmitted to the system. Before the first initial image is acquired by using the flying shot mode, the image acquisition of the micro display in a static state is needed to be performed by using the positioning component to obtain the second initial image.

[0073] It should be noted that when the camera completes the shooting in the flying shot mode, the first initial image is usually interfered and affected by various noises in the acquisition, transmission and storage process, and thus the image quality is reduced. Therefore, in order to improve the image quality of the first initial image, the mean filter algorithm or the Gaussian filter algorithm can be used to perform the noise reduction processing. When the mean filter algorithm is used, a 3x3 mean filter window can be used to perform the filter processing on the first initial image. When the Gaussian filter algorithm is used, the Gaussian function can be used as a filter to perform the weighted average on the signal to smooth the signal and reduce the noise, and the filter processing is performed on the first initial image.

[0074] 102. The system acquires first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and acquires second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image.

[0075] In this embodiment, the system extracts the lamp bead feature information of all the lamp beads on the first initial image, and determines the first initial coordinate parameters corresponding to each lamp bead on the first initial image according to the extracted lamp bead feature information. Similarly, the system extracts the lamp bead feature information of all the lamp beads on the second initial image, and determines the second initial coordinate parameters corresponding to each lamp bead on the second initial image according to the extracted lamp bead feature information.

[0076] In this embodiment, the first initial coordinate parameters include the coordinate value of the geometric center of the lamp bead and the distance from the geometric center of the lamp bead to the boundary of the lamp bead. For example, in the row direction, the distance from the geometric center of the lamp bead to the boundary of the lamp bead is L1, and in the column direction, the distance from the geometric center of the lamp bead to the boundary of the lamp bead is L2. Wherein, L1 and L2 can be equal or not equal.

[0077] 103、The system obtains the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters according to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image.

[0078] In this embodiment, since the first initial image obtained by the system in the snapshot mode may be offset, deformed, blurred and the like, the first initial coordinate parameters obtained by the system according to the first initial image are rough coordinate parameters, which include the first effective coordinate parameters with high positioning accuracy and the abnormal coordinate parameters with low positioning accuracy. At this time, the system can select an image with intact lamp beads, obvious features and clear background from a plurality of finished micro display images as a standard image, and compare the standard image with the first initial image to determine the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters.

[0079] 104、The system obtains the second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters.

[0080] In this embodiment, after the system determines the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters, it needs to determine the second effective coordinate parameters corresponding to the abnormal coordinate parameters. Specifically, the interpolation method can be used to fit the second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the known first effective coordinate parameters.

[0081] 105、The system obtains the first physical coordinate parameters of any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters.

[0082] In this embodiment, the system obtains the first effective coordinate parameters and the second effective coordinate parameters as the effective coordinates of each lamp bead on the first initial image in the image coordinate system, and needs to convert the effective coordinates into the first physical coordinate parameters by using the preset calibration coefficient, where the first physical coordinate parameters are the real physical coordinates of each lamp bead on the first initial image relative to the flying point position. It can be understood that when the first effective coordinate parameters and the second effective coordinate parameters are converted into the first physical coordinate parameters, each lamp bead can be further positioned, that is, a more accurate coordinate of each lamp bead is determined by using the method of finding the edge region of the minimum circumscribed rectangle, and then the coordinate is converted into the first physical coordinate parameters by using the preset calibration coefficient, and the specific implementation process will be described in subsequent embodiments.

[0083] 106. The system obtains the second physical coordinate parameters of any lamp bead according to the second initial coordinate parameters and the standard coordinate parameters.

[0084] In this embodiment, the system first performs a difference between the first initial coordinate parameters and the standard coordinate parameters, and then obtains the second physical coordinate parameters of each lamp bead according to the obtained difference, where the second physical coordinate parameters are the real physical coordinates of the calibration point positions corresponding to each lamp bead relative to the flying point position.

[0085] 107. The system calculates the coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters.

[0086] The system can obtain the coordinate offset corresponding to each lamp bead by performing a difference between the calculated first physical coordinate parameters and the second physical coordinate parameters, where the coordinate offset is the real physical offset of the lamp bead.

