WO2022246643A1 - 图像获取方法及装置、存储介质 - Google Patents
图像获取方法及装置、存储介质 Download PDFInfo
- Publication number
- WO2022246643A1 WO2022246643A1 PCT/CN2021/095789 CN2021095789W WO2022246643A1 WO 2022246643 A1 WO2022246643 A1 WO 2022246643A1 CN 2021095789 W CN2021095789 W CN 2021095789W WO 2022246643 A1 WO2022246643 A1 WO 2022246643A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- defect
- defects
- display panel
- target
- image
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000007547 defect Effects 0.000 claims abstract description 553
- 238000012423 maintenance Methods 0.000 claims abstract description 122
- 238000012216 screening Methods 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims description 53
- 238000000605 extraction Methods 0.000 claims description 11
- 230000007812 deficiency Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 11
- 230000008439 repair process Effects 0.000 description 16
- 230000008569 process Effects 0.000 description 12
- 238000004519 manufacturing process Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 238000013527 convolutional neural network Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 239000012528 membrane Substances 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 210000003454 tympanic membrane Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
-
- G—PHYSICS
- G02—OPTICS
- G02F—OPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
- G02F1/00—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics
- G02F1/01—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour
- G02F1/13—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour based on liquid crystals, e.g. single liquid crystal display cells
Definitions
- the present application relates to the field of display technology, and in particular to an image acquisition method and device, and a storage medium.
- display panels usually need to be manufactured through a complex process. During the manufacturing process of the display panel, defects such as open circuit or short circuit are very easy to occur in its circuit.
- AOI Automatic Optic Inspection
- the AOI system After detecting a defect in the display panel, the AOI system also needs to photograph the display panel to obtain a repair reference image containing the defect. Subsequent maintenance personnel can perform maintenance on the display panel according to the image, so as to greatly improve the product yield of the display panel.
- the current AOI system has low accuracy in obtaining repair reference images, resulting in a low yield rate of the display panel.
- Embodiments of the present application provide an image acquisition method and device, and a storage medium.
- the problem of low yield rate of the current display panel can be solved, and the technical solution is as follows:
- an image acquisition method comprising:
- the target defect of the display panel is photographed to obtain a maintenance reference image including the target defect.
- the maintenance instruction information includes: at least one of location information and defect type information.
- the maintenance instruction information includes: the location information, and based on the maintenance instruction information of the plurality of defects, screening target defects among the plurality of defects includes:
- a defect located in a specified area of the display panel is determined as the target defect.
- the designated area includes: at least one of: an area where transistors in the display panel are located, an intersection area of two types of signal lines, and a connection area between two connected electrodes.
- the maintenance instruction information includes: the defect type information, and based on the maintenance instruction information of the plurality of defects, screening target defects among the plurality of defects includes:
- a defect whose defect type is a specified type is determined as the target defect.
- determining a defect whose defect type is a specified type as the target defect includes:
- the extracted first defect feature is compared with a second defect feature to screen a target defect among the plurality of defects, and the second defect feature is used to characterize the defect type information of the specified type.
- comparing each of the first defect features with the second defect features to screen the target defect among multiple defects including:
- a defect corresponding to the first defect feature whose similarity is greater than a similarity threshold is determined as the target defect.
- the maintenance instruction information includes: the location information and the defect type information, and based on the maintenance instruction information of the plurality of defects, screening target defects among the plurality of defects includes:
- a defect whose defect type is a specified type is determined as the target defect.
- photographing the target defect of the display panel to obtain a maintenance reference image containing the target defect including:
- the image containing the target defect is intercepted in the second image to obtain the maintenance reference image.
- an image acquisition device comprising:
- an acquisition module configured to acquire the first image containing the display panel
- a determining module configured to determine maintenance instruction information of a plurality of defects of the display panel in the first image, where the maintenance instruction information is used to reflect the degree of maintenance demand for defects;
- a screening module configured to screen target defects among the plurality of defects based on the maintenance instruction information of the plurality of defects
- the photographing module is configured to photograph the target defect of the display panel to obtain a maintenance reference image including the target defect.
- the maintenance instruction information includes: location information
- the screening module is configured to: determine, based on the location information of the plurality of defects, a defect located in a designated area of the display panel as the target defect .
- the maintenance instruction information includes: defect type information
- the screening module is configured to determine, based on the defect type information of the plurality of defects, a defect whose defect type is a specified type as the target defect.
- the maintenance instruction information includes: location information and defect type information
- the screening module is configured to: based on the location information of the plurality of defects, among the plurality of defects, the Defects within the specified area of the selected defect are determined as candidate defects; based on the defect type information of the candidate defects, among the plurality of candidate defects, a defect whose defect type is a specified type is determined as the target defect.
- the shooting module is configured to: acquire a second image containing the display panel; based on the position information of the target defect, intercept the image containing the target defect in the second image, to obtain The repair reference image.
- a computer-readable storage medium wherein instructions are stored in the computer-readable storage medium, and when the readable storage medium is run on a processing component, the processing component performs the above-mentioned image acquisition method.
- a maintenance reference image including the target defect may be acquired subsequently.
- the target defects screened out through the maintenance instruction information have a high probability of being repaired, and other defects not screened out have a low probability of being repaired, which effectively improves the accuracy of the AOI system to obtain maintenance reference images and ensures The yield rate of the display panel is relatively high.
- FIG. 1 is a schematic structural diagram of an AOI system involved in an image acquisition method provided by an embodiment of the present application
- Fig. 2 is a flow chart of an image acquisition method provided by an embodiment of the present application.
- Fig. 3 is a flow chart of another image acquisition method provided by the embodiment of the present application.
- FIG. 4 is an effect diagram of a defect of a display panel provided by an embodiment of the present application.
- Fig. 5 is a flow chart of a method for screening target defects among multiple defects provided by the embodiment of the present application.
