Calibration method and device based on screen body foreign matter detection and storage medium

Abstract

The invention discloses a calibration method and device based on screen body foreign matter detection and a storage medium. The calibration method and device are used for improving the calibration effect of a lighting detection camera and a layered camera in screen body foreign matter detection. Obtaining a calibration screen shot image; carrying out calibration point coordinate extraction to generate calibration point position information; the calibration screen is moved to a station of the layered camera through the displacement carrying table; focusing the layered camera to a pixel layer of the calibration screen, collecting a central calibration point and a positioning calibration point on the calibration screen, and generating a calibration point image; converting the calibration point image into a grayscale image; determining a gray threshold value, and performing pixel point screening processing on the calibration point area of the gray image through the gray threshold value to generate a binary image; performing adjacent pixel gap elimination processing on the binarized image, extracting positioning pixel point coordinates from the binarized image according to the feature information of the positioning pixel points, and generating positioning pixel point position information; calculating position deviation information; and calculating a position calibration matrix.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of display screen foreign matter detection, and in particular, to a calibration method, device and storage medium based on screen foreign matter detection. Background Art

[0002] With the development of LCD screens, LCD screens gradually have more functionality. The increase in functionality will increase the number of layers of LCD screens, or improve each layer. This makes the structure of LCD screens complex. Existing LCD screens may include multiple components such as CG cover plate, OCA glue, upper polarizer, TFT, CF, lower polarizer and backlight. Therefore, during the production and manufacturing process of LCD screens, foreign matter may be generated in many production links due to the complex structure. These foreign matter may be generated between any two adjacent layers. However, the impact of these foreign matter on the quality of the final product is inconsistent. For example, the impact of backlight foreign matter is not as great as that of polarizer foreign matter. This makes it very important to determine the location and layer position of foreign matter.

[0003] During the production and preparation process, although each link is strictly controlled, it is inevitable that there will be mixing between the two layers, resulting in a screen with foreign body defects. Therefore, foreign body defect detection has become an indispensable link in the LCD screen production process. Traditional AOI spot lighting inspection uses a camera to align to the pixel layer of the LCD screen and take an image of the entire LCD screen with preset pixels. In the image obtained in this way, foreign bodies between two adjacent layers cannot be accurately distinguished, and foreign bodies on the upper and lower sides of the pixel layer cannot be accurately detected. Therefore, the inspection performed by the above-mentioned captured images can only be controlled according to the same standard to detect foreign bodies. Some foreign bodies that have little impact on the quality of the LCD screen will also be screened according to high standards, resulting in huge waste.

[0004] In order to reduce losses and improve product utilization. After the lighting inspection is completed, a defect layering inspection can be added to accurately distinguish the location of foreign matter. The layering inspection uses a 25M camera with a depth of field of 75um to move to the top of the foreign matter. By adjusting the height of the layering camera, the clearest position of the defect is obtained and the layer of the defect is calculated. In order for the layering camera to be accurately moved to the top of the foreign body defect, the lighting inspection camera and the layering camera need to calculate the positional relationship between each other through calibration. However, with the continuous updating and improvement of LCD screens, the functionality of LCD screens has been continuously developed, and structures such as the internal circuit of the display screen have been added, and the degree of bending and folding of the display screen has also been upgraded. These have made the LCD screen structure gradually complex. The traditional camera calibration method is easily affected by these structures, which affects the point positioning process of the two cameras during calibration, thereby reducing the calibration effect of the lighting inspection camera and the layering camera in the screen foreign body detection. Summary of the invention

[0005] The present application discloses a calibration method, device and storage medium based on screen foreign matter detection, which are used to improve the calibration effect of a lighting detection camera and a layered camera in screen foreign matter detection.

[0006] The first aspect of the present application proposes a calibration method based on screen foreign body detection, comprising: Input the calibration image to the calibration screen at the center of the stage, use the lighting detection camera to shoot, and obtain the calibration screen captured image. There are 1 center calibration point and several positioning calibration points on the calibration image. The center calibration point is provided with a cross mark area with a different grayscale from the center calibration point. The positioning calibration point is provided with a cross mark area with a different grayscale from the positioning calibration point. A positioning pixel point is provided at the exact center of the cross mark area. The center calibration point is located at the center of the calibration image. There are center calibration points and positioning calibration points at corresponding positions of the calibration screen captured image. Extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; The calibration screen is moved to the station of the layered camera by means of a displacement stage, and the layered camera is a color camera; The layered camera is focused to the pixel layer of the calibration screen according to the layer information of the calibration screen, and the center calibration point and the positioning calibration point on the calibration screen are collected by the layered camera according to the calibration point position information to generate a calibration point image; Convert the calibration point image into a grayscale image; Determine the grayscale threshold, and use the grayscale threshold to perform pixel screening processing on the calibration point area of ​​the grayscale image to generate a binary image; Performing adjacent pixel gap elimination processing on the binary image, and extracting the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points, and generating the position information of the positioning pixel points; Calculate the position deviation information according to the position information of the positioning pixel point and the center coordinate position of the layered camera on the calibration point image; The position calibration matrix of the lighting detection camera and the layered camera is calculated through the calibration point position information, the positioning pixel position information and the position deviation information.

[0007] Optionally, extracting the coordinates of the calibration points from the image captured by the calibration screen according to the preset calibration point feature quantity information to generate the calibration point position information includes: Binarize the image captured by the calibration screen, and remove the cross mark on the image captured by the calibration screen through morphological expansion; The calibration point feature information of each block on the calibration screen shooting image is filtered to obtain all calibration point sets, and the calibration point feature information includes true roundness information and area information; The row and column coordinates of the 9 calibration points are sorted in the order of row first and column later to generate the calibration point position information.

[0008] Optionally, the calibration screen is a display screen provided with a circuit area, and the calibration point feature information also includes circuit structure information; The calibration point feature information of each block on the calibration screen image is filtered out to obtain all calibration point sets, including: Marking the circuit area on the image captured by the calibration screen according to the circuit distribution data in the circuit structure information; Determine the interference area in the circuit area according to the circuit reflection data and the circuit morphology data in the circuit structure information; Determine a set of preliminary screening calibration points that meet the calibration points in each block on the image captured by the calibration screen according to the true roundness information and the area information; The circuit area interference is filtered out from the initial screening calibration point set through the interference area, and the calibration point set that meets the requirements is selected.