[0087] In this embodiment, first, the first initial image of the micro display and the second initial image are acquired, the first initial image is the image collected by the camera in the fly-shooting mode, and the second initial image is the image collected by the positioning component. Then, the first initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the first initial image, the first initial coordinate parameters include the first effective coordinate parameters and the abnormal coordinate parameters, and the second initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the second initial image. Then, the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters are acquired according to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image. The second effective coordinate parameters corresponding to the abnormal coordinate parameters are acquired according to the first effective coordinate parameters. The first physical coordinate parameters of any lamp bead are further acquired according to the first effective coordinate parameters and the second effective coordinate parameters. The second physical coordinate parameters of any lamp bead are further acquired according to the second initial coordinate parameters and the standard coordinate parameters. Finally, the coordinate offset of any lamp bead is calculated according to the first physical coordinate parameters and the second physical coordinate parameters. In this way, the first initial image collected by the micro display in the fly-shooting mode can be preliminarily positioned, the first initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the preliminary positioning result, the first effective coordinate parameters in the first initial coordinate parameters and the second effective coordinate parameters corresponding to the abnormal coordinate parameters are determined by using the preset standard image, then the first initial image is further positioned according to the first effective coordinate parameters and the second effective coordinate parameters in the first initial coordinate parameters, and the first physical coordinate parameters of the plurality of lamp beads are acquired. The reference points of the plurality of lamp beads are positioned according to the second initial coordinate parameters corresponding to the second initial image collected by the positioning component and the standard coordinate parameters corresponding to the standard image, and the second physical coordinate parameters corresponding to the plurality of lamp beads are obtained. Finally, the coordinate offset corresponding to the plurality of lamp beads is determined according to the first physical coordinate parameters and the second physical coordinate parameters. In this way, the inaccurate positioning of the lamp beads caused by image offset can be reduced by multiple positioning, and the defect detection precision of the micro display can be improved.

[0088] Please refer to FIG. 2-1, FIG. 2-2, FIG. 2-3 and FIG. 2-4, another embodiment of the defect detection method of the micro display in the application includes:

[0089] 201, the system acquires the first initial image and the second initial image of the micro display, the first initial image is the image collected by the camera in the fly-shooting mode, and the second initial image is the image collected by the positioning component.

[0090] 202, the system acquires the first initial coordinate parameters corresponding to the plurality of lamp beads according to the first initial image, and acquires the second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image.

[0091] Steps 201-202 in this embodiment are similar to steps 101-102 in the embodiment shown in FIG. 1, and thus are not described again here.

[0092] 203. The system acquires a standard image of the micro display.

[0093] 204. The system acquires standard coordinate parameters corresponding to the plurality of lamp beads according to the standard image.

[0094] 205. The system acquires a correlation coefficient between the standard coordinate parameters and the first initial coordinate parameters.

[0095] 206. The system determines whether the correlation coefficient is within a preset range. If yes, step 207 is performed. If no, step 207 is performed.

[0096] 207. The system determines whether the first initial coordinate parameters of each lamp bead match the corresponding standard coordinate parameters. If yes, step 208 is performed. If no, step 209 is performed.

[0097] 208. The system determines that the first initial coordinate parameters of the lamp beads are first effective coordinate parameters.

[0098] 209. The system determines that the first initial coordinate parameters of the lamp beads are abnormal coordinate parameters.

[0099] Optionally, in this embodiment, after the system acquires the standard image, the system extracts image features of each lamp bead on the standard image, and determines standard coordinate parameters corresponding to each lamp bead according to the extracted image feature information. The standard image is slid on the first initial image along the possible position area of each lamp bead one by one, and the correlation coefficient between the standard coordinate parameters and the first initial coordinate parameters of the current lamp bead is calculated in each sliding process. When calculating the correlation coefficient, the standard coordinate parameters and the first initial coordinate parameters need to be zero-meaned, and then the normalized cross-correlation between them is calculated. The calculation can be performed according to the following formula:

[0100] where (x', y') represents the standard coordinate parameters of any lamp bead on the standard image, (x, y) represents the first initial coordinate parameters of any lamp bead on the first initial image, T(x+x', y+y') represents the current pixel value of the standard image, I(x', y') represents the current pixel value of the first initial image, represents the average pixel value of the standard image, NCC(x, y) = (x - x) (y - y) (x - x) (y - y) where, μ1(x, y) represents the average pixel value of the first initial image, and NCC(x, y) represents the correlation coefficient of the standard coordinate parameter and the first initial coordinate parameter. The numerator part of the formula is the covariance of the two signals, and the denominator part is the product of their respective standard deviations, for normalizing the cross-correlation value result to the range of [-1, 1]. If NCC(x, y) is close to 1, it indicates that the current possible position area of the lamp bead in the first initial image is highly similar to the standard image. If NCC(x, y) is close to -1, it indicates that the current possible position area of the lamp bead in the first initial image is highly opposite to the standard image. If NCC(x, y) is close to 0, it indicates that the current possible position area of the lamp bead in the first initial image has no obvious correlation with the standard image. Therefore, the standard image can be correlated with all the possible position areas of the lamp beads near a certain lamp bead A one by one to calculate the correlation coefficient, and it is judged whether the correlation coefficient is within the preset range. If the correlation coefficient is within the preset range, it is continued to be judged whether the first initial coordinate parameter of the lamp bead A matches the corresponding standard coordinate parameter. If the first initial coordinate parameter of the lamp bead A matches the corresponding standard coordinate parameter, it indicates that the first initial coordinate parameter of the lamp bead A is the first effective coordinate parameter. If the first initial coordinate parameter of the lamp bead A does not match the corresponding standard coordinate parameter, it indicates that the first initial coordinate parameter of the lamp bead A is the abnormal coordinate parameter.