- Fig. 6 is a structural block diagram of an image acquisition device provided by an embodiment of the present application.
- FIG. 1 is a schematic structural diagram of an AOI system involved in an image acquisition method provided in an embodiment of the present application.
- the AOI system 100 may include: an image acquisition device 101 and a processing device 102 .
- the image capture device 101 is used to capture images of the display panel 103 . In this way, during the manufacturing process of the display panel 103 , the images of the display panel 103 after various processes can be collected by the image capturing device 101 .
- the processing device 102 may be a computer device, or a server, or a server cluster composed of several servers, or a cloud computing service center.
- the image acquisition device 101 is communicatively connected with the processing device 102 .
- the image capture device 101 can send the images it captures to the processing device 102 .
- the so-called communication connection in the embodiment of the present application may be a communication connection established through a wired network or a wireless network.
- the image capture device 101 can capture grayscale images and color images including the display panel 103 .
- the processing device 102 can complete the process of detecting defects in the display panel based on the grayscale image, and can also complete the process of obtaining a repair reference image for defects based on the color image.
- the AOI system after the AOI system detects the defects existing in the display panel, the AOI system also needs to photograph each defect in the display panel to obtain a maintenance reference image including each defect.
- the number of defects of the display panel detected by the AOI system is relatively large (usually hundreds or even thousands of defects can be detected). Since it takes a long time for the AOI system to obtain the repair reference image of the defect, and some defects of the display panel detected by the AOI system do not need to be repaired, and these defects will not affect the normal display of the display panel. Therefore, if the AOI system needs to obtain repair reference images of all the defects of the display panel, it will inevitably result in a low detection efficiency of the display panel by the AOI system, thereby resulting in low production capacity of the display panel.
- the AOI system can inspect 22,000 display panels per month; when each display panel has 100 defects, the AOI system can only inspect 17,000 display panels per month to test.
- the AOI system acquires all the defects in the display panel, it is necessary to randomly select a specified number (for example, 20) of the defects from these defects, and The screened out defects are photographed to obtain a maintenance reference image including the screened out defects.
- FIG. 2 is a flow chart of an image acquisition method provided by an embodiment of the present application.
- the image acquisition method is applied to the processing device 102 in the AOI system 100 shown in FIG. 1 .
- the image acquisition method may include:
- Step 201 Acquire a first image including a display panel.
- Step 202 Determine maintenance instruction information of multiple defects of the display panel in the first image.
- the repair instruction information is used to reflect the degree of repair need for the defect.
- Step 203 based on the maintenance instruction information of multiple defects, screening target defects among multiple defects.
- Step 204 photographing the target defect of the display panel to obtain a maintenance reference image including the target defect.
- the image acquisition method obtains multiple defect maintenance instruction information of the display panel, and based on the maintenance instruction information, screens the target defect among the multiple defects, and subsequently acquires the maintenance information containing the target defect.
- the image There is no need to obtain maintenance reference images of all defects of the display panel, so that the detection efficiency of the display panel by the AOI system is high, and the production capacity of the display panel is high.
- the target defects screened out through the maintenance instruction information have a high probability of being repaired, and other defects not screened out have a low probability of being repaired, which effectively improves the accuracy of the AOI system to obtain maintenance reference images and ensures The yield rate of the display panel is relatively high.
- FIG. 3 is a flowchart of another image acquisition method provided by an embodiment of the present application.
- the image acquisition method is applied to the processing device 102 in the AOI system 100 shown in FIG. 1 .
- the image acquisition method may include:
- Step 301 Acquire a first image including a display panel.
- the processing device may acquire the first image including the display panel.
- the image acquisition device in the AOI system can acquire the first image of the display panel after each process. Moreover, the AOI system can send the collected first image including the display panel to the processing device, so that the processing device can acquire the image including the display panel.
- the first image is usually a grayscale image.
- the image acquisition device may include: a scanning camera, and the first image acquired by the scanning camera is a grayscale image.
- the display panel can be loaded on the carrier base in the AOI system, and the display panel on the carrier base can be aligned by the mechanical arm in the AOI system, so as to Make sure that the image acquisition device can face the display panel directly.
- the first image including the display panel can be acquired through the scanning camera in the image acquisition device.
- the image acquisition device may send the acquired image to the storage server, so that the storage server may store the first image including the display panel. Subsequently, if it is necessary to detect defects in the display panel after a certain process, the processing device downloads the corresponding first image from the storage server, and detects the defects in the display panel in the first image .
- Step 302 acquiring multiple defects of the display panel in the first image.
- the processing device may further acquire multiple defects of the display panel in the first image.
- the processing device may acquire multiple defects of the display panel in the first image by means of periodic comparison. For example, you can compare the average gray value of the first row of 20 pixels in the region to be detected in the first image with the average gray value of the second row of 20 pixels to see if the difference is within within the preset range. If the difference is within the preset range, it means that the gray value of the 20 pixels in the first row or the gray value of the 20 pixels in the second row is relatively balanced, and there is no pixel with a sudden change in the gray value. There is no defect in the detection area; otherwise, it is determined that there is a defect in the area to be detected.
- Step 303 Determine maintenance instruction information of the plurality of defects.
- the processing device may further acquire maintenance instruction information for the multiple defects.
- the maintenance instruction information is used to reflect the degree of maintenance demand for the defect. For defects that will affect the display effect of the display panel, the degree of maintenance demand for the defect is relatively high, and the probability that it needs to be repaired is high; for defects that do not affect the display effect of the display panel, the maintenance demand for the defect is high. The lower the degree, the less likely it needs to be repaired. It should be noted that the processing device needs to acquire maintenance instruction information for each of the multiple defects.
- the indication information of the defect of the display panel may include: at least one of defect location information and defect type information.
- Step 304 based on the maintenance instruction information of multiple defects, screen the target defect among multiple defects.
- the processing device may filter the target defect among the multiple defects based on the maintenance instruction information of the multiple defects.