[0009] Optionally, determining a grayscale threshold, and performing pixel point screening processing on a calibration point area of ​​a grayscale image by using the grayscale threshold to generate a binary image includes: When there is a calibration point located on the circuit area, channel data is selected from the RGB channel information of the grayscale image according to the circuit reflection data in the circuit structure information; Generate a grayscale threshold according to the selected channel pixel data; The grayscale image is filtered by grayscale threshold to generate a binary image.

[0010] Optionally, after converting the calibration point image into a grayscale image, determining a grayscale threshold, and performing pixel point screening processing on the calibration point area of ​​the grayscale image by the grayscale threshold, before generating a binary image, the calibration method further includes: Acquire several light reflection images of the calibration screen collected under spherical integral light sources of different brightness; Detecting brightness information of the display screen in the non-circuit area with the calibration image input by a brightness meter; Determine a corresponding target light reflection image among a plurality of light reflection images according to brightness information; Determine the calibration point to be adjusted located in the circuit area; Determine the reflection area corresponding to the calibration point to be adjusted on the target light reflection image; The circuit interference elimination process is performed on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean value of the reflection area.

[0011] Optionally, performing adjacent pixel gap elimination processing on the binary image, and extracting the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points, including: Performing morphological dilation processing on the pixels of the calibration point area in the binary image, connecting the pixels of the calibration point area to form a pixel whole, the calibration point area includes a central calibration point area and a positioning calibration point area; The coordinate position of the positioning pixel point is extracted from the binary image after morphological expansion according to the feature information of the positioning pixel point, and the position information of the positioning pixel point is generated.

[0012] The second aspect of the present application provides a calibration device based on screen foreign body detection, comprising: The first acquisition unit is used to input a calibration image to the calibration screen at the center of the stage, use a lighting detection camera to shoot, and obtain the calibration screen shot image, wherein a central calibration point and a plurality of positioning calibration points are set on the calibration image, a cross Mark area with a grayscale different from that of the central calibration point is set on the central calibration point, a cross Mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point, a positioning pixel point is set at the exact center of the cross Mark area, the central calibration point is located at the center of the calibration image, and there are a central calibration point and a positioning calibration point at the corresponding position of the calibration screen shot image; A first generating unit is used to extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, wherein the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; A displacement unit, used to move the calibration screen to the station of the layered camera through a displacement stage, and the layered camera is a color camera; The second generating unit is used to focus the layered camera to the pixel layer of the calibration screen according to the layer information of the calibration screen, collect the central calibration point and the positioning calibration point on the calibration screen through the layered camera according to the calibration point position information, and generate a calibration point image; A conversion unit, used for converting the calibration point image into a grayscale image; A first determination unit is used to determine a grayscale threshold, and to perform pixel point screening processing on a calibration point area of ​​a grayscale image by using the grayscale threshold to generate a binary image; The third generating unit is used to perform adjacent pixel gap elimination processing on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points; A first calculation unit is used to calculate the position deviation information according to the position information of the positioning pixel point and the center coordinate position of the layered camera on the calibration point image; The second calculation unit is used to calculate the position calibration matrix of the lighting detection camera and the layered camera through the calibration point position information, the positioning pixel point position information and the position deviation information.

[0013] Optionally, the first generating unit includes: A processing module is used to perform binarization processing on the image captured by the calibration screen, and remove the cross mark on the image captured by the calibration screen through morphological expansion; A screening module is used to screen out all calibration point sets from the calibration point feature information of each block on the calibration screen captured image, where the calibration point feature information includes true roundness information and area information; The generation module is used to sort the row and column coordinates of the 9 calibration points in the order of row first and column later to generate the calibration point position information.

[0014] Optionally, the calibration screen is a display screen provided with a circuit area, and the calibration point feature information also includes circuit structure information; Screening modules include: Marking the circuit area on the image captured by the calibration screen according to the circuit distribution data in the circuit structure information; Determine the interference area in the circuit area according to the circuit reflection data and the circuit morphology data in the circuit structure information; Determine a set of preliminary screening calibration points that meet the calibration points in each block on the image captured by the calibration screen according to the true roundness information and the area information; The circuit area interference is filtered out from the initial screening calibration point set through the interference area, and the calibration point set that meets the requirements is selected.

[0015] Optionally, the first determining unit includes: When there is a calibration point located on the circuit area, channel data is selected from the RGB channel information of the grayscale image according to the circuit reflection data in the circuit structure information; Generate a grayscale threshold according to the selected channel pixel data; The grayscale image is filtered by grayscale threshold to generate a binary image.

[0016] Optionally, after the conversion unit and before the first determination unit, the calibration device further includes: A first acquisition unit is used to acquire a plurality of light reflection images collected by the calibration screen under a spherical integral light source of different brightness; A detection unit, used to detect brightness information of a display screen in a non-circuit area input with a calibration image by using a brightness meter; A second determining unit, configured to determine a corresponding target light reflection image among the plurality of light reflection images according to the brightness information; A third determining unit, used to determine a calibration point to be adjusted located in the circuit area; A fourth determining unit, used to determine a reflection area corresponding to the calibration point to be adjusted on the target light reflection image; The elimination unit is used to perform circuit interference elimination processing on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean value of the reflection area.

[0017] Optionally, the third generating unit includes: Performing morphological dilation processing on the pixels of the calibration point area in the binary image, connecting the pixels of the calibration point area to form a pixel whole, the calibration point area includes a central calibration point area and a positioning calibration point area; The coordinate position of the positioning pixel point is extracted from the binary image after morphological expansion according to the feature information of the positioning pixel point, and the position information of the positioning pixel point is generated.

[0018] The third aspect of the present application provides a calibration device based on screen foreign body detection, comprising: Processor, memory, input-output unit, and bus; The processor is connected to the memory, the input and output unit, and the bus; The memory stores a program, and the processor calls the program to execute the first aspect and any optional calibration method of the first aspect.

[0019] A fourth aspect of the present application provides a computer-readable storage medium, on which a program is stored. When the program is executed on a computer, the program executes the first aspect and any optional calibration method of the first aspect.