[0101] 210、The system obtains the first effective coordinate parameters corresponding to the two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameter.

[0102] 211、The system calculates the second effective coordinate parameter corresponding to the abnormal coordinate parameter by using the interpolation method.

[0103] Optionally, in the embodiment, the system can first arrange all the first initial coordinate parameters according to the distribution of the lamp beads, and then select all the first initial coordinate parameters located in the same row as the abnormal coordinate parameter or all the first initial coordinate parameters located in the same column as the abnormal coordinate parameter. Then, two first effective coordinate parameters adjacent to the abnormal coordinate parameter are selected from all the selected first initial coordinate parameters, and finally the interpolation method is used to coordinate fit the two first effective coordinate parameters to fit the second effective coordinate parameter corresponding to the abnormal coordinate parameter.

[0104] 212、The system obtains the minimum circumscribed rectangle corresponding to any lamp bead according to the first effective coordinate parameter and the second effective coordinate parameter.

[0105] 213、The system obtains the peripheral rectangle located in the periphery of the minimum circumscribed rectangle according to the preset interval.

[0106] 214、The system determines the transition area where the boundary of the lamp bead is located from the peripheral rectangle to the direction of the minimum circumscribed rectangle.

[0107] 215、The system performs threshold segmentation on the transition region and screens boundary pixel points meeting preset conditions.

[0108] 216、The system fits a target rectangle corresponding to any lamp bead according to the obtained plurality of boundary pixel points.

[0109] 217、The system obtains a coordinate value of a center pixel point located in the target rectangle.

[0110] 218、The system obtains a first physical coordinate parameter of any lamp bead according to the coordinate value of the center pixel point and a preset calibration coefficient.

[0111] Optionally, in the embodiment, the minimum circumscribed rectangle of the lamp bead refers to a minimum area rectangle that can contain the target lamp bead. After obtaining the first effective coordinate parameter and the second effective coordinate parameter of all the lamp beads, the system can solve the minimum circumscribed rectangle of the corresponding lamp bead according to the first effective coordinate parameter or the second effective coordinate parameter of each lamp bead. The rotation and clamping method can be used to achieve this, that is, the convex hull of the target lamp bead region is first solved, and then the minimum circumscribed rectangle of the convex hull is solved. The specific minimum circumscribed rectangle solving algorithm is not limited here. Each side of the minimum circumscribed rectangle is a vertical direction or a parallel direction. The vertical direction is perpendicular to a coordinate axis in the image coordinate system of the first initial image, and the parallel direction is parallel to the coordinate axis.

[0112] The minimum circumscribed rectangle is expanded by a preset interval to obtain a peripheral rectangle of the periphery, for example, 30 pixels are expanded. The specific expansion parameter is not limited here according to actual needs. Then, taking the rectangular side of the peripheral rectangle as a base point, edge detection is performed in the direction of the minimum circumscribed rectangle to determine the transition region where the boundary of the lamp bead is located. Threshold segmentation is then performed on the transition region to screen out boundary pixel points meeting preset conditions, and a target rectangle corresponding to the lamp bead is fitted according to the boundary pixel points. The target rectangle is the minimum circumscribed rectangle corresponding to the true contour of the lamp bead. Finally, the coordinate value of the center pixel point in the target rectangle is obtained. The coordinate value of the center pixel point is the true coordinate value of any lamp bead in the image coordinate system of the first initial image. Let the coordinate value of the center pixel point of any lamp bead be (x, y), and the calibration coefficient be β. The first physical coordinate parameter of the corresponding lamp bead is wherein

[0113] 219、The system calculates the difference between the standard coordinate parameter of any lamp bead and the corresponding second initial coordinate parameter.

[0114] 220、The system calculates the second physical coordinate parameter of any lamp bead according to the second initial coordinate parameter, the difference value, and the frame value corresponding to the obtained second initial image.

[0115] ​Optionally, in the embodiment, the second initial coordinate parameter of any lamp bead is The difference between the standard coordinate parameter and the second initial coordinate parameter of the corresponding lamp bead is The second physical coordinate parameter of the corresponding lamp bead is The calculation formula of the second physical coordinate parameter is as follows:

[0116] Wherein, n is the frame value corresponding to the current second initial image, a mx and are the difference values of the standard coordinate parameter and the second initial coordinate parameter in the x direction and the y direction, respectively.

[0117] 221、The system calculates the coordinate offset of any lamp bead according to the first physical coordinate parameter and the second physical coordinate parameter.

[0118] In the embodiment, after obtaining the first physical coordinate parameter and the second physical coordinate parameter, the first physical coordinate parameter and the second physical coordinate parameter can be subtracted to determine the coordinate offset of the lamp bead. The coordinate offset of any lamp bead is offset (offset x , offset y ), and the corresponding calculation formula is:

[0119] Wherein, x1 is the x-axis coordinate value of the first physical coordinate parameter, is the y-axis coordinate value of the first physical coordinate parameter, is the x-axis coordinate value of the second physical coordinate parameter, is the y-axis coordinate value of the second physical coordinate parameter.