- the target defect is a defect that may affect the display effect of the display panel, and the degree of maintenance requirement for the target defect is relatively high.
- the maintenance instruction information of the defect includes: location information of the defect.
- the position information of the defect is used to reflect the position of the defect in the display panel.
- the above step 304 may include: based on the location information of multiple defects, determining a defect located in a specified area of the display panel as a target defect.
- the designated area of the display panel is an artificially predetermined area.
- the designated area may include: at least one of the area where transistors in the display panel are located, the intersection area of two signal lines, and the connection area of two connected electrodes. The defects existing in these designated areas have a high probability of affecting the normal display of the display panel.
- the processing device may determine a defect located in a specified area of the display panel as a target defect based on the location information of the plurality of defects.
- the target defect located in the designated area has a higher probability of needing to be repaired, and a repair reference image including the target defect can be acquired subsequently.
- other defects located outside the designated area have a low probability of affecting the normal display of the display panel, and have a low probability of being repaired, and there is no need to obtain repair reference images of these defects in the future.
- FIG. 4 is an effect diagram of a defect of a display panel provided by an embodiment of the present application.
- the processing device obtains and determines the coordinate points of each pixel point in the first image in a certain coordinate system.
- the origin in the coordinate system may coincide with the center point of the first image
- the X axis in the coordinate system is parallel to the width direction of the first image
- the Y axis in the coordinate system is parallel to the length direction of the first image. Since the position of the designated area P in the first image is fixed, the position of the designated area P in the coordinate system is also fixed.
- the processing device determines that defect a is located outside the designated area P, and defect b is located within the designated area P, then the processing device may determine defect b as the target defect.
- the maintenance instruction information of the defect includes: defect type information of the defect.
- the defect type information of the defect is used to reflect the type of the defect.
- the defect type may include: foreign body residue, tympanic membrane, membrane rupture, membrane layer detachment, membrane layer loss, disconnection or short circuit, etc.
- the above step 304 may include: based on the defect type information of multiple defects, determining a defect whose defect type is a specified type as a target defect.
- the specified type of defects may be artificially predetermined defects that have a high degree of influence on the display effect of the display panel.
- the specified type of defect may include: at least one of film peeling, film missing, disconnection and short circuit.
- the processing device may determine a defect whose defect type is a specified type as a target defect based on the defect type information of the plurality of defects.
- a target defect whose defect type is a specified type has a high probability of needing to be repaired, and a repair reference image including the target defect can be acquired subsequently.
- other defects whose defect type is not the specified type have a low probability of affecting the normal display of the display panel, and have a low probability of being repaired, and there is no need to obtain repair reference images of these defects in the future.
- the processing device determines a defect whose defect type is a specified type as a target defect based on the defect type information of multiple defects, which may include the following steps:
- Step A performing feature extraction on multiple defects in the first image to obtain first defect features of the multiple defects.
- the processing device may perform feature extraction on each defect in the first image to obtain first defect features of multiple defects.
- the first defect feature is used to characterize the defect type information of the defect, which may include grayscale information and boundary information of the displayed content in the image.
- the defect feature can be an array or a vector.
- the way for the processing device to obtain the boundary information may be: use a Gaussian filter to smooth the image; use the finite difference of the first-order partial derivative to calculate the gradient Magnitude and direction; non-maximum suppression of gradient magnitude; detection and connection of edges with a dual-threshold algorithm.
- the feature extraction process can be implemented by a convolutional neural network (English: convolutional neural network; referred to as: CNN), for example, the first image can be directly input to the CNN, and the CNN calculates and outputs the image. A first defect feature for each defect in the first image.
- the feature extraction process can also be implemented by other computing modules or feature extractors. For example, a convolution operation can be performed on the first image, and the result of the operation can be used as the result of the first image. defect characteristics. It should be noted that there may be other manners of feature extraction processing, which are not limited in this embodiment of the present application.
- Step B Comparing the extracted first defect feature with the second defect feature to screen the target defect among multiple defects.
- the processing device may compare the extracted first defect features with the second defect features respectively, so as to screen the target defect among multiple defects.
- the second defect feature is used to characterize the type information of a specified type of defect.
- the second defect feature is acquired in advance.
- multiple reference images containing specified types of defects can be screened out in advance from a large number of images collected by the image acquisition device in the AOI system, and feature extraction can be performed on the specified types of defects in the multiple images, and the extracted After the average operation is performed on the obtained defect features, the result of the operation can be used as the second defect feature.
- the processing equipment extracts the first defect feature of each defect, it directly compares the first defect feature with the second defect feature, so that the target defect can be screened out from multiple defects.
- the obtained defect features are also different.
- the multiple second defect features correspond to various types of defects one by one.
- the number of first defect features obtained after feature extraction of multiple defects in the first image is also multiple, and after subsequent comparisons of each first defect feature with each second defect feature, that is Defects whose defect type is a specified type can be filtered out among multiple defects.
- step B may include:
- Step B Determine the degree of similarity between each first defect feature and the second defect feature.
- the processing device may determine the similarity between each first defect feature and the second defect feature.
- the processing device may use a similarity calculation formula to calculate the similarity between each first defect feature and the second defect feature.
- Step B2 detecting whether the similarity between each first defect feature and the second defect feature is greater than a similarity threshold.
- the processing device can detect whether the similarity between each first defect feature and the second defect feature is greater than the similarity threshold.
- step B3 when the processing device determines that the similarity between a certain first defect feature and the second defect feature is greater than the similarity threshold, step B3 is performed; when the processing device determines that the similarity between a certain first defect feature and the second defect feature When the degree is not greater than the similarity threshold, perform step B4.
- Step B3 determining the defect corresponding to the first defect feature whose similarity is greater than the similarity threshold as the target defect.
- the processing device when the processing device determines that the similarity between a first defect feature and a second defect feature is greater than the similarity threshold, the processing device may determine the defect corresponding to the first defect feature as the target defect.