[0020] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages: In the present application, firstly, a calibration image is input to the calibration screen at the center of the stage, and the calibration screen image is obtained by shooting with a lighting detection camera. A central calibration point and several positioning calibration points are set on the pre-set calibration image, and a cross mark area with a grayscale different from the central calibration point is set on each central calibration point, and a cross mark area with a grayscale different from the positioning calibration point is set on the positioning calibration point. A positioning pixel point is set in the center of the cross mark area, and the central calibration point is located at the center of the calibration image. There are central calibration points and positioning calibration points at corresponding positions in the captured calibration screen image. At this time, different blocks on the calibration screen image have at least one calibration point. Then, according to the preset calibration point feature quantity information, the calibration point coordinates are extracted in the calibration screen image to generate the calibration point position information, and the calibration point information includes the feature quantity information of the central calibration point and the feature quantity information of the positioning calibration point. At this time, the calibration point position information is the calibration point coordinates determined with reference to the lighting detection camera. Next, the calibration screen is moved to the station of the layered camera by the displacement stage. Among them, the layered camera is a color camera. The area captured by the layered camera is limited, but it can capture clearer images. First, according to the hierarchical information of the calibration screen, the layered camera is focused on the pixel layer of the calibration screen. After alignment, the center calibration point and the positioning calibration point on the calibration screen are collected through the layered camera according to the calibration point position information to generate a calibration point image. At this time, the shooting is performed with the layered camera as the reference system. The calibration point image is converted into a grayscale image so that the difference of the calibration points in the entire shooting area is highlighted. Next, the grayscale threshold for distinguishing the calibration point area, the cross Mark area and the positioning pixel point is determined. The grayscale threshold is used to screen the pixels in the calibration point area of ​​the grayscale image to generate a binary image. At this time, because the layered camera focuses on the pixel layer and highlights the gaps between the pixels, it is necessary to eliminate the gaps between adjacent pixels on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the positioning pixel point position information. Finally, the position deviation information is calculated based on the positioning pixel point position information and the center coordinate position of the layered camera on the calibration point image. After obtaining the position deviation, the position calibration matrix of the lighting detection camera and the layered camera can be calculated through the calibration point position information, the positioning pixel position information and the position deviation information.

[0021] First, the image for calibration is taken by the lighting detection camera at the workstation, and then the layered camera is focused on the pixel layer at the workstation of the layered camera to collect images of each calibration point. The location information of the positioning pixel point is determined by processing image conversion, pixel point screening processing within the calibration point, and adjacent pixel gap elimination processing. Then, the deviation is calculated according to the center coordinate position of the layered camera to further reduce the error. Finally, the position calibration matrix is ​​generated by the calibration point position and deviation determined on the two images. The position calibration matrix can be used as the position relationship data of the current layered camera and the lighting detection camera to improve the calibration effect of the lighting detection camera and the layered camera in the screen foreign body detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0023] Figure 1 This is a schematic diagram of an embodiment of a calibration method based on screen foreign matter detection in the present application; Figure 2 A schematic diagram of an embodiment of a method for generating calibration point position information for the present application; Figure 3 A schematic diagram of an embodiment of a method for screening a set of calibration points in the present application; Figure 4 A schematic diagram of an embodiment of a method for selecting pixels in the present application; Figure 5 A schematic diagram of an embodiment of a method for eliminating circuit area interference in the present application; Figure 6 A schematic diagram of an embodiment of a method for generating location pixel position information for the present application; Figure 7 This is a schematic diagram of an embodiment of a calibration device based on screen foreign matter detection in the present application; Figure 8 This is a schematic diagram of another embodiment of a calibration device based on screen foreign matter detection in the present application; Fig. 9 A structural schematic diagram of the machine of this application; Fig.10 A schematic diagram of calibrating an image for this application; Fig.11 A schematic diagram of an image captured for the calibration screen of this application; Fig.12 A schematic diagram of a binary image after feature quantity screening processing in this application; Fig.13 A schematic diagram of the calibration point images of 9 calibration points in this application. DETAILED DESCRIPTION

[0024] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0025] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0026] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0027] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0028] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0029] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0030] In the prior art, in order to reduce losses and improve product utilization. After the lighting detection is completed, a defect layering detection can be added to accurately distinguish the location of the foreign body. The layering detection uses a 25M camera with a depth of field of 75um to move to the top of the foreign body. By adjusting the height of the layering camera, the clearest position of the defect is obtained and the layer of the defect is calculated. In order for the layering camera to be accurately moved to the top of the foreign body defect, the lighting detection camera and the layering camera need to calculate the positional relationship between each other through calibration. However, with the continuous updating and improvement of LCD screens, the functionality of LCD screens has been continuously developed, and structures such as the internal circuit of the display screen have been added, and the degree of bending and folding of the display screen has also been upgraded. These have made the LCD screen structure gradually complex. The traditional camera calibration method is easily affected by these structures, which affects the point positioning process of the two cameras during calibration, thereby reducing the calibration effect of the lighting detection camera and the layering camera in the detection of foreign bodies on the screen.

[0031] Based on this, the present application discloses a calibration method, device and storage medium based on screen foreign matter detection, which are used to improve the calibration effect of the lighting detection camera and the layered camera in screen foreign matter detection.

[0032] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0033] The method of the present application can be applied to a server, a device, a terminal or other devices with logic processing capabilities, and the present application does not limit this. For the convenience of description, the following description is made by taking the execution subject as an example of a terminal.

[0034] See also Figure 1 The present application provides an embodiment of a calibration method based on screen foreign body detection, including: 101. Input a calibration image to the calibration screen at the center of the stage, use a lighting detection camera to shoot, and obtain a calibration screen shot image. A center calibration point and several positioning calibration points are set on the calibration image. A cross mark area with a grayscale different from that of the center calibration point is set on the center calibration point, and a cross mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point. A positioning pixel point is set at the exact center of the cross mark area. The center calibration point is located at the center of the calibration image, and there are a center calibration point and a positioning calibration point at the corresponding position of the calibration screen shot image. In this embodiment, the core is to calculate the position relationship matrix between the lighting detection camera (AOI camera, shooting parameter is 151M) and the layered camera (shooting parameter is 25M). The lighting detection camera is a fixed position, and the layered camera can be moved from the origin position to any position within the specified range. The current calculation method is to extract the row and column coordinates of the nine points of the lighting detection camera through an algorithm, and then move the layered camera to the position of the corresponding calibration point manually or by machine equipment, record the physical distance, and calculate the relationship matrix by matching the coordinates of the nine calibration points (center calibration point and positioning calibration point) one by one. For a general LCD screen, this calibration process takes about 1 hour, which is very unfriendly to on-site production. In order to improve efficiency, this embodiment provides an automatic calibration method that can complete the calibration in a few minutes, greatly improving work efficiency.

[0035] Please refer to Fig. 9 , Fig. 9 The figure is a schematic diagram of the machine. The lighting detection camera (151M) and the layering camera (25M) are located on both sides of the machine. The LCD screen is rotated from the 151M camera station to the 25M camera station through a turntable. The 151M station only needs to take an image once to obtain all calibration points (center calibration point and positioning calibration point), while the 25M camera has a relatively small field of view and can only take pictures one by one. Therefore, we considered that if we can know the physical coordinates of one of the calibration points at the 25M camera station in advance, then we can determine the physical coordinates of the remaining points based on the positional relationship between the calibration points.