[0120] 222、The system divides each lamp bead on the first initial image into a sub-region according to the first effective coordinate parameter and the second effective coordinate parameter.

[0121] 223、The system obtains the feature information of the lamp bead in any sub-region.

[0122] 224、The system calculates the restoration coordinate parameter of the lamp bead according to the feature information.

[0123] 225、The system obtains the restoration image according to the restoration coordinate parameter.

[0124] Optionally, in the high-speed snapshot mode, the high-speed movement of the detection platform or the industrial camera can cause blurring, slight deformation and distortion of the target image, so that the real defects on the surface of the lamp bead cannot be extracted. Therefore, the Lucas-Kanade algorithm can be used to correct the slight deformation of the lamp bead, and then the real surface defects are detected. The Lucas-Kanade algorithm is a method for finding the corresponding relationship between the previous frame and the current frame by using the change of pixels in the time domain and the correlation between adjacent frames, so as to calculate the motion information of the object between adjacent frames. The basic idea is based on the following three assumptions:

[0125] 1. Constant brightness: the brightness of the same point does not change over time.

[0126] 2. Motion: the position does not change dramatically over time, and only in the case of small motion can the gray scale change caused by the unit position change between the previous and subsequent frames be used to approximate the partial derivative of the gray scale to the position.

[0127] 3. Spatial consistency: adjacent points on a scene are also adjacent points in the projection image, and the adjacent points have consistent speed. Because the basic equation of the optical flow method has only one constraint, and the speed in the x and y directions has two unknown variables, it is necessary to solve a plurality of equations simultaneously.

[0128] The basic constraint equation is:

[0129] Let a pixel I(x, y, t) have a light intensity in the first frame (where t represents the time dimension in which it is located). It moves a distance (dx, dy) to the next frame, and uses dt time. Because it is the same pixel point, it is assumed that the light intensity before and after the movement of the pixel is unchanged, that is:

[0130] I(x, y, t) = I(x + dx, y + dy, t + dt)

[0131] After expanding the right side of the above formula using the Taylor formula, we can get:

[0132] Let u and v be the velocity vectors of the x and y axes, respectively, that is:

[0133] Let Then we have:

[0134] I x dx+I y dy+I t dt=0

[0135] 1 x u+Iy v = -I t

[0136] In matrix form, it is expressed as:

[0137] Finally, the least square method is used to obtain:

[0138] wherein, is the final optical flow.

[0139] The system first cuts each lamp bead on the first initial image according to the first effective coordinate parameter or the second effective coordinate parameter corresponding to a plurality of lamp beads, so as to divide each lamp bead into a sub-region. Then, the system extracts features of the lamp bead in each sub-region to obtain corresponding feature information. Then, the feature information is input into the basic constraint equation of the Lucas-Kanade algorithm, and the restored coordinate parameter of the corresponding lamp bead is determined according to the solved optical flow. Finally, the restored coordinate parameters of all lamp beads are mapped to the first initial image to obtain the restored image of the micro display.

[0140] 226、The system differentiates the restored image from the standard image to obtain a difference image.

[0141] 227、The system performs image preprocessing on the difference image.

[0142] 228、The system performs binaryzation processing on the image-preprocessed difference image.

[0143] 229、The system performs connected region marking on the binaryzation-processed difference image.

[0144] 230、The system obtains feature information of each connected region.

[0145] 231、The system judges whether the lamp bead located in any connected region has a defect according to the feature information.

[0146] Optionally, in the embodiment, first, the reduced image is differentiated with the standard image to obtain the difference image of the two. Then the obtained difference image is preprocessed, and the image preprocessing process can include grayscale processing and noise removal processing. Among them, the grayscale processing is to convert the color image into a grayscale image, simplifying the subsequent processing process. The noise removal processing is to remove the noise in the image using a filter (such as a Gaussian filter). After the image preprocessing is completed, the binarization processing is performed to convert the difference image into a black and white image, so that the defects and the background can be clearly distinguished, and the threshold segmentation or adaptive threshold segmentation method can be used to complete the binarization processing, and the specific place is not limited. Then, the difference image after binarization processing is labeled with a connected region, and each connected region in the image is identified, and the connected component labeling algorithm can be used to complete the connected region labeling. Then the feature information of each labeled connected region is extracted, such as area, perimeter, shape and other feature information, which can be used to distinguish the normal part and the defect part in the image. Finally, according to the extracted feature information, a suitable rule or model is designed to judge whether each connected region has a defect, for example, the area size, perimeter and other characteristics can be used for judgment, and the judgment result is analyzed and recorded to generate the lamp bead surface defect detection result of the micro display. It is worth mentioning that after the lamp bead surface defect detection is completed, the performance evaluation of the defect detection process can be performed, including accuracy, recall rate, false detection rate and other indicators, to verify its effectiveness and reliability.