- Step B4 determining the defects corresponding to the first defect features whose similarity is not greater than the similarity threshold as other defects.
- the processing device when the processing device determines that the similarity between a certain first defect feature and the second defect feature is not greater than the similarity threshold, the processing device may determine the defect corresponding to the first defect feature as multiple Defects other than target defects.
- the defect maintenance instruction information includes: defect location information and defect type information.
- location information of the defect refer to the corresponding content in the above-mentioned first implementable manner
- defect type information of the defect refer to the corresponding content in the above-mentioned second implementable manner.
- FIG. 5 is a flowchart of a method for screening target defects among multiple defects provided by an embodiment of the present application.
- the above step 304 may include:
- Step 3041 based on the location information of the plurality of defects, among the plurality of defects, determine a defect located in a specified area of the display panel as a defect to be selected.
- Step 3042 based on the defect type information of the candidate defect, among the plurality of candidate defects, determine a defect whose defect type is a specified type as the target defect.
- the probability of the defect affecting the normal display of the display panel is relatively small.
- the type of the defect is not a specified type of defect, even if the defect is located in the specified area, the probability that the defect will affect the normal display of the display panel is relatively small.
- the defect maintenance instruction information includes: defect location information and defect type information
- the defect located in the designated area and whose defect type is the designated type can be used as the target defect, which not only ensures the screening
- the number of detected target defects can be further reduced, so as to further improve the detection efficiency of the display panel by the AOI system. It can also be ensured that the selected target defects have a higher probability of being repaired, and other defects not screened out have a lower probability of being repaired, so as to ensure a higher yield rate of the display panel.
- the above embodiment is schematically illustrated by taking a preliminary screening based on the defect location information first, and then performing a secondary screening based on the defect type information as an example.
- a preliminary screening may be performed based on the defect type information of the defect, and then a secondary screening may be performed based on the defect location information.
- This embodiment of the present application does not limit it.
- Step 305 Acquire a second image including the display panel.
- the processing device may acquire the second image including the display panel through the image acquisition device.
- the repair reference image is usually a color image.
- the maintenance reference image is obtained by intercepting the second image, and therefore, the second image is usually a color image.
- the image acquisition device may further include: an optical camera for acquiring color images.
- the processing device can acquire the second image through the optical camera. It should be noted that the first image and the second image including the display panel may be acquired simultaneously, that is, the above step 305 and step 301 may be performed simultaneously.
- Step 306 based on the location information of the target defect, intercept the image containing the target defect from the second image to obtain a maintenance reference image.
- an image containing the target defect may be intercepted from the second image based on the position information of the target defect.
- the image containing the target defect is the maintenance reference image of the target defect. Subsequently, maintenance personnel can perform maintenance on the target defect based on the maintenance reference image, so as to improve the yield rate of the display panel.
- the image acquisition method obtains multiple defect maintenance instruction information of the display panel, and based on the maintenance instruction information, screens the target defect among the multiple defects, and subsequently acquires the maintenance information containing the target defect.
- the image There is no need to obtain repair reference images for all defects of the display panel, so as to ensure that the AOI system can detect the display panel with high efficiency, so as to ensure a high production capacity of the display panel.
- the target defects screened out through the maintenance instruction information have a high probability of being repaired, and other defects not screened out have a low probability of being repaired, which effectively improves the accuracy of the AOI system to obtain maintenance reference images and ensures The yield rate of the display panel is relatively high.
- FIG. 6 is a structural block diagram of an image acquisition device provided in an embodiment of the present application.
- the image acquisition apparatus 400 may be integrated into the processing device 102 in the AOI system 100 shown in FIG. 1 .
- the image acquisition device 400 may include:
- An acquiring module 401 configured to acquire a first image including a display panel.
- the determining module 402 is configured to determine maintenance instruction information of multiple defects of the display panel in the first image.
- the maintenance instruction information is used to reflect the degree of maintenance demand for the defect.
- the screening module 403 is configured to screen target defects among multiple defects based on the maintenance instruction information of multiple defects.
- the photographing module 404 is configured to photograph the target defect of the display panel to obtain a maintenance reference image including the target defect.
- the image acquisition device acquires multiple defect maintenance instruction information of the display panel, and based on the maintenance instruction information, screens the target defect among the multiple defects, and subsequently acquires the maintenance information including the target defect.
- the image There is no need to obtain maintenance reference images of all defects of the display panel, so that the detection efficiency of the display panel by the AOI system is high, and the production capacity of the display panel is high.
- the target defects screened out through the maintenance instruction information have a high probability of being repaired, and other defects that have not been screened out have a low probability of being repaired, which effectively improves the accuracy of the AOI system to obtain maintenance reference images and ensures The yield rate of the display panel is relatively high.
- the maintenance instruction information includes: location information.
- the screening module 403 is configured to: determine a defect located in a specified area of the display panel as a target defect based on the location information of multiple defects.
- the designated area includes: at least one of: an area where transistors in the display panel are located, an intersection area of two types of signal lines, and a connection area between two connected electrodes.
- the maintenance instruction information includes: defect type information.
- the screening module 403 is configured to: based on defect type information of multiple defects, determine a defect whose defect type is a specified type as a target defect.
- the screening module 403 is configured to: perform feature extraction on multiple defects in the first image to obtain first defect features of multiple defects, and the first defect features are used to represent defect type information of defects; extract The obtained first defect feature is compared with the second defect feature to screen the target defect among multiple defects, and the second defect feature is used to represent defect type information of a specified type.
- the screening module 403 is configured to: determine the similarity between each first defect feature and the second defect feature; determine the defect corresponding to the first defect feature whose similarity is greater than a similarity threshold as a target defect.
- the maintenance instruction information includes: location information and defect type information.