[0036] In the actual production process, the size and resolution of the LCD screens of different models are constantly changing, and in order to ensure that the 25M camera can be moved above each calibration point. Taking the above factors into consideration, this embodiment uses a special displacement stage. This displacement stage is rectangular in shape, with scales on the two horizontal and vertical sides of the displacement stage, and high scales are distributed along the center scale 0 to both sides. Taking into account the different sizes of LCD screens, the displacement stage can be flexibly adjusted up and down and left and right to ensure that the LCD screen can be within the active range of the 25M camera. After testing, it was found that when the intersection B of the lines connecting the horizontal and vertical directions along the 0 scale is used as the center of the LCD screen, all models can remain within the active range of the camera, and the distance from the intersection B to the camera origin remains unchanged. Please refer to Fig.10 , Fig.10 Therefore, in this embodiment, by ensuring that the middle point A (center calibration point) among the nine calibration points coincides with the intersection point B, and then reversely calculating the distances between the eight surrounding points (positioning calibration points) and point A (the center point A of the calibration image is the center calibration point) and the origin of the camera, the physical coordinates of the nine points are finally obtained.

[0037] Among them, a cross mark area with a grayscale different from that of the center calibration point is set on the center calibration point, and a cross mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point. The cross mark area is used to detect the position of the calibration point on the calibration screen image, so the cross mark area uses a grayscale different from that of the calibration point. A positioning pixel point is set at the center of the cross mark area. The positioning pixel point is difficult to display in the calibration screen image, but can be clearly displayed under the layered camera. The positioning pixel point is used to locate the center position of the calibration point under the layered camera image. There are 1 center calibration point and several positioning calibration points on the calibration image. When displayed on the calibration screen, 1 center calibration point and several positioning calibration points are also displayed on the calibration screen. After shooting with the lighting detection camera, the generated calibration screen image will also have a center calibration point and a positioning calibration point at the corresponding position.

[0038] It should be noted that the morphological structures of the center calibration point and the positioning calibration point are usually set to be the same. The center calibration point needs to calibrate the image center, while the positioning calibration point is set according to the needs of the researcher.

[0039] 102. Extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, wherein the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; In this embodiment, it is necessary to extract the calibration point coordinates (calibration point position information) from the calibration screen image captured by the lighting detection camera. Specifically, the terminal first inputs the calibration image to the calibration screen at the center of the stage, uses the lighting detection camera to shoot, and obtains the calibration screen image. Then, the calibration point coordinates are extracted based on the calibration point feature information in the calibration screen image to generate the calibration point position information. Finally, the row and column coordinates {Rowi, Coli} of the calibration point under the lighting detection camera are obtained, where i={0-8}, and then the coordinates are passed to the layered camera. The feature quantity information of the calibration points includes information of the calibration points such as area, roundness, etc. For detailed calibration methods, please refer to the subsequent embodiments.

[0040] Specifically, the calibration screen capture image can cover all 9 calibration points, please refer to Fig.11 , Fig.11 This is a schematic diagram of the calibration screen image. The dot matrix (calibration image) used in this embodiment has a clear contrast and little background interference. The calibration points can be extracted by binarization. However, in order to facilitate the layered camera to obtain the center coordinates of the calibration points, a cross mark area is created on each calibration point. Please refer to Fig.12 , Fig.12 It is a binary image after feature filtering.

[0041] There is a difference in the grayscale of the cross mark area and the calibration point. Therefore, the binary area obtained in the calibration screen image needs to be morphologically expanded to remove the cross mark area so that the calibration point becomes a complete circle. Then all the calibration points are screened out by feature quantities, and the row and column coordinates {Rowi, Coli} of the calibration points under the lighting detection camera are obtained, where i={0-8}, and then the coordinates are passed to the layered camera.

[0042] 103. Move the calibration screen to the station of the layered camera by means of a displacement stage, wherein the layered camera is a color camera; After obtaining the calibration point position information in the calibration screen detection image, it is necessary to move the calibration screen to the position of the layered camera through the displacement stage so that the layered camera can take pictures. The layered camera in this embodiment is a color camera.

[0043] 104. Focusing the layered camera to the pixel layer of the calibration screen according to the layer information of the calibration screen, collecting the center calibration point and the positioning calibration point on the calibration screen through the layered camera according to the calibration point position information, and generating a calibration point image; The terminal first focuses the layered camera to the pixel layer of the calibration screen according to the hierarchical information of the calibration screen, and collects each calibration point on the calibration screen through the layered camera and according to the calibration point position information to generate a calibration point image.

[0044] As mentioned above, after the LCD screen rotates to the position of the layered camera, the layered camera first moves to the intersection of the 0 scale line of the stage (i.e. the center of the stage), and the center point of the calibration image is also located here. In this embodiment, the calibration image is burned into each model through PG. Although the resolution of different models is different, when making the calibration map, it is arranged in proportion to the screen size. Therefore, the center position and the eight surrounding points can be multiplied by the length and width according to the ratio to calculate the physical coordinates of the remaining points, and the camera is controlled to move to the corresponding position to take pictures, and finally the image of the nine calibration points is obtained. Please refer to Fig.13 , Fig.13 Schematic diagram of the calibration point image of 9 calibration points.

[0045] As can be seen from the figure, the layered camera cannot capture all the calibration points due to the limited field of view, and cannot even capture the entire calibration point. When making the calibration point, a cross Mark area is designed, and a small positioning pixel point with a size of one display screen pixel is made in the center. In this way, the center position {Xi, Yi} of each calibration point (calibration point in the calibration image) can be calculated through the center point (positioning pixel point), where i={0-8}. In this embodiment, due to the constraint of a small depth of field, the layered camera needs to be uniformly focused on the pixel layer, so that the calibration point can be captured clearly, so that the sub-pixels of the LCD screen can also be clearly presented.

[0046] 105. Convert the calibration point image into a grayscale image; 106. Determine a grayscale threshold, and perform pixel screening processing on the calibration point area of ​​the grayscale image by using the grayscale threshold to generate a binary image; 107. Performing adjacent pixel gap elimination processing on the binary image, and extracting the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points, to generate the position information of the positioning pixel points; 108. Calculate position deviation information according to the position information of the positioning pixel point and the center coordinate position of the layered camera on the calibration point image; In this embodiment, the terminal performs image processing according to the following steps to obtain the center coordinates (position of the positioning pixel point) of each calibration point (central calibration point and positioning calibration point).