[0147] Referring to FIG. 3, one embodiment of the defect detection system of the micro display in the application includes:

[0148] The acquisition unit 301 is configured to acquire a first initial image and a second initial image of the micro display, the first initial image is an image collected by a camera in a flying shutter mode, the second initial image is an image collected by a positioning component, and the acquisition unit 301 is further configured to acquire first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and acquire second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image.

[0149] The calculation unit 302 is configured to acquire first effective coordinate parameters and abnormal coordinate parameters in the first initial coordinate parameters according to standard coordinate parameters corresponding to a plurality of lamp beads in a preset standard image, and acquire second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters, and the calculation unit 302 is further configured to acquire first physical coordinate parameters of any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters, and acquire second physical coordinate parameters of any lamp bead according to the second initial coordinate parameters and the standard coordinate parameters, and calculate a coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters.

[0150] In this embodiment, the acquisition unit 301 acquires a first initial image and a second initial image of the micro display, the first initial image is an image collected by the camera in the fly-shooting mode, the second initial image is an image collected by the positioning assembly, and the first initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the first initial image, and the second initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the second initial image; the calculation unit 302 acquires the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters according to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image, and acquires the second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters, the calculation unit is also used for acquiring the first physical coordinate parameters of any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters, and is used for acquiring the second physical coordinate parameters of any lamp bead according to the second initial coordinate parameters and the standard coordinate parameters, and is used for calculating the coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters. In this way, the first initial image collected by the micro display in the fly-shooting mode can be preliminarily positioned, the first initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the preliminary positioning result, the first effective coordinate parameters in the first initial coordinate parameters and the second effective coordinate parameters corresponding to the abnormal coordinate parameters are determined by using the preset standard image, and then the first initial image is further positioned according to the first effective coordinate parameters and the second effective coordinate parameters in the first initial coordinate parameters, and the first physical coordinate parameters of the plurality of lamp beads are acquired. According to the second initial coordinate parameters corresponding to the second initial image collected by the positioning assembly and the standard coordinate parameters corresponding to the standard image, the reference points of the plurality of lamp beads are positioned, and the second physical coordinate parameters corresponding to the plurality of lamp beads are obtained, and finally the coordinate offset corresponding to the plurality of lamp beads is determined according to the first physical coordinate parameters and the second physical coordinate parameters. In this way, the inaccurate positioning of the lamp beads caused by image offset can be reduced by multiple positioning, and the defect detection precision of the micro display can be improved.

[0151] Please refer to FIG. 4, another embodiment of the defect detection system of the micro display in the application includes:

[0152] The acquisition unit 401 is used for acquiring a first initial image and a second initial image of the micro display, the first initial image is an image collected by the camera in the fly-shooting mode, the second initial image is an image collected by the positioning assembly, and the first initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the first initial image, and the second initial coordinate parameters corresponding to the plurality of lamp beads are acquired according to the second initial image.

[0153] The computing unit 402 is specifically configured to acquire a standard image of the micro display. A plurality of standard coordinate parameters corresponding to the lamp beads are acquired according to the standard image. A correlation coefficient of the standard coordinate parameters and the first initial coordinate parameters is acquired. It is judged whether the correlation coefficient is located in a preset range. If the correlation coefficient is located in the preset range, it is judged whether the first initial coordinate parameters of each lamp bead match the corresponding standard coordinate parameters. If the first initial coordinate parameters of the lamp bead match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp bead are determined as the first effective coordinate parameters. If the first initial coordinate parameters of the lamp bead do not match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp bead are determined as the abnormal coordinate parameters. The second effective coordinate parameters corresponding to the abnormal coordinate parameters are acquired according to the first effective coordinate parameters. The first effective coordinate parameters corresponding to two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameters are acquired. The interpolation method is used to calculate the second effective coordinate parameters corresponding to the abnormal coordinate parameters. The computing unit is also specifically configured to acquire a minimum circumscribed rectangle corresponding to any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters. An outer rectangle located at the periphery of the minimum circumscribed rectangle is acquired according to a preset interval. A transition area where the boundary of the lamp bead is located is determined from the outer rectangle to the direction of the minimum circumscribed rectangle. The transition area is subjected to threshold segmentation, and boundary pixel points meeting a preset condition are screened. A target rectangle corresponding to any lamp bead is fitted according to the acquired plurality of boundary pixel points. A coordinate value of a center pixel point located in the target rectangle is acquired. The first physical coordinate parameters of any lamp bead are acquired according to the coordinate value of the center pixel point and a preset calibration coefficient. The computing unit is also specifically configured to calculate a difference value between the standard coordinate parameters of any lamp bead and the corresponding second initial coordinate parameters. The second physical coordinate parameters of any lamp bead are calculated according to the second initial coordinate parameters, the difference value and a frame value corresponding to the acquired second initial image. The computing unit is also configured to calculate the coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters.

