Disclosure of Invention
In order to solve the problem of poor universality of the traditional liquid crystal display detection method, the invention provides a detection method and a detection system for cell in-plane and edge defects.
In order to solve the above technical problems, the present invention provides a method for detecting in-cell and edge defects, comprising the following steps:
Acquiring a first image aiming at a liquid crystal screen, wherein the first image is obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a first set value;
According to the first image, determining an edge to-be-detected area on the first image, wherein the edge to-be-detected area comprises at least one of a grinding area, a corner area, a mark point area and a defect area, the grinding area is an area with grinding marks on the liquid crystal screen, the corner area is an area at the corner of the liquid crystal screen, the mark point area is an area where mark points are manually drawn on the liquid crystal screen, and the defect area is an area with damages, spots or scratches on the liquid crystal screen;
detecting the edge of the liquid crystal screen according to the area to be detected of the edge to obtain a first detection result;
acquiring a second image aiming at the liquid crystal screen, wherein the second image is obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a second set value, and the first set value is smaller than the second set value;
Determining a panel display area on the second image according to the second image, wherein the panel display area is an area which is used for displaying a picture and is used for removing an edge part on the liquid crystal screen;
according to the panel display area, determining characteristic points meeting setting conditions, wherein the setting conditions are pixel points in the panel display area, the pixel values of which are larger than the area standard deviation or the gray values of which are larger than the gray threshold;
determining a display to-be-detected area according to the characteristic points meeting the set conditions, wherein the display to-be-detected area is an area with defects on the panel display area;
and detecting the cell surface of the liquid crystal screen according to the display to-be-detected area to obtain a second detection result.
The method for detecting the defects in the cell surface and the edges has the advantages that the edge to-be-detected area is determined through the first image, the edge of the liquid crystal screen is detected through the edge to-be-detected area, the panel display area is determined through the second image, the area with the defects on the panel display area is extracted to obtain the display to-be-detected area, the cell surface of the liquid crystal screen is detected through the display to-be-detected area, and for different liquid crystal screens, only line scanning cameras for acquiring the first image and the second image are needed to be replaced.
Based on the technical scheme, the detection method for the cell in-plane and edge defects can be improved as follows.
Further, determining the edge to-be-detected area on the first image according to the first image includes:
And determining an edge region to be detected on the first image through a deep learning segmentation algorithm according to the first image, wherein the deep learning segmentation algorithm is Bisenet-v2 segmentation network.
The method has the advantages that the characteristics can be automatically extracted aiming at various defects through the deep learning segmentation algorithm, and the method has good detection rate for the defect characteristics, so that the defect characteristics on the first image are extracted by utilizing the deep learning to obtain the edge region to be detected.
Further, when the edge to be detected area includes a grinding area, the method further includes:
Determining the width of grinding traces in the grinding area according to the grinding area;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the width of the grinding trace is larger than the preset width, judging that the first detection result is that the edge of the liquid crystal display has defects.
The further scheme has the beneficial effect that when the edge to-be-detected area is a grinding area, the edge of the liquid crystal screen is detected according to the width of the grinding trace.
Further, when the edge to be detected area includes an angular area, the method further includes:
constructing a preset coordinate system according to the region to be detected of the edge;
According to the corner area, determining an abscissa and an ordinate of corners in the corner area on a preset coordinate system;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the abscissa is greater than the first threshold or the ordinate is greater than the second threshold, judging that the first detection result is that the edge of the liquid crystal display has defects.
The further scheme has the beneficial effects that when the region to be detected of the edge is an angular region, the edge of the liquid crystal screen is detected according to the positions of the horizontal coordinate and the vertical coordinate of the corners in the angular region on a preset coordinate system.
Further, when the edge to-be-detected area includes a mark point area, the method further includes:
determining an edge line through a straight line fitting algorithm according to the marked point area;
Determining the distance between the mark point and the edge line according to the positions of the edge line and the mark point in the mark point area;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the distance is greater than the third threshold value, judging that the first detection result is that the edge of the liquid crystal display has defects.
The further scheme has the beneficial effect that when the area to be detected of the edge is the mark point area, the edge of the liquid crystal screen is detected through the distance between the edge line and the mark point.