[0047] First, convert the calibration point image into a grayscale image (because the 25M layered camera is a color camera). Next, it is necessary to determine the grayscale threshold based on the calibration point area and the cross mark area, and then use the grayscale threshold to filter the pixels in the calibration point area of ​​the grayscale image to generate a binary image. Because when the layered camera focuses on the pixel layer, the pixel grid of the LCD screen is magnified by about 150 times, and the gaps between all pixels are also revealed. The binary image is composed of many independent pixels. In order to distinguish the central positioning pixel, the binary image is expanded to connect the pixels around the center point (the pixels in the calibration area) into a whole. There is a difference between the grayscale of the cross mark area and the grayscale of the positioning pixel, and the difference is large. Usually, the grayscale value of the positioning pixel is set to be close to the calibration area.

[0048] The terminal will perform adjacent pixel gap elimination processing on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points.

[0049] The calculated center coordinates (positioning pixel coordinates) are subtracted from the center coordinates (half width, half height) of the corresponding calibration point image (25M image) to calculate the position deviation of the calibration point from the center of the camera field of view. However, this coordinate difference is a pixel difference, which needs to be divided by the magnification factor of each pixel and then multiplied by the physical size of a single pixel to obtain the physical coordinate deviation of the calibration point {ΔXi, ΔYi}, where i={0-8}.

[0050] 109. Calculate the position calibration matrix of the lighting detection camera and the layered camera through the calibration point position information, the positioning pixel position information and the position deviation information.

[0051] Finally, the terminal calculates the position calibration matrix of the lighting detection camera and the layered camera through the calibration point position information, positioning pixel position information and position deviation information.

[0052] Add {Xi, Yi} (i={0-8}) and {ΔXi, ΔYi} (i={0-8}) to obtain the physical coordinates {Xi+ΔX, Yi+ΔYi} (i={0-8}) of the nine calibration points at the layered camera station. Because the theoretical coordinates of the calibration points are calculated, their order is known and can be adjusted at will. In this embodiment, the coordinate order is ensured to be consistent with that of 151M. In this way, the row and column coordinates {Rowi, Coli} (i={0-8}) of the calibration points obtained by the 151M camera correspond one to one with their physical coordinates {Xi+ΔX, Yi+ΔYi} (i={0-8}). Then, the position relationship matrix R between the two cameras is calculated according to the following formula to complete all the work of automatic point calibration.

[0053]

[0054] Among them, the formula are the parameters in {Xi,Yi} (i={0-8}), are the parameters in {ΔXi,ΔYi} (i={0-8}), are the parameters in {Rowi,Coli} (i={0-8}) respectively.

[0055] In this embodiment, first, a calibration image is input to the calibration screen at the center of the stage, and the camera for lighting detection is used to shoot and obtain the calibration screen shot image. The pre-set calibration image is provided with one central calibration point and several positioning calibration points, and each central calibration point is provided with a cross mark area with a different grayscale from the central calibration point, and the positioning calibration point is provided with a cross mark area with a different grayscale from the positioning calibration point. A positioning pixel point is provided in the exact center of the cross mark area, and the central calibration point is located at the center of the calibration image. The captured calibration screen shot image has a central calibration point and a positioning calibration point at the corresponding position. At this time, different blocks on the calibration screen shot image have at least one calibration point. Then, according to the preset calibration point feature quantity information, the calibration point coordinates are extracted in the calibration screen shot image to generate the calibration point position information, and the calibration point information includes the feature quantity information of the central calibration point and the feature quantity information of the positioning calibration point. At this time, the calibration point position information is the calibration point coordinates determined with reference to the lighting detection camera. Next, the calibration screen is moved to the station of the layered camera by the displacement stage. Among them, the layered camera is a color camera. The area captured by the layered camera is limited, but it can capture clearer images. First, according to the hierarchical information of the calibration screen, the layered camera is focused on the pixel layer of the calibration screen. After alignment, the center calibration point and the positioning calibration point on the calibration screen are collected through the layered camera according to the calibration point position information to generate a calibration point image. At this time, the shooting is performed with the layered camera as the reference system. The calibration point image is converted into a grayscale image so that the difference of the calibration points in the entire shooting area is highlighted. Next, the grayscale threshold for distinguishing the calibration point area, the cross Mark area and the positioning pixel point is determined. The grayscale threshold is used to screen the pixels in the calibration point area of ​​the grayscale image to generate a binary image. At this time, because the layered camera focuses on the pixel layer and highlights the gaps between the pixels, it is necessary to eliminate the gaps between adjacent pixels on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the positioning pixel point position information. Finally, the position deviation information is calculated based on the positioning pixel point position information and the center coordinate position of the layered camera on the calibration point image. After obtaining the position deviation, the position calibration matrix of the lighting detection camera and the layered camera can be calculated through the calibration point position information, the positioning pixel position information and the position deviation information.

[0056] First, the image for calibration is taken by the lighting detection camera at the workstation, and then the layered camera is focused on the pixel layer at the workstation of the layered camera to collect images of each calibration point. The location information of the positioning pixel point is determined by processing image conversion, pixel point screening processing within the calibration point, and adjacent pixel gap elimination processing. Then, the deviation is calculated according to the center coordinate position of the layered camera to further reduce the error. Finally, the position calibration matrix is ​​generated by the calibration point position and deviation determined on the two images. The position calibration matrix can be used as the position relationship data of the current layered camera and the lighting detection camera to improve the calibration effect of the lighting detection camera and the layered camera in the screen foreign body detection.

[0057] See also Figure 2 The present application provides an embodiment of a method for generating calibration point position information, comprising: 201. Binarize the image captured by the calibration screen, and remove the cross mark on the image captured by the calibration screen by morphological dilation; 202. Filter out all calibration point sets based on calibration point feature information of each block on the calibration screen captured image, where the calibration point feature information includes true roundness information and area information; 203. Sort the row and column coordinates of the 9 calibration points in the order of row first and column later to generate calibration point position information.

[0058] In this embodiment, the image needs to be binarized first to remove a large amount of background areas. There is a difference in the grayscale of the cross mark area and the calibration point, so the binary area obtained in the calibration screen image needs to be morphologically expanded to remove the cross mark area, so that the calibration point becomes a complete circle, and then all calibration points are filtered out by feature quantities (such as area, true roundness).