[0154] The matching module 403 is configured to divide each lamp bead on the first initial image into a sub-region according to the first effective coordinate parameters and the second effective coordinate parameters. The matching module is also configured to acquire feature information of the lamp bead in any sub-region. The matching module is also configured to calculate the restoration coordinate parameters of the lamp bead according to the feature information. The matching module is also configured to acquire the restoration image according to the restoration coordinate parameters.

[0155] The detection module 404 is configured to perform difference between the restoration image and the standard image to obtain a difference image. The detection module is also configured to perform image preprocessing on the difference image. The detection module is also configured to perform binaryzation processing on the difference image after the image preprocessing. The detection module is also configured to perform connected region marking on the difference image after the binaryzation processing. The detection module is also configured to acquire feature information of each connected region. The detection module is also configured to judge whether the lamp bead located in any connected region has a defect according to the feature information.

[0156] In this embodiment, the functions of the units and modules are similar to those of steps 201 to 231 in the embodiments shown in FIG. 2-1, FIG. 2-2, FIG. 2-3 and FIG. 2-4, and will not be described here.

[0157] Referring to FIG. 5, one embodiment of the micro-display defect detection device in the present application includes the micro-display defect detection system in the embodiments shown in FIG. 3 or FIG. 4 and a workbench for placing the micro-display to be tested, and specifically includes a central processing unit 502, a memory 501, an input / output interface 503, a wired or wireless network interface 504 and a power supply 505.

[0158] The memory 501 is a volatile memory or a persistent memory.

[0159] The central processing unit 502 is configured to communicate with the memory 501 and execute the instructions in the memory 501 to perform the steps in the embodiments shown in FIG. 1 to FIG. 2-1, FIG. 2-2, FIG. 2-3 and FIG. 2-4.

[0160] The micro-display defect detection device also provides a computer readable storage medium including instructions, which, when executed on a computer, cause the computer to perform the steps in the embodiments shown in FIG. 1 to FIG. 2-1, FIG. 2-2, FIG. 2-3 and FIG. 2-4.

[0161] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0162] In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the units is only a logical function division. There can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or in other forms.

[0163] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0164] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0165] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, read-only memory), a random access memory (RAM, random access memory), a magnetic disk or an optical disk, and various media that can store program codes.

Claims

1. A method for detecting defects of a micro display, comprising: acquiring a first initial image and a second initial image of the micro display, the first initial image being an image captured by a camera in a fly-by mode, and the second initial image being an image captured by a positioning component; acquiring first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and acquiring second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image; acquiring first effective coordinate parameters and abnormal coordinate parameters in the first initial coordinate parameters according to standard coordinate parameters corresponding to the plurality of lamp beads in a preset standard image; acquiring second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters; acquiring first physical coordinate parameters of any of the lamp beads according to the first effective coordinate parameters and the second effective coordinate parameters; acquiring second physical coordinate parameters of any of the lamp beads according to the second initial coordinate parameters and the standard coordinate parameters; calculating coordinate offset of any of the lamp beads according to the first physical coordinate parameters and the second physical coordinate parameters.

2. The defect detection method as claimed in claim 1, wherein, The step of acquiring the first effective coordinate parameters and the abnormal coordinate parameters in the first initial coordinate parameters according to the standard coordinate parameters corresponding to the plurality of lamp beads in the preset standard image comprises: acquiring the standard image of the micro display; acquiring the standard coordinate parameters corresponding to the plurality of lamp beads according to the standard image; acquiring a correlation coefficient of the standard coordinate parameters and the first initial coordinate parameters; judging whether the correlation coefficient is within a preset range; if the correlation coefficient is within the preset range, judging whether the first initial coordinate parameters of each of the lamp beads match the corresponding standard coordinate parameters; if the first initial coordinate parameters of the lamp bead match the corresponding standard coordinate parameters, determining that the first initial coordinate parameters of the lamp bead are first effective coordinate parameters, and if the first initial coordinate parameters of the lamp bead do not match the corresponding standard coordinate parameters, determining that the first initial coordinate parameters of the lamp bead are abnormal coordinate parameters.

3. The defect detection method as claimed in claim 1, wherein, The step of acquiring the second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters comprises: acquiring the first effective coordinate parameters corresponding to two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameters; calculating the second effective coordinate parameters corresponding to the abnormal coordinate parameters by using an interpolation method.

4. The defect detection method as claimed in claim 1, wherein, The step of acquiring the first physical coordinate parameters of any of the lamp beads according to the first effective coordinate parameters and the second effective coordinate parameters comprises: acquiring a minimum circumscribed rectangle corresponding to any of the lamp beads according to the first effective coordinate parameters and the second effective coordinate parameters; acquiring a peripheral rectangle located at a periphery of the minimum circumscribed rectangle according to a preset interval; determining a transition area in which a boundary of the lamp bead is located from a direction of the minimum circumscribed rectangle to the peripheral rectangle; performing threshold segmentation on the transition area, and screening boundary pixel points meeting a preset condition; fitting a target rectangle corresponding to any of the lamp beads according to the acquired plurality of boundary pixel points; Obtaining a coordinate value of a center pixel point located in the target rectangle; Obtaining a first physical coordinate parameter of any of the lamp beads according to the coordinate value of the center pixel point and a preset calibration coefficient.