Further, detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, including:
if the edge to-be-detected area comprises a defect area, judging that the first detection result is that the edge of the liquid crystal display has defects.
The further scheme has the beneficial effect that when the edge detection area is a defect area, the edge of the liquid crystal display screen is directly judged to have the defect.
Further, the determining, according to the panel display area, the feature points meeting the setting conditions includes:
According to the panel display area, determining an area standard deviation through a first formula, wherein the first formula is as follows:
Wherein δ represents the area standard deviation, xi represents the i-th pixel point in the panel display area, N represents the total number of pixel points in the panel display area, and l represents the panel display area;
And determining the pixel points with pixel values larger than the area standard deviation in each pixel point in the panel display area as the characteristic points meeting the set condition.
The technical scheme has the beneficial effects that the area standard deviation is determined through the first formula, when the pixel value of the pixel point in the panel display area is larger than the area standard deviation, the pixel point is indicated to have abnormality, and the pixel point is taken as the feature point meeting the condition.
Further, detecting the cell surface of the liquid crystal screen according to the display to-be-detected area to obtain a second detection result, including:
Determining classification labels according to the display to-be-detected areas through a deep learning classification algorithm, wherein the classification labels are background classes, damaged classes, spots classes, scratches classes or water stains classes;
When the classification labels are of the damage type, the spot type, the scratch type or the water stain type, judging that the second detection result is that the cell surface of the liquid crystal screen has defects.
The deep learning classification algorithm has the advantages that the deep learning classification algorithm can directly judge the classification of the characteristics of the display region to be detected without manually extracting the characteristics, so that the deep learning classification algorithm is utilized to directly classify the characteristics of the display region to be detected to obtain the classification label, and the cell surface is detected through the classification label.
In a second aspect, the present invention provides a detection system for in-cell and edge defects, comprising:
The first image acquisition module is used for acquiring a first image aiming at the liquid crystal screen, wherein the first image is an image obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a first set value;
The edge to-be-detected area acquisition module is used for determining an edge to-be-detected area on the first image according to the first image, wherein the edge to-be-detected area comprises at least one of a grinding area, a corner area, a mark point area and a defect area, the grinding area is an area with grinding marks on the liquid crystal screen, the corner area is an area at the corner of the liquid crystal screen, the mark point area is an area where a mark point is manually drawn on the liquid crystal screen, and the defect area is an area with damages, spots or scratches on the liquid crystal screen;
the first detection result acquisition module is used for detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result;
the second image acquisition module is used for acquiring a second image aiming at the liquid crystal screen, wherein the second image is an image obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a second set value, and the first set value is smaller than the second set value;
The panel display area acquisition module is used for determining a panel display area on the second image according to the second image, wherein the panel display area is an area which is used for displaying a picture and is used for removing an edge part on the liquid crystal screen;
The characteristic point acquisition module is used for determining characteristic points meeting set conditions according to the panel display area, wherein the set conditions are pixel points with pixel values larger than the area standard deviation or gray values larger than the gray threshold value in the panel display area;
the display to-be-detected area acquisition module is used for determining a display to-be-detected area according to the characteristic points meeting the set conditions, wherein the display to-be-detected area is an area with defects on the panel display area;
And the second detection result acquisition module is used for detecting the cell surface of the liquid crystal screen according to the display to-be-detected area to obtain a second detection result.
In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a program stored in the memory and running on the processor, where the processor implements the steps of a method for detecting in-cell and edge defects as described above when the processor executes the program.
In a fourth aspect, the present invention also provides a computer readable storage medium having instructions stored therein which, when executed on a terminal device, cause the terminal device to perform steps of a method for detecting in-cell and edge defects.
Detailed Description
The following examples are further illustrative and supplementary of the present invention and are not intended to limit the invention in any way.
The following describes a method and a system for detecting in-cell and edge defects according to an embodiment of the present invention with reference to the accompanying drawings.
The method for detecting the in-cell and edge defects is applied to terminal equipment, the terminal equipment is taken as an execution main body in the scheme of the application, the scheme of the application is explained, the terminal equipment can be a computer, a server and the like and is used for executing the steps of the method for detecting the in-cell and edge defects, the terminal equipment is also connected with each line scanning camera, and each line scanning camera is used for acquiring a first image and a second image of a liquid crystal screen.