[0059] If the LCD screen does not have a circuit area, you can directly use the area size of the binarized block and the true roundness to analyze which part matches the calibration point. Only after the corresponding area is analyzed can the point calculation be performed.

[0060] Finally, the row and column coordinates of the 9 calibration points are sorted in the order of row first and column later, and finally the row and column coordinates {Rowi, Coli} of the calibration points under the lighting detection camera are obtained, where i={0-8}, and then the coordinates are passed to the layered camera.

[0061] See also Figure 3 The present application provides an embodiment of a method for screening a set of calibration points, comprising: 301. Marking a circuit area on the image captured by the calibration screen according to the circuit distribution data in the circuit structure information; 302. Determine an interference area in the circuit area according to the circuit reflection data and the circuit morphology data in the circuit structure information; 303. Determine a set of pre-screened calibration points that meet the calibration points in each block on the image captured by the calibration screen according to the true roundness information and the area information; 304. The circuit area interference is filtered out from the initial screening calibration point set through the interference area, and a calibration point set that meets the requirements is selected.

[0062] When shooting with a lighting detection camera, if there is an internal circuit of the display screen inside the LCD screen (the internal circuit is usually set very thin, but the presence of a composite circuit will increase the thickness and make it have better reflective ability), and the calibration point is located on the internal circuit area, there may be reflection deviation when the calibration screen displays the calibration image. This is because when the internal circuit structure is complex, part of the internal circuit reflects the light generated when the pixel layer is lit, and the calibration screen image captured by the lighting detection camera may have grayscale deviation, especially around the calibration point. At this time, there may be a false calibration point area caused by the internal circuit reflection, and specific circuit interference processing is required for the circuit area.

[0063] First, the terminal marks the circuit area on the image captured by the calibration screen according to the circuit distribution data in the circuit structure information, then calculates the grayscale value of the calibration point in the calibration image, and then determines the interference area in the circuit area according to the circuit reflection data and circuit morphology data in the circuit structure information. The circuit reflection data refers to the degree to which the material and thickness used in a certain part in the circuit area can reflect the light source. The reflected part can interfere with the display effect of a specific area on the LCD screen. In this embodiment, it is necessary to screen the circuit area that can cause interference, and then analyze it according to the circuit morphology data. If the circuit area in the part of the interfering circuit area is small or very narrow, the reflection effect it brings can be ignored. However, if there is an area with a circular or circular shape, that is, an area similar to the shape of the calibration point, it is determined to be an interference area.

[0064] The terminal will determine the initial screening calibration point set that meets the calibration point in each block on the calibration screen image based on the true roundness information and area information. Then, the circuit area interference is filtered out from the initial screening calibration point set through the interference area, and finally a set of calibration points that meets the requirements is selected.

[0065] It should be noted that, in addition to screening, if the interference area and the initial screening calibration point set coincide with each other, after the true roundness and area detection, the real calibration point area is missing due to the filtering of the interference area and cannot be detected. In this case, it is necessary to adjust the calibration point position of this block and re-test.

[0066] See also Figure 4The present application provides an embodiment of a method for screening pixels, comprising: 401. When there is a calibration point located on the circuit area, select channel data from RGB channel information of the grayscale image according to circuit reflection data in the circuit structure information; 402. Generate a grayscale threshold according to the selected channel pixel data; 403. Perform pixel screening processing on the grayscale image through a grayscale threshold to generate a binary image.

[0067] In this embodiment, the depth of field of the layered camera is small, so that the pixels in the pixel layer can be presented quite clearly, which makes the grayscale between the calibration point area and the cross mark area more susceptible to the influence of the internal circuit. In the calibration screen image taken by the lighting detection camera, it is affected by the circuit area of ​​Class A circuit (strong reflection ability), but when the calibration point is located in the Class B (with reflection ability, medium reflection ability) circuit area and the calibration point position overlap, in the calibration point image taken by the layered camera with a small depth of field, the grayscale of the calibration point area and the cross mark area is more susceptible to the influence of the Class B circuit area.

[0068] When the terminal converts the calibration point image into a grayscale image, the grayscale of the RGB channels in the grayscale image is quite different. At this time, it is necessary to analyze which calibration point overlaps with the Class B circuit area, and select channel data from the RGB channel information of the grayscale image based on the circuit reflection data in the circuit structure information. Specifically, the grayscale values ​​of the two sub-pixels R and B are used as the grayscale threshold, usually as the lower limit of the grayscale for binarization. Finally, the grayscale image is pixel-filtered by the grayscale threshold to generate a binarized image. This method can effectively reduce the impact of the Class B circuit area on the calibration point area and the cross mark area in the grayscale image, and accurately separate the pixels of the two areas.

[0069] See also Figure 5 The present application provides an embodiment of a method for eliminating circuit area interference, comprising: 501. Acquire a plurality of light reflection images collected by the calibration screen under spherical integral light sources of different brightness; 502. Detecting brightness information of a display screen having a calibration image inputted thereto in a non-circuit area by means of a brightness meter; 503. Determine a corresponding target light reflection image among the plurality of light reflection images according to the brightness information; 504. Determine the calibration point to be adjusted located in the circuit area; 505. Determine a reflection area corresponding to the calibration point to be adjusted on the target light reflection image; 506. Perform circuit interference elimination processing on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean value of the reflection area.

[0070] In this embodiment, it is necessary to consider the influence of the reflection of the circuit area on the reflection of the calibration area and adjust the grayscale of the calibration point area and the cross mark area so that the subsequent binarization operation can be more accurate. Specifically, it is necessary to first obtain several light reflection images collected by the calibration screen under spherical integral light sources of different brightness.

[0071] Then, the brightness information of the display screen with the calibration image input in the non-circuit area is detected by a brightness meter, and the target light reflection image with consistent brightness is found through the brightness information.

[0072] Next, determine the calibration points to be adjusted located in the circuit area, that is, the calibration point area and the cross Mark point area on the grayscale image and located on the circuit area. Next, determine the reflection area corresponding to the calibration point to be adjusted on the target light reflection image, and finally perform circuit interference elimination processing on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean of the reflection area, that is, subtract the grayscale mean of the corresponding reflection area from the grayscale of the pixel point on the calibration point area, and at the same time subtract the grayscale mean of the corresponding reflection area from the grayscale of the pixel point in the cross Mark point area. This further eliminates the influence of the reflection of the circuit area on the grayscale image, and improves the accuracy of the subsequent binarization processing.