5. The defect detection method as claimed in claim 1, wherein, The step of obtaining a second physical coordinate parameter of any of the lamp beads according to the second initial coordinate parameter and the standard coordinate parameter comprises: Calculating a difference value between the standard coordinate parameter of any of the lamp beads and the corresponding second initial coordinate parameter; Obtaining a second physical coordinate parameter of any of the lamp beads according to the second initial coordinate parameter, the difference value and a frame value corresponding to the second initial image.

6. The defect detection method as claimed in claim 1, wherein, The defect detection method further comprises: Dividing each of the lamp beads on the first initial image into a sub-region according to the first effective coordinate parameter and the second effective coordinate parameter; Obtaining feature information of the lamp bead in any of the sub-regions; Calculating a restoration coordinate parameter of the lamp bead according to the feature information; Obtaining a restoration image according to the restoration coordinate parameter.

7. The defect detection method as claimed in claim 6, wherein, After the step of obtaining a restoration image according to the restoration coordinate parameter, the defect detection method further comprises: Differencing the restoration image and the standard image to obtain a difference image; Performing image preprocessing on the difference image; Performing binaryzation processing on the image-preprocessed difference image; Performing connected region marking on the binaryzation-processed difference image; Obtaining feature information of each of the connected regions; Judging whether the lamp bead located in any of the connected regions has a defect according to the feature information. 8.A defect detection system of a micro display, comprising: an obtaining unit, configured to obtain a first initial image and a second initial image of the micro display, the first initial image being an image collected by a camera in a flying shutter mode, the second initial image being an image collected by a positioning component, and configured to obtain first initial coordinate parameters corresponding to a plurality of lamp beads according to the first initial image, and obtain second initial coordinate parameters corresponding to the plurality of lamp beads according to the second initial image; a calculating unit, configured to obtain a first effective coordinate parameter and an abnormal coordinate parameter in the first initial coordinate parameters according to standard coordinate parameters corresponding to a plurality of lamp beads in a preset standard image, and obtain a second effective coordinate parameter corresponding to the abnormal coordinate parameter according to the first effective coordinate parameter, and configured to obtain a first physical coordinate parameter of any of the lamp beads according to the first effective coordinate parameter and the second effective coordinate parameter, and obtain a second physical coordinate parameter of any of the lamp beads according to the second initial coordinate parameter and the standard coordinate parameter, and configured to calculate a coordinate offset of any of the lamp beads according to the first physical coordinate parameter and the second physical coordinate parameter.

9. The defect detection system as claimed in claim 8, wherein, The obtaining unit is specifically configured to obtain the standard image of the micro display; obtain the standard coordinate parameters corresponding to the plurality of lamp beads according to the standard image; obtain a correlation coefficient between the standard coordinate parameters and the first initial coordinate parameters; judge whether the correlation coefficient is located in a preset range; If the correlation coefficient is within the preset range, it is determined whether the first initial coordinate parameter of each lamp bead matches the corresponding standard coordinate parameter; If the first initial coordinate parameter of the lamp bead matches the corresponding standard coordinate parameter, it is determined that the first initial coordinate parameter of the lamp bead is a first effective coordinate parameter, and if the first initial coordinate parameter of the lamp bead does not match the corresponding standard coordinate parameter, it is determined that the first initial coordinate parameter of the lamp bead is an abnormal coordinate parameter.

10. The defect detection system as claimed in claim 8, wherein, The computing unit is specifically configured to: obtain the first effective coordinate parameters corresponding to the two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameter; An interpolation method is used to calculate a second effective coordinate parameter corresponding to the abnormal coordinate parameter.

11. The defect detection system as claimed in claim 8, wherein, The computing unit is further specifically configured to: obtain a minimum circumscribed rectangle corresponding to any lamp bead according to the first effective coordinate parameter and the second effective coordinate parameter; A peripheral rectangle located on the periphery of the minimum circumscribed rectangle is obtained according to a preset interval; A transition area in which the boundary of the lamp bead is located is determined from the peripheral rectangle to the direction of the minimum circumscribed rectangle; The transition area is threshold segmented, and boundary pixel points meeting a preset condition are screened; A target rectangle corresponding to any lamp bead is fitted according to the obtained plurality of boundary pixel points; A coordinate value of a center pixel point located in the target rectangle is obtained; A first physical coordinate parameter of any lamp bead is obtained according to the coordinate value of the center pixel point and a preset calibration coefficient.

12. The defect detection system as claimed in claim 8, wherein, The computing unit is further specifically configured to: calculate a difference value between the standard coordinate parameter of any lamp bead and the corresponding second initial coordinate parameter; A second physical coordinate parameter of any lamp bead is calculated according to the second initial coordinate parameter, the difference value, and a frame value corresponding to the second initial image.