The detection of the liquid crystal screen mainly comprises two aspects, namely edge detection of the liquid crystal screen and detection of a cell surface, namely a panel display area.
Optionally, since the edge detection of the lcd needs to measure the parameters of each side (the parameters include the width of the grinding trace of the grinding area, the abscissa and ordinate of the corners in the preset coordinate system, the positions of the edge line and the mark point) and detect the defects, the detection time is long, based on this, two channels can be set to perform edge detection simultaneously, so that the detection time is shortened while each side of the lcd can be detected, as shown in fig. 1, the lcd 1 enters the channel a and the channel B to perform edge detection respectively (in order to shorten the detection time, when one channel a performs edge detection on the lcd, the other channel B may perform edge detection on the other lcd, and then exchange the lcd of the channel a and the channel B), in the channel a, the first image of the liquid crystal screen 1 is acquired through the line scanning cameras 2-1 and 2-2, the first edge of the liquid crystal screen 1 is detected through the detection method for the in-cell surface and edge defects provided by the scheme of the application, then the first image of the liquid crystal screen 1 is acquired through the line scanning cameras 3-1 and 3-2 through rotation, the second edge of the liquid crystal screen 1 is detected through the detection method for the in-cell surface and edge defects provided by the scheme of the application, and so on, the liquid crystal screen 1 enters the channel B, the first image of the liquid crystal screen 1 is acquired through the line scanning cameras 4-1 and 4-2, the third edge of the liquid crystal screen 1 is detected through the detection method for the in-cell surface and edge defects provided by the scheme of the application, the first image of the liquid crystal screen 1 is acquired through the line scanning cameras 5-1 and 5-2 through rotation, the fourth edge of the liquid crystal screen 1 is detected by the detection method for the in-cell and edge defects provided by the scheme of the application, so far, four edges of the liquid crystal screen 1 are detected, finally, the second image of the liquid crystal screen 1 is acquired by the line scanning cameras 6-1 and 6-2, and the in-cell and edge defects of the liquid crystal screen 1 are detected by the detection method provided by the scheme of the application, so that all detection of the liquid crystal screen is completed.
As shown in fig. 2, the method for detecting the in-cell and edge defects provided by the invention comprises the following steps:
S1, acquiring a first image aiming at a liquid crystal screen, wherein the first image is obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a first set value;
s2, determining an edge to-be-detected area on the first image according to the first image, wherein the edge to-be-detected area comprises at least one of a grinding area, a corner area, a mark point area and a defect area, the grinding area is an area with grinding marks on the liquid crystal screen, the corner area is an area at the corner of the liquid crystal screen, the mark point area is an area where a mark point is manually drawn on the liquid crystal screen, and the defect area is an area with damage, spots or scratches on the liquid crystal screen;
S3, detecting the edge of the liquid crystal screen according to the area to be detected of the edge to obtain a first detection result;
S4, acquiring a second image aiming at the liquid crystal screen, wherein the second image is obtained by scanning the liquid crystal screen through a line scanning camera with resolution lower than a second set value, and the first set value is smaller than the second set value;
s5, determining a panel display area on the second image according to the second image, wherein the panel display area is an area which is used for displaying a picture and is used for removing an edge part on the liquid crystal screen;
S6, determining characteristic points meeting set conditions according to the panel display area, wherein the set conditions are pixel points with pixel values larger than the area standard deviation or gray values larger than the gray threshold value in the panel display area;
S7, determining a display area to be detected according to the characteristic points meeting the set conditions, wherein the display area to be detected is an area with defects on the panel display area;
and S8, detecting the cell surface of the liquid crystal screen according to the display region to be detected, and obtaining a second detection result.
Optionally, determining the region to be detected of the edge on the first image according to the first image includes:
And determining an edge region to be detected on the first image through a deep learning segmentation algorithm according to the first image, wherein the deep learning segmentation algorithm is Bisenet-v2 segmentation network.
In this embodiment, the Bisenet-v2 segmentation network includes a spatial path with a small step size for preserving spatial position information so as to generate a high-resolution feature map, a semantic path with a fast downsampling rate for obtaining a feature map of an objective receptive field of an original image after the spatial path, and a feature fusion module for fusing outputs of the spatial path and the semantic path, and outputting a final result to achieve a balance between speed and accuracy.