[0073] See also Figure 6 The present application provides an embodiment of a method for generating location information of a positioning pixel point, comprising: 601. Perform morphological dilation processing on the pixels of the calibration point area in the binary image, and connect the pixels of the calibration point area to form a pixel whole, wherein the calibration point area includes a central calibration point area and a positioning calibration point area; 602. Extract the coordinate position of the positioning pixel point from the binary image after morphological expansion according to the feature information of the positioning pixel point, and generate the position information of the positioning pixel point.

[0074] In this embodiment, the terminal first performs morphological dilation processing on the pixels in the calibration point area in the binary image, connects the pixels around the positioning pixel (mainly the pixels in the calibration area, the pixels in the cross mark area will not be affected due to the difference in grayscale, and the positioning pixel does not exist around adjacent pixels of the same grayscale, so the dilation will not affect the positioning pixel) to form a pixel as a whole, and eliminates the pixel gaps generated after the binarization. This has no effect on the positioning pixel, because there are no other lit pixels around the positioning pixel, so the surrounding area cannot eliminate gaps with other pixels after binarization.

[0075] Finally, the positioning points are extracted from the binary image after morphological expansion according to the feature information of the positioning pixels, that is, according to, and the coordinates of the positioning pixels are determined to generate the position information of the positioning pixels.

[0076] See also Figure 7 The present application provides an embodiment of a calibration device based on screen foreign body detection, including: The first acquisition unit 701 is used to input a calibration image to the calibration screen at the center of the stage, use a light detection camera to shoot, and obtain a calibration screen shot image, wherein one center calibration point and several positioning calibration points are set on the calibration image, a cross Mark area with a grayscale different from that of the center calibration point is set on the center calibration point, a cross Mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point, a positioning pixel point is set at the exact center of the cross Mark area, the center calibration point is located at the center of the calibration image, and there are a center calibration point and a positioning calibration point at the corresponding position of the calibration screen shot image; The first generating unit 702 is used to extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, wherein the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; Optionally, the first generating unit 702 includes: The processing module 7021 is used to perform binarization processing on the calibration screen captured image, and remove the cross mark on the calibration screen captured image by morphological dilation; A screening module 7022 is used to screen out all calibration point sets from the calibration point feature information of each block on the calibration screen captured image, where the calibration point feature information includes true roundness information and area information; Optionally, the calibration screen is a display screen provided with a circuit area, and the calibration point feature information also includes circuit structure information; Screening module 7022, including: Marking the circuit area on the image captured by the calibration screen according to the circuit distribution data in the circuit structure information; Determine the interference area in the circuit area according to the circuit reflection data and the circuit morphology data in the circuit structure information; Determine a set of preliminary screening calibration points that meet the calibration points in each block on the image captured by the calibration screen according to the true roundness information and the area information; The circuit area interference is filtered out from the initial screening calibration point set through the interference area, and the calibration point set that meets the requirements is selected.

[0077] The generating module 7023 is used to sort the row and column coordinates of the 9 calibration points in the order of row first and column later to generate the calibration point position information.

[0078] The displacement unit 703 is used to move the calibration screen to the station of the layered camera through the displacement stage, and the layered camera is a color camera; The second generating unit 704 is used to focus the layered camera to the pixel layer of the calibration screen according to the layer information of the calibration screen, and collect the central calibration point and the positioning calibration point on the calibration screen through the layered camera according to the calibration point position information to generate a calibration point image; A conversion unit 705, used to convert the calibration point image into a grayscale image; A first acquisition unit 706 is used to acquire a plurality of light reflection images collected by the calibration screen under spherical integral light sources of different brightness; A detection unit 707, configured to detect brightness information of a display screen having a calibration image inputted thereto in a non-circuit area by using a brightness meter; A second determining unit 708, configured to determine a corresponding target light reflection image among the plurality of light reflection images according to the brightness information; A third determining unit 709, configured to determine a calibration point to be adjusted located in the circuit area; The fourth determining unit 710 is used to determine the reflection area corresponding to the calibration point to be adjusted on the target light reflection image; The elimination unit 711 is used to perform circuit interference elimination processing on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean value of the reflection area; A first determining unit 712 is used to determine a grayscale threshold, and to perform pixel point screening processing on the calibration point area of ​​the grayscale image by using the grayscale threshold to generate a binary image; Optionally, the first determining unit 712 includes: When there is a calibration point located on the circuit area, channel data is selected from the RGB channel information of the grayscale image according to the circuit reflection data in the circuit structure information; Generate a grayscale threshold according to the selected channel pixel data; The grayscale image is filtered by grayscale threshold to generate a binary image.

[0079] The third generating unit 713 is used to perform adjacent pixel gap elimination processing on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points; Optionally, the third generating unit 713 includes: Performing morphological dilation processing on the pixels of the calibration point area in the binary image, connecting the pixels of the calibration point area to form a pixel whole, the calibration point area includes a central calibration point area and a positioning calibration point area; The coordinate position of the positioning pixel point is extracted from the binary image after morphological expansion according to the feature information of the positioning pixel point, and the position information of the positioning pixel point is generated.

[0080] A first calculation unit 714 is used to calculate the position deviation information according to the position information of the positioning pixel point and the center coordinate position of the layered camera on the calibration point image; The second calculation unit 715 is used to calculate the position calibration matrix of the lighting detection camera and the layered camera through the calibration point position information, the positioning pixel point position information and the position deviation information.

[0081] See also Figure 8 The present application provides a calibration device based on screen foreign body detection, comprising: Processor 801 , memory 802 , input-output unit 803 , and bus 804 .

[0082] The processor 801 is connected to the memory 802 , the input and output unit 803 , and the bus 804 .

[0083] The memory 802 stores a program, and the processor 801 calls the program to execute the following steps: Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 The calibration method in .

[0084] The present application provides a computer-readable storage medium, on which a program is stored, and when the program is executed on a computer, the program performs the following steps: Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 The calibration method in .

[0085] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0086] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0087] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0088] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0089] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A calibration method based on screen foreign body detection, characterized in that: include: Input a calibration image to the calibration screen at the center of the stage, use a lighting detection camera to shoot, and obtain a calibration screen shot image, wherein one center calibration point and several positioning calibration points are set on the calibration image, a cross Mark area with a grayscale different from that of the center calibration point is set on the center calibration point, a cross Mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point, and a positioning pixel point is set at the exact center of the cross Mark area, the center calibration point is located at the center of the calibration image, and there are a center calibration point and a positioning calibration point at the corresponding position of the calibration screen shot image; Extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, wherein the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; The calibration screen is moved to the working position of a layered camera by a displacement stage, and the layered camera is a color camera; Focusing the layered camera to the pixel layer of the calibration screen according to the layer information of the calibration screen, collecting the central calibration point and the positioning calibration point on the calibration screen through the layered camera according to the calibration point position information, and generating a calibration point image; Converting the calibration point image into a grayscale image; Determine a grayscale threshold, and perform pixel point screening processing on the calibration point area of ​​the grayscale image by using the grayscale threshold to generate a binary image; Performing adjacent pixel gap elimination processing on the binary image, and extracting the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points, and generating the position information of the positioning pixel points; Calculate the position deviation information according to the positioning pixel point position information and the center coordinate position of the layered camera on the calibration point image; The position calibration matrix of the lighting detection camera and the layered camera is calculated through the calibration point position information, the positioning pixel point position information and the position deviation information.