- the screening module 403 is configured to: determine, among the plurality of defects, a defect located in a specified area of the display panel as a defect to be selected based on the position information of the plurality of defects; Among the defects to be selected, a defect whose defect type is a specified type is determined as a target defect.
- the photographing module 404 is configured to: acquire a second image containing the display panel; based on the location information of the target defect, intercept the image containing the target defect in the second image to obtain a maintenance reference map.
- the image acquisition device acquires multiple defect maintenance instruction information of the display panel, and based on the maintenance instruction information, screens the target defect among the multiple defects, and subsequently acquires the maintenance information including the target defect.
- the image There is no need to obtain maintenance reference images of all defects of the display panel, so that the detection efficiency of the display panel by the AOI system is high, and the production capacity of the display panel is high.
- the target defects screened out through the maintenance instruction information have a high probability of being repaired, and other defects not screened out have a low probability of being repaired, which effectively improves the accuracy of the AOI system to obtain maintenance reference images and ensures The yield rate of the display panel is relatively high.
- the embodiment of the present application also provides an image acquisition device.
- the image acquisition device may include: a processor, and a memory for storing executable instructions of the processor.
- the processor is configured to execute the image acquisition method shown in FIG. 2 or FIG. 3 .
- the embodiment of the present application also provides a computer-readable storage medium. Instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on the processing component, the processing component is made to execute the image acquisition method shown in FIG. 2 or FIG. 3 .
- first and second are used for descriptive purposes only, and should not be construed as indicating or implying relative importance.
- the term “plurality” means two or more, unless otherwise clearly defined.
- the program can be stored in a computer-readable storage medium.
- the above-mentioned The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, and the like.
Abstract
一种图像获取方法及装置、存储介质，方法包括：获取包含显示面板的第一图像；确定第一图像中的显示面板的多个缺陷的维修指示信息，维修指示信息用于反映对缺陷的维修需求的程度；基于多个缺陷的维修指示信息，在多个缺陷中筛选目标缺陷；对显示面板的目标缺陷进行拍摄，以得到包含目标缺陷的维修参考图像。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率较高。并且，通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，有效地提高了AOI系统获取维修参考图像的准确度，保证显示面板的良品率较高。
Description
本申请涉及显示技术领域，特别涉及一种图像获取方法及装置、存储介质。
目前，显示面板通常需要经过复杂的工艺流程制造。在显示面板的制造过程中，其线路极易出现断路或短路等缺陷。