Based on the above, the first image is input to the Bisenet-v2 dividing network, the edge to-be-detected area is output, the edge to-be-detected area includes at least one of a grinding area, an angle area, a mark point area and a defect area, for example, as shown in fig. 3, the first image is input to the Bisenet-v2 dividing network, to obtain the edge to-be-detected area corresponding to the grinding area shown in fig. 4, at this time, the line width in fig. 4 is measured to be the width of the grinding trace, and the edge of the liquid crystal screen can be detected according to the width of the grinding trace.
Optionally, when the edge to be detected area includes a grinding area, the method further includes:
Determining the width of grinding traces in the grinding area according to the grinding area;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the width of the grinding trace is larger than the preset width, judging that the first detection result is that the edge of the liquid crystal display has defects.
In this embodiment, in the production process of the liquid crystal display, a grinding process is involved, the edge portion of the liquid crystal display is ground, so that the edge is smoother, and a part of the liquid crystal display is defective due to defects of the liquid crystal display or defects of grinding equipment, during the grinding process, grinding marks are left, so that the liquid crystal display cannot meet a qualification standard, therefore, as shown in fig. 4, the grinding marks (a line in the middle portion of fig. 4) obtained by a Bisenet-v2 partition network are required to be detected, if the width of the grinding marks is greater than a preset width, the first detection result is determined as that the edge of the liquid crystal display is defective, and if the width of the grinding marks is less than the preset width, the edge detection of the liquid crystal display is determined to be qualified.
Optionally, the preset width is a specification requirement defined by a customer.
Optionally, when the edge to be detected area includes an angular area, the method further includes:
constructing a preset coordinate system according to the region to be detected of the edge;
According to the corner area, determining an abscissa and an ordinate of corners in the corner area on a preset coordinate system;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the abscissa is greater than the first threshold or the ordinate is greater than the second threshold, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, because the sizes of the images of the first images are the same, the sizes of the images of the acquired corner areas are the same, and the placement positions of the liquid crystal screens in the channels are the same, therefore, corners of different liquid crystal screens are located at the same positions in the corner areas, a preset coordinate system is constructed based on the same, if the abscissa is greater than a first threshold or the ordinate is greater than a second threshold, the first detection result is judged to be that the edge of the liquid crystal screen is defective, and if the abscissa is smaller than the first threshold and the ordinate is smaller than the second threshold, the first detection result is judged to be that the edge of the liquid crystal screen is detected to be qualified.
Optionally, the first threshold and the second threshold are custom specifications of the client.
Optionally, when the area to be detected of the edge includes a mark point area, the method further includes:
determining an edge line through a straight line fitting algorithm according to the marked point area;
Determining the distance between the mark point and the edge line according to the positions of the edge line and the mark point in the mark point area;
Detecting the edge of the liquid crystal screen according to the edge to-be-detected area to obtain a first detection result, wherein the detection result comprises the following steps:
if the distance is greater than the third threshold value, judging that the first detection result is that the edge of the liquid crystal display has defects.
In this embodiment, the straight line fitting algorithm may employ a RANSAC method, where the edge line is determined mainly by the following method:
S11, randomly selecting two points according to the edge line area, and determining a line l by the two points;
S12, determining the geometric distance between the rest points in the edge line area and the straight line l according to a threshold t, taking the points with the geometric distance smaller than t as a data point set S (l), and calling the data point set S (l) as a consistent set of the straight line l, wherein each point in the data point set S (l) is called an inner point;
s13, repeating random selection for a plurality of times, namely S11-S12, to obtain a consistent set S (l 1), S (l 2) and S (ln) corresponding to straight lines l1, l 2;
S14, taking the data point set with the largest number of inner points in each consistent set as a target data point set, and fitting the inner points in the target data point set into a straight line by least square as an edge line.
Optionally, if the distance between the mark point and the edge line is greater than the third threshold, the first detection result is judged to be that the edge of the liquid crystal screen has a defect, and if the distance between the mark point and the edge line is less than the third threshold, the first detection result is judged to be that the edge of the liquid crystal screen is qualified.
Optionally, the third threshold is a custom specification of the customer.