2. The calibration method according to claim 1, characterized in that: Extracting the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature quantity information to generate the calibration point position information includes: Binarization is performed on the calibration screen captured image, and the cross mark on the calibration screen captured image is removed by morphological dilation; Filtering out all calibration point sets from calibration point feature information of each block on the captured image of the calibration screen, wherein the calibration point feature information includes true roundness information and area information; The row and column coordinates of the 9 calibration points are sorted in the order of row first and column later to generate the calibration point position information.

3. The calibration method according to claim 2, characterized in that: The calibration screen is a display screen provided with a circuit area, and the calibration point feature information also includes circuit structure information; The calibration point feature information of each block on the calibration screen captured image is filtered to obtain all calibration point sets, including: Marking a circuit area on the calibration screen captured image according to the circuit distribution data in the circuit structure information; Determine the interference area in the circuit area according to the circuit reflection data and the circuit morphology data in the circuit structure information; Determine a set of preliminary screening calibration points that meet the calibration points in each block on the image captured by the calibration screen according to the true roundness information and the area information; The circuit area interference is filtered out from the initial screening calibration point set through the interference area, and a calibration point set that meets the requirements is selected.

4. The calibration method according to claim 3, characterized in that: Determining a grayscale threshold, performing pixel point screening processing on the calibration point area of ​​the grayscale image by using the grayscale threshold, and generating a binary image, including: When there is a calibration point located on the circuit area, selecting channel data from the RGB channel information of the grayscale image according to the circuit reflection data in the circuit structure information; Generate a grayscale threshold according to the selected channel pixel data; The grayscale image is subjected to pixel screening processing by using a grayscale threshold to generate a binary image.

5. The calibration method according to claim 4, characterized in that: After converting the calibration point image into a grayscale image, determining a grayscale threshold, and performing pixel point screening processing on the calibration point area of ​​the grayscale image by the grayscale threshold, before generating a binary image, the calibration method further includes: Acquire a plurality of light reflection images collected by the calibration screen under spherical integral light sources of different brightness; Detecting brightness information of the display screen in the non-circuit area with the calibration image input by a brightness meter; Determine a corresponding target light reflection image among a plurality of light reflection images according to brightness information; Determine the calibration point to be adjusted located in the circuit area; Determine a reflection area corresponding to the calibration point to be adjusted on the target light reflection image; A circuit interference elimination process is performed on the grayscale value of the corresponding area of ​​the grayscale image according to the grayscale mean value of the reflective area.

6. The calibration method according to any one of claims 1 to 5, characterized in that: The binary image is processed to eliminate gaps between adjacent pixels, and the coordinates of the positioning pixel points are extracted from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points, including: Performing morphological dilation processing on the pixels of the calibration point area in the binary image, connecting the pixels of the calibration point area to form a pixel whole, the calibration point area including a central calibration point area and a positioning calibration point area; The coordinate position of the positioning pixel point is extracted from the binary image after morphological expansion according to the feature information of the positioning pixel point, and the position information of the positioning pixel point is generated.

7. A calibration device based on screen foreign body detection, characterized in that: include: A first acquisition unit is used to input a calibration image to a calibration screen at the center of the stage, shoot with a lighting detection camera, and acquire a calibration screen shot image, wherein one center calibration point and several positioning calibration points are set on the calibration image, a cross Mark area with a grayscale different from that of the center calibration point is set on the center calibration point, a cross Mark area with a grayscale different from that of the positioning calibration point is set on the positioning calibration point, a positioning pixel point is set at the exact center of the cross Mark area, the center calibration point is located at the center of the calibration image, and there are a center calibration point and a positioning calibration point at corresponding positions of the calibration screen shot image; A first generating unit is used to extract the coordinates of the calibration points in the image captured by the calibration screen according to the preset calibration point feature information, and generate the calibration point position information, wherein the calibration point information includes the feature information of the center calibration point and the feature information of the positioning calibration point; A displacement unit, used for moving the calibration screen to a station of a layered camera via a displacement stage, wherein the layered camera is a color camera; A second generating unit is used to focus the layered camera to the pixel layer of the calibration screen according to the layer information of the calibration screen, collect the central calibration point and the positioning calibration point on the calibration screen through the layered camera according to the calibration point position information, and generate a calibration point image; A conversion unit, used for converting the calibration point image into a grayscale image; A first determining unit is used to determine a grayscale threshold, and to perform pixel point screening processing on the calibration point area of ​​the grayscale image by using the grayscale threshold to generate a binary image; The third generating unit is used to perform adjacent pixel gap elimination processing on the binary image, and extract the coordinates of the positioning pixel points from the binary image according to the feature information of the positioning pixel points to generate the position information of the positioning pixel points; A first calculation unit, used for calculating position deviation information according to the position information of the positioning pixel point and the center coordinate position of the layered camera on the calibration point image; The second calculation unit is used to calculate the position calibration matrix of the lighting detection camera and the layered camera according to the calibration point position information, the positioning pixel point position information and the position deviation information.

8. The calibration device according to claim 7, characterized in that: The first generation unit comprises: A processing module, used for performing binarization processing on the image captured by the calibration screen, and removing the cross mark on the image captured by the calibration screen by morphological dilation; A screening module, used to screen out all calibration point sets from the calibration point feature information of each block on the image captured by the calibration screen, wherein the calibration point feature information includes true roundness information and area information; The generation module is used to sort the row and column coordinates of the 9 calibration points in the order of row first and column later to generate the calibration point position information.

9. A calibration device based on screen foreign body detection, characterized in that: Includes processor, memory, input and output unit and bus; The processor is connected to the memory, the input and output unit, and the bus; The memory stores a program, and the processor calls the program to execute the calibration device according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, and when the program is executed on a computer, the calibration device according to any one of claims 1 to 6 is executed.

Images