为此，在显示面板的制造过程中的各个工序后，需要采用自动光学检测(英文：Automatic Optic Inspection；简称：AOI)系统对显示面板中存在的缺陷进行检测。在检测到显示面板存在缺陷后，AOI系统还需要对该显示面板进行拍摄，以获取包含缺陷的维修参考图像。后续维修人员能够根据该图像对显示面板进行维修，以大幅提高显示面板的产品良率。
然而，目前的AOI系统获取维修参考图像的准确度较低，导致显示面板的良率较低。
发明内容
本申请实施例提供了一种图像获取方法及装置、存储介质。可以解决目前的显示面板的良率较低的问题，所述技术方案如下：
一方面，提供了一种图像获取方法，所述方法包括：
获取包含显示面板的第一图像；
确定所述第一图像中的所述显示面板的多个缺陷的维修指示信息，所述维修指示信息用于反映对缺陷的维修需求的程度；
基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷；
对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像。
可选的，所述维修指示信息包括：位置信息和缺陷类型信息中的至少一种。
可选的，所述维修指示信息包括：所述位置信息，基于所述多个缺陷的维 修指示信息，在所述多个缺陷中筛选目标缺陷，包括：
基于所述多个缺陷的位置信息，将位于所述显示面板的指定区域内的缺陷确定为所述目标缺陷。
可选的，所述指定区域包括：所述显示面板中的晶体管所在区域、两种信号线的交叉区域和相连的两个电极的连接区域中的至少一种。
可选的，所述维修指示信息包括：所述缺陷类型信息，基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷，包括：
基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
可选的，基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷，包括：
对所述第一图像中的所述多个缺陷进行特征提取，得到所述多个缺陷的第一缺陷特征，所述第一缺陷特征用于表征所述缺陷的缺陷类型信息；
将提取得到所述第一缺陷特征与第二缺陷特征进行比对，以在所述多个缺陷中筛选目标缺陷，所述第二缺陷特征用于表征所述指定类型的缺陷类型信息。
可选的，将各个所述第一缺陷特征与第二缺陷特征进行比对，以在多个所述缺陷中筛选目标缺陷，包括：
确定各个所述第一缺陷特征与所述第二缺陷特征的相似度；
将相似度大于相似度阈值的第一缺陷特征所对应的缺陷确定为所述目标缺陷。
可选的，所述维修指示信息包括：所述位置信息和所述缺陷类型信息，基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷，包括：
基于所述多个缺陷的位置信息，在所述多个缺陷中，将位于所述显示面板的指定区域内的缺陷确定为待选缺陷；
基于所述待选缺陷的缺陷类型信息，在多个所述待选缺陷中，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
可选的，对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像，包括：
获取包含所述显示面板的第二图像；
基于所述目标缺陷的位置信息，在所述第二图像中截取包含所述目标缺陷 的图像，以得到所述维修参考图像。
另一方面，提供了一种图像获取装置，所述装置包括：
获取模块，用于获取包含显示面板的第一图像；
确定模块，用于确定所述第一图像中的所述显示面板的多个缺陷的维修指示信息，所述维修指示信息用于反映对缺陷的维修需求的程度；
筛选模块，用于基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷；
拍摄模块，用于对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像。
可选的，所述维修指示信息包括：位置信息，所述筛选模块，用于：基于所述多个缺陷的位置信息，将位于所述显示面板的指定区域内的缺陷确定为所述目标缺陷。
可选的，所述维修指示信息包括：缺陷类型信息，所述筛选模块，用于：基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
可选的，所述维修指示信息包括：位置信息和缺陷类型信息，所述筛选模块，用于：基于所述多个缺陷的位置信息，在所述多个缺陷中，将位于所述显示面板的指定区域内的缺陷确定为待选缺陷；基于所述待选缺陷的缺陷类型信息，在多个所述待选缺陷中，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
可选的，所述拍摄模块，用于：获取包含所述显示面板的第二图像；基于所述目标缺陷的位置信息，在所述第二图像中截取包含所述目标缺陷的图像，以得到所述维修参考图像。
又一方面，提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述可读存储介质在处理组件上运行时，使得所述处理组件执行上述的图像获取方法。
本申请实施例提供的技术方案带来的有益效果至少包括：
通过获取显示面板的多个缺陷维修指示信息，并基于该维修指示信息在多个缺陷中筛选目标缺陷，后续获取包含该目标缺陷的维修参考图像即可。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率 较高，以保证显示面板的产能较高。并且，在通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，有效的提高了AOI系统获取维修参考图像的准确度，保证该显示面板的良品率较高。
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的一种图像获取方法所涉及的AOI系统的结构示意图；
图2是本申请实施例提供的一种图像获取方法的流程图；
图3是本申请实施例提供的另一种图像获取方法的流程图；
图4是本申请实施例提供的一种显示面板的缺陷的效果图；
图5是本申请实施例提供的一种在多个缺陷中筛选目标缺陷的方法流程图；
图6是本申请实施例提供的一种图像获取装置的结构框图。
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。
请参考图1，图1是本申请实施例提供的一种图像获取方法所涉及的AOI系统的结构示意图。该AOI系统100可以包括：图像采集设备101和处理设备102。
图像采集设备101用于采集显示面板103的图像。如此，在显示面板103的制造过程中，通过该图像采集设备101能够采集到该显示面板103在各个工序后的图像。
处理设备102可以为计算机设备，或者是一台服务器，或者由若干台服务器组成的服务器集群，或者是一个云计算服务中心等。
其中，图像采集设备101与处理设备102通信连接。该图像采集设备101 能够将其采集的图像发送给处理设备102。需要说明的是，本申请实施例中所谓的通信连接，可以是通过有线网络或者无线网络建立的通信连接。