13. The defect detection system as claimed in claim 8, wherein, The defect detection system further comprises a matching module, which is configured to divide each lamp bead on the first initial image into a sub-region according to the first effective coordinate parameter and the second effective coordinate parameter. The matching module is further configured to obtain feature information of the lamp bead in any sub-region; The matching module is further configured to calculate a restored coordinate parameter of the lamp bead according to the feature information; The matching module is further configured to obtain a restored image according to the restored coordinate parameter.

14. The defect detection system as claimed in claim 8, wherein, The defect detection system further comprises a detection module, which is configured to perform difference between the restored image and the standard image to obtain a difference image; The detection module is further configured to perform image preprocessing on the difference image; The detection module is further configured to perform binaryzation processing on the difference image after image preprocessing; The detection module is further configured to perform connected region marking on the difference image after binaryzation processing; The detection module is further configured to obtain feature information of each connected region; The detection module is further configured to determine whether the lamp bead located in any connected region has a defect according to the feature information.

15. An apparatus for detecting defects of a micro display, comprising a workbench for placing a micro display to be detected and a system for detecting defects of the micro display, the system for detecting defects of the micro display comprising: an acquisition unit configured to acquire a first initial image and a second initial image of the micro display, the first initial image being an image captured by a camera in a burst mode, the second initial image being an image captured by a positioning assembly, and configured to acquire first initial coordinate parameters corresponding to a plurality of lamp beads from the first initial image and second initial coordinate parameters corresponding to the plurality of lamp beads from the second initial image; a calculation unit configured to acquire first effective coordinate parameters and abnormal coordinate parameters in the first initial coordinate parameters according to standard coordinate parameters corresponding to the plurality of lamp beads in a preset standard image, and acquire second effective coordinate parameters corresponding to the abnormal coordinate parameters according to the first effective coordinate parameters, and configured to acquire first physical coordinate parameters of any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters, and second physical coordinate parameters of any lamp bead according to the second initial coordinate parameters and the standard coordinate parameters, and calculate a coordinate offset of any lamp bead according to the first physical coordinate parameters and the second physical coordinate parameters.

16. The defect detection apparatus as claimed in claim 15, wherein The acquisition unit is specifically configured to acquire the standard image of the micro display. The standard coordinate parameters corresponding to the plurality of lamp beads are acquired according to the standard image. A correlation coefficient of the standard coordinate parameters and the first initial coordinate parameters is acquired. It is determined whether the correlation coefficient is within a preset range. If the correlation coefficient is within the preset range, it is determined whether the first initial coordinate parameters of each lamp bead match corresponding standard coordinate parameters. If the first initial coordinate parameters of the lamp bead match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp bead are determined as first effective coordinate parameters, and if the first initial coordinate parameters of the lamp bead do not match the corresponding standard coordinate parameters, the first initial coordinate parameters of the lamp bead are determined as abnormal coordinate parameters.

17. The defect detection apparatus as claimed in claim 15, wherein The calculation unit is specifically configured to acquire the first effective coordinate parameters corresponding to two lamp beads adjacent to the lamp bead corresponding to the abnormal coordinate parameters. An interpolation method is used to calculate second effective coordinate parameters corresponding to the abnormal coordinate parameters.

18. The defect detection apparatus as claimed in claim 15, wherein The calculation unit is further specifically configured to acquire a minimum circumscribed rectangle corresponding to any lamp bead according to the first effective coordinate parameters and the second effective coordinate parameters. A peripheral rectangle located at a periphery of the minimum circumscribed rectangle is acquired according to a preset interval. A transition area in which a boundary of the lamp bead is located is determined from a direction of the minimum circumscribed rectangle to the peripheral rectangle. Threshold segmentation is performed on the transition area, and boundary pixel points meeting a preset condition are screened. A target rectangle corresponding to any lamp bead is fitted according to the acquired plurality of boundary pixel points. A coordinate value of a center pixel point located in the target rectangle is acquired. According to the coordinate value of the center pixel point and a preset calibration coefficient, a first physical coordinate parameter of any of the lamp beads is obtained.

19. The defect detection apparatus as claimed in claim 15, wherein The calculation unit is also specifically configured to calculate a difference between the standard coordinate parameter of any of the lamp beads and the corresponding second initial coordinate parameter. According to the second initial coordinate parameter, the difference, and a frame value corresponding to the second initial image that is obtained, a second physical coordinate parameter of any of the lamp beads is calculated.

20. The defect detection apparatus as claimed in claim 15, wherein The defect detection system further comprises a matching module, which is configured to divide each of the lamp beads on the first initial image into a sub-region according to the first effective coordinate parameter and the second effective coordinate parameter. The matching module is further configured to obtain feature information of the lamp beads in any of the sub-regions. The matching module is further configured to calculate a restored coordinate parameter of the lamp beads according to the feature information. The matching module is further configured to obtain a restored image according to the restored coordinate parameter.

Images