Optionally, detecting the edge of the liquid crystal screen according to the area to be detected to obtain a first detection result, including:
if the edge to-be-detected area comprises a defect area, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, if the edge detection area is a defect area, it is directly determined that the first detection result is that the edge of the liquid crystal display has a defect, as shown in fig. 5, and in order to obtain a first image corresponding to the defect area, the first image is input Bisenet-v2 into a segmentation network to obtain an edge to-be-detected area corresponding to the defect area, as shown in fig. 6, where at this time, a defect can be clearly seen from the solid background of fig. 6, that is, it is determined that the first detection result is that the edge of the liquid crystal display has a defect.
Optionally, when interference such as water stain exists on the liquid crystal screen, the water stain can affect the recognition of the Bisenet-v2 segmentation network to the first image, so that misjudgment is caused, therefore, the output result of the Bisenet-v2 segmentation network can be screened again by using a deep learning classification algorithm, the interference image is removed, and the accuracy of the edge detection of the liquid crystal screen is improved.
Optionally, the determining, according to the panel display area, the feature points meeting the set condition includes:
According to the panel display area, determining an area standard deviation through a first formula, wherein the first formula is as follows:
Wherein δ represents the area standard deviation, xi represents the i-th pixel point in the panel display area, N represents the total number of pixel points in the panel display area, and l represents the panel display area;
And determining the pixel points with pixel values larger than the area standard deviation in each pixel point in the panel display area as the characteristic points meeting the set condition.
Optionally, the gray threshold is a custom specification of the client, the gray value of the panel display area is a numerical value of 0-255, and the display area to be detected corresponding to the gray value can be determined through a second formula, where the second formula is:
wherein g (x, y) is a gray value of a pixel point of a display area of the panel, f (x, y) is a gray threshold, and T represents a gray threshold.
Optionally, because the background of the panel display area is single, if a defect occurs, the pixel value and the gray value of the image of the panel display area are abnormal, so that the area standard deviation and the gray threshold can be determined by adopting the first formula and the second formula, and the area to be detected is determined and displayed through the area standard deviation and the gray threshold.
Optionally, detecting the cell surface of the liquid crystal screen according to the display to-be-detected area to obtain a second detection result, including:
Determining classification labels according to the display to-be-detected areas through a deep learning classification algorithm, wherein the classification labels are background classes, damaged classes, spots classes, scratches classes or water stains classes;
When the classification labels are of the damage type, the spot type, the scratch type or the water stain type, judging that the second detection result is that the cell surface of the liquid crystal screen has defects.
Optionally, the deep learning classification algorithm adopts repvgg models of a CNN convolutional neural network, and inputs a display region to be detected into repvgg models to obtain classification labels, wherein the background type indicates that the display region to be detected is a background image, namely, the second detection result is that the cell surface of the liquid crystal screen has defects, the breakage type indicates that the display region to be detected has damages, namely, the second detection result is that the cell surface of the liquid crystal screen has defects, the spot type indicates that the display region to be detected has spots, namely, the second detection result is that the cell surface of the liquid crystal screen has defects, the scratch type indicates that the display region to be detected has scratches, namely, the second detection result is that the cell surface of the liquid crystal screen has defects, and the water stain type indicates that the display region to be detected has water stains, namely, the second detection result is that the cell surface of the liquid crystal screen has defects.