示例的，图像采集设备101能够采集包含显示面板103的灰度图像和彩色图像。处理设备102能够基于该灰度图像完成对显示面板中的缺陷的检测过程，还能够基于该彩色图像完成对缺陷的维修参考图像获取的过程。
在相关技术中，在AOI系统对显示面板中存在的缺陷进行检测后，AOI系统还需要对显示面板中的每个缺陷进行拍摄，以得到包含各个缺陷的维修参考图像。但是，AOI系统所检测出的显示面板的缺陷的个数较多(通常能够检测出成百上千个缺陷)。由于AOI系统获取缺陷的维修参考图像的耗时较长，并且，通过AOI系统检测出的显示面板的部分缺陷是无需进行维修的，这部分缺陷并不会对显示面板的正常显示造成影响。因此，如果AOI系统需要获取显示面板的全部缺陷的维修参考图像时，势必会导致AOI系统对显示面板的检测效率较低，进而导致显示面板的产能较低。
例如，当每个显示面板存在50个缺陷时，AOI系统每个月能够对22000块显示面板进行检测；当每个显示面板存在100个缺陷时，AOI系统每个月仅能够对17000块显示面板进行检测。
目前，为了提高AOI系统对显示面板的检测效率，在AOI系统获取到显示按面板中存在的全部缺陷后，需要随机在这些缺陷中筛选出指定个数(例如，20个)的缺陷，并对筛选出的缺陷进行拍摄，以得到包含筛选出的缺陷的维修参考图像。
然而，当采用随机筛选方式筛选缺陷时，有些会影响显示面板的显示效果的缺陷可能不会被筛选出，进而后续无法对该缺陷进行维修。而有些不会影响显示面板的显示效果的缺陷可能会被筛选出来。因此，该AOI系统获取维修参考图像的准确度较低，导致显示面板的良率较低。
请参考图2，图2是本申请实施例提供的一种图像获取方法的流程图。该图像获取方法应用于图1示出的AOI系统100中的处理设备102。该图像获取方法可以包括：
步骤201、获取包含显示面板的第一图像。
步骤202、确定该第一图像中的显示面板的多个缺陷的维修指示信息。该维 修指示信息用于反映对缺陷的维修需求的程度。
步骤203、基于多个缺陷的维修指示信息，在多个缺陷中筛选目标缺陷。
步骤204、对显示面板的目标缺陷进行拍摄，以得到包含该目标缺陷的维修参考图像。
综上所述，本申请实施例提供的图像获取方法，通过获取显示面板的多个缺陷维修指示信息，并基于该维修指示信息在多个缺陷中筛选目标缺陷，后续获取包含该目标缺陷的维修参考图像即可。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率较高，以保证显示面板的产能较高。并且，在通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，有效的提高了AOI系统获取维修参考图像的准确度，保证该显示面板的良品率较高。
请参考图3，图3是本申请实施例提供的另一种图像获取方法的流程图。该图像获取方法应用于图1示出的AOI系统100中的处理设备102。该图像获取方法可以包括：
步骤301、获取包含显示面板的第一图像。
在本申请实施例中，处理设备可以获取包含显示面板的第一图像。
在本申请中，在显示面板的制造过程中，AOI系统中的图像采集设备可以采集显示面板在各个工序后的第一图像。并且，该AOI系统能够将其采集的包含显示面板的第一图像发送给处理设备，使得该处理设备能够获取到包含显示面板的图像。需要说明的是，为了便于后续对该第一图像中的显示面板存在的缺陷进行检测，该第一图像通常为灰度图像。为此，该图像采集设备可以包括：扫描相机，通过该扫描相机所获取的第一图像为灰度图像。
示例的，在显示面板制造过程中的各个工序后，可以将显示面板加载在AOI系统中的承载基台上，并通过AOI系统中的机械臂对承载基台上的显示面板进行对位，以保证图像采集设备能够正对于显示面板。之后，通过图像采集设备中的扫描相机即可获取到包含显示面板的第一图像。
需要说明的是，在图像采集设备采集到包含显示面板的第一图像后，该图像采集设备可以将其采集到的图像发送给存储服务器，使得存储服务器可以存储该包含显示面板的第一图像。后续，若需要对某个工序后的显示面板中存在 的缺陷进行检测，则处理设备在存储服务器中下载相应的第一图像，并对第一图像中的显示面板中存在的缺陷进行检测即可。
步骤302、获取第一图像中的显示面板的多个缺陷。
在本申请实施例中，在处理设备获取到包含显示面板的第一图像后，该处理设备还可以获取第一图像中的显示面板的多个缺陷。
示例的，处理设备可以通过周期性对比的方式，获取第一图像中的显示面板的多个缺陷。例如，可以将第一图像中的待检测区域内的第一行20个像素的灰度值的平均值与第二行20个像素的灰度值的平均值做比较，看其差值是否在预设范围内。若差值在预设范围内，说明第一行20个像素的灰度值或者第二行20个像素的灰度值比较均衡，并未有灰度值发生突变的像素点，可以确定该待检测区域未出现缺陷；否则，确定该待检测区域存在缺陷。
步骤303、确定该多个缺陷的维修指示信息。
在本申请实施例中，在处理设备获取到第一图像中的显示面板的多个缺陷后，处理设备还可以获取该多个缺陷的维修指示信息。其中，维修指示信息用于反映对缺陷的维修需求的程度。对于会影响显示面板的显示效果的缺陷，对该缺陷的维修需求的程度较高，其需要被维修的概率较高；对于不会影响显示面板的显示效果的缺陷，对该缺陷的维修需求的程度较低，其需要被维修的概率较低。需要说明的是，处理设备需要获取多个缺陷中的每个缺陷的维修指示信息。
在本申请中，显示面板的缺陷的指示信息可以包括：缺陷的位置信息和缺陷类型信息中的至少一种。
步骤304、基于多个缺陷的维修指示信息，在多个缺陷中筛选目标缺陷。
在本申请实施例中，在处理设备获取到多个缺陷的维修指示信息后，处理设备可以基于该多个缺陷的维修指示信息，在多个缺陷中筛选目标缺陷。其中，该目标缺陷为会影响显示面板的显示效果的缺陷，对该目标缺陷的维修需求的程度较高。
在本申请中，缺陷的维修指示信息的类型有多种，而对于不同的类型的维修指示信息，在多个缺陷中筛选目标缺陷的方式也不同。为此，本申请将列举以下三种可实现方式对筛选目标缺陷的方式进行示意性的说明：
第一种可实现方式，缺陷的维修指示信息包括：缺陷的位置信息。其中， 该缺陷的位置信息用于反映缺陷在显示面板中的位置。在这种情况下，上述步骤304可以包括：基于多个缺陷的位置信息，将位于显示面板的指定区域内的缺陷确定为目标缺陷。
在本申请实施例中，显示面板的指定区域为人为预先规定的区域。例如，该指定区域可以包括：显示面板中的晶体管所在区域、两种信号线的交叉区域和相连的两个电极的连接区域中的至少一种。这些指定区域内存在的缺陷对显示面板的正常显示造成影响的概率较大。
为此，处理设备可以基于多个缺陷的位置信息，将位于显示面板的指定区域内的缺陷确定为目标缺陷。其中，该多个缺陷中位于指定区域内的目标缺陷需要被维修的概率较高，后续可以获取包含该目标缺陷的维修参考图像。而多个缺陷中位于指定区域外的其他缺陷对显示面板的正常显示造成影响的概率较低，其需要被维修的概率也较低，后续无需获取这些缺陷的维修参考图像。