As shown in fig. 7, the embodiment of the present invention further provides a system for detecting in-cell and edge defects, including:
a first image obtaining module 201, configured to obtain a first image for a liquid crystal screen, where the first image is an image obtained by scanning the liquid crystal screen with a line scanning camera with a resolution lower than a first set value;
The edge to-be-detected area obtaining module 202 is configured to determine an edge to-be-detected area on the first image according to the first image, where the edge to-be-detected area includes at least one of a grinding area, a corner area, a mark point area and a defect area, the grinding area is an area where grinding marks exist on the liquid crystal screen, the corner area is an area at a corner of the liquid crystal screen, the mark point area is an area where mark points are manually drawn on the liquid crystal screen, and the defect area is an area where damage, spots or scratches exist on the liquid crystal screen;
the first detection result obtaining module 203 is configured to detect an edge of the liquid crystal screen according to the edge to-be-detected area, so as to obtain a first detection result;
A second image obtaining module 204, configured to obtain a second image for the liquid crystal screen, where the second image is an image obtained by scanning the liquid crystal screen with a line scanning camera with a resolution lower than a second set value, and the first set value is smaller than the second set value;
A panel display area obtaining module 205, configured to determine, according to the second image, a panel display area on the second image, where the panel display area is an area on the liquid crystal screen from which an edge portion is removed and is used for displaying a picture;
The feature point obtaining module 206 is configured to determine, according to the panel display area, a feature point that meets a setting condition, where the setting condition is a pixel point in the panel display area where a pixel value is greater than an area standard deviation or where a gray value is greater than a gray threshold;
A display region to be detected acquisition module 207, configured to determine a display region to be detected according to the feature points that meet the set condition, where the display region to be detected is a region with a defect on the panel display region;
And the second detection result obtaining module 208 is configured to detect a cell surface of the liquid crystal screen according to the display to-be-detected area, so as to obtain a second detection result.
Optionally, the edge to-be-detected area obtaining module 202 is specifically configured to:
And determining an edge region to be detected on the first image through a deep learning segmentation algorithm according to the first image, wherein the deep learning segmentation algorithm is Bisenet-v2 segmentation network.
Optionally, the first detection result obtaining module 203 further includes:
The grinding trace acquisition module is used for determining the width of the grinding trace in the grinding region according to the grinding region;
the first detection result obtaining module 203 is specifically configured to:
if the width of the grinding trace is larger than the preset width, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, the first detection result obtaining module 203 further includes:
The preset coordinate system acquisition module is used for constructing a preset coordinate system according to the region to be detected of the edge;
the coordinate acquisition module is used for determining the horizontal coordinate and the vertical coordinate of the corners in the corner area on a preset coordinate system according to the corner area;
The first detection result obtaining module 203 is specifically configured to:
if the abscissa is greater than the first threshold or the ordinate is greater than the second threshold, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, the first detection result obtaining module 203 further includes:
The edge line acquisition module is used for determining an edge line through a straight line fitting algorithm according to the marked point area;
the distance acquisition module is used for determining the distance between the mark point and the edge line according to the positions of the mark point in the edge line and the mark point area;
The first detection result obtaining module 203 is specifically configured to:
if the distance is greater than the third threshold value, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, the first detection result obtaining module 203 is specifically configured to:
if the edge to-be-detected area comprises a defect area, judging that the first detection result is that the edge of the liquid crystal display has defects.
Optionally, the feature point obtaining module 206 specifically includes:
The area standard deviation obtaining module is used for determining an area standard deviation according to a panel display area through a first formula, wherein the first formula is as follows:
Wherein δ represents the area standard deviation, xi represents the i-th pixel point in the panel display area, N represents the total number of pixel points in the panel display area, and l represents the panel display area;
and the first judging module is used for determining the pixel points with the pixel values larger than the area standard deviation in each pixel point in the panel display area as the characteristic points meeting the set conditions.
Optionally, the second detection result obtaining module 208 specifically includes:
The classification label module is used for determining classification labels according to the display to-be-detected area through a deep learning classification algorithm, wherein the classification labels are background classes, damaged classes, spots classes, scratches classes or water stains classes;
and the second judging module is used for judging that the cell surface of the liquid crystal screen has defects when the classification labels are of the damage type, the spot type, the scratch type or the water stain type.
The electronic equipment comprises a memory, a processor and a program stored in the memory and running on the processor, wherein the processor realizes part or all of the steps of the method for detecting the in-cell and edge defects when executing the program.
The electronic device may be a computer, and correspondingly, the program is computer software, and the parameters and steps in the above-mentioned electronic device of the present invention may refer to the parameters and steps in the above-mentioned embodiment of a method for detecting a cell in-plane and edge defects, which are not described herein.
Those skilled in the art will appreciate that the present invention may be implemented as a system, method, or computer program product. Accordingly, the present disclosure may be embodied in either entirely hardware, entirely software (including firmware, resident software, micro-code, etc.), or in a combination of hardware and software, referred to herein generally as a "circuit," module, "or" system. Furthermore, in some embodiments, the invention may also be embodied in the form of a computer program product in one or more computer-readable media, which contain computer-readable program code. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.