示例的，如图4所示，图4是本申请实施例提供的一种显示面板的缺陷的效果图。在处理设备获取到第一图像后，该处理设备获取确定出该第一图像中的各个像素点在某个坐标系下的坐标点。例如，该坐标系中的原点可以与第一图像的中心点重合，该坐标系中的X轴与第一图像的宽度方向平行，该坐标系中的Y轴与第一图像的长度方向平行。由于指定区域P在第一图像中的位置是固定的，因此，指定区域P在该坐标系中的位置也是固定的。当第一图像中的显示面板的缺陷的位置信息采用在该坐标系的坐标点进行表示时，可以基于缺陷在该坐标系中的坐标点与指定区域P内的各个坐标点的关系，确定出缺陷是否在指定区域P内。例如，在图4中，处理设备确定出缺陷a位于指定区域P外，缺陷b位于指定区域P内，则，处理设备可以将缺陷b确定为目标缺陷。
第二种可实现方式，缺陷的维修指示信息包括：缺陷的缺陷类型信息。其中，该缺陷的缺陷类型信息用于反映缺陷的类型。例如，该缺陷类型可以包括：异物残留、鼓膜、膜破、膜层脱落、膜层缺失、断线或短路等。在这种情况下，上述步骤304可以包括：基于多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为目标缺陷。
在本申请实施例中，不同类型的缺陷对显示面板的显示效果的影响程度也不同。该指定类型的缺陷可以是人为预先规定的对显示面板的显示效果的影响程度较高的缺陷。例如，该指定类型的缺陷可以包括：膜层脱落、膜层缺失、 断线和短路中的至少一种。
为此，处理设备可以基于多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为目标缺陷。其中，该多个缺陷中缺陷类型为指定类型的目标缺陷需要被维修的概率较高，后续可以获取包含该目标缺陷的维修参考图像。而多个缺陷中缺陷类型不为指定类型的其他缺陷对显示面板的正常显示造成影响的概率较低，其需要被维修的概率也较低，后续无需获取这些缺陷的维修参考图像。
示例的，处理设备基于多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为目标缺陷，可以包括以下几个步骤：
步骤A、对第一图像中的多个缺陷进行特征提取，得到多个缺陷的第一缺陷特征。
在本申请实施例中，处理设备可以对第一图像中的各个缺陷进行特征提取，得到多个缺陷的第一缺陷特征。
其中，该第一缺陷特征用于表征缺陷的缺陷类型信息，其可以包含图像中的显示内容的灰度信息和边界信息等。其中，该缺陷特征可以为数组或向量。
示例的，当缺陷特征包括图像中的显示内容的边界信息时，处理设备获取边界信息的方式可以为：采用高斯滤波器对图象进行平滑处理；用一阶偏导的有限差分来计算梯度的幅值和方向；对梯度幅值进行非极大值抑制；用双阈值算法检测和连接边缘。
在一种可选的实现方式中，特征提取处理可以由卷积神经网络(英文：convolutional neural network；简称：CNN)来实现，例如，可以直接向CNN中输入第一图像，由CNN计算输出该第一图像中的各个缺陷的第一缺陷特征。在另一种可选的实现方式中，特征提取处理也可以由其他计算模块或者特征提取器来实现，例如，可以对第一图像进行卷积运算，将运算得到的结果作为该第一图像的缺陷特征。需要说明的是，特征提取处理的方式还可以有其他方式，本申请实施例对此不做限定。
步骤B、将提取得到的第一缺陷特征与第二缺陷特征进行比对，以在多个缺陷中筛选目标缺陷。
在本申请实施例中，处理设备可以将提取得到的各个第一缺陷特征分别与第二缺陷特征进行比对，以在多个缺陷中筛选目标缺陷。其中，该第二缺陷特 征用于表征指定类型的缺陷的类型信息。
需要说明的是，第二缺陷特征是预先获取到的。示例的，可以预先在AOI系统中的图像采集设备采集到大量的图像中筛选出多张包含指定类型的缺陷的参考图像，并对多张图像中的指定类型的缺陷进行特征提取，且对提取到的缺陷特征进行平均运算后，可以将运算得到的结果作为第二缺陷特征。如此，在处理设备提取到各个缺陷的第一缺陷特征后，直接将该第一缺陷特征与第二缺陷特征进行比对，便能够在多个缺陷中筛选出目标缺陷。
在本申请中，人为规定的指定类型的缺陷的种类有多种，对不同种类的缺陷进行特征提取后，得到的缺陷特征也不同。为此，第二缺陷特征的个数为多个，该多个第二缺陷特征与多种类型的缺陷一一对应。且，对第一图像中的多个缺陷进行特征提取后得到的第一缺陷特征的个数也为多个，后续依次对每个第一缺陷特征与各个第二缺陷特征进行比对后，即可在多个缺陷中筛选出缺陷类型为指定类型的缺陷。
示例的，上述步骤B可以包括：
步骤B1、确定各个第一缺陷特征与第二缺陷特征的相似度。
在本申请实施例中，处理设备在确定出多个缺陷的第一缺陷特征后，该处理设备可以确定各个第一缺陷特征与第二缺陷特征的相似度。
示例的，处理设备可以采用相似度计算公式计算各个第一缺陷特征与第二缺陷特征的相似度。
步骤B2、检测每个第一缺陷特征与第二缺陷特征的相似度是否大于相似度阈值。
在本申请实施例中，在处理设备确定出各个第一缺陷特征与第二缺陷的特征的相似度后，处理设备可以检测每个第一缺陷特征与第二缺陷特征的相似度是否大于相似度阈值。
示例的，在处理设备确定出某个第一缺陷特征与第二缺陷特征的相似度大于相似度阈值时，执行步骤B3；在处理设备确定出某个第一缺陷特征与第二缺陷特征的相似度不大于相似度阈值时，执行步骤B4。
步骤B3、将相似度大于相似度阈值的第一缺陷特征所对应的缺陷确定为目标缺陷。
在本申请实施例中，在处理设备确定出某个第一缺陷特征与第二缺陷特征 的相似度大于相似度阈值时，该处理设备可以将第一缺陷特征所对应的缺陷确定为目标缺陷。
步骤B4、将相似度不大于相似度阈值的第一缺陷特征所对应的缺陷确定为其他缺陷。
在本申请实施例中，在处理设备确定出某个第一缺陷特征与第二缺陷特征的相似度不大于相似度阈值时，该处理设备可以将第一缺陷特征所对应的缺陷确定为多个缺陷中除目标缺陷之外的其他缺陷。
第三种可实现方式，缺陷的维修指示信息包括：缺陷的位置信息和缺陷类型信息。其中，该缺陷的位置信息的具体含义可以参考上述第一种可实现方式中的对应内容，该缺陷的缺陷类型信息的具体含义可以参考上述第二种可实现方式中的对应内容。在这种情况下，如图5所示，图5是本申请实施例提供的一种在多个缺陷中筛选目标缺陷的方法流程图。上述步骤304可以包括：
步骤3041、基于多个缺陷的位置信息，在多个缺陷中，将位于显示面板的指定区域内的缺陷确定为待选缺陷。
其中，在多个缺陷中筛选待选缺陷的具体流程及原理，可以参考上述第一种可实现方式中在多个缺陷中筛选目标缺陷的对应内容，本申请实施例在此不再赘述。需要说明的是，由于显示面板的缺陷的个数通常较多，因此，在多个缺陷中筛选出的待选缺陷的个数也为多个。
步骤3042、基于待选缺陷的缺陷类型信息，在多个待选缺陷中，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
其中，在多个待选缺陷中筛选目标缺陷的具体流程及原理，可以参考上述第二种可实现方式中在多个缺陷中筛选目标缺陷的对应内容，本申请实施例在此不再赘述。
在本申请实施例中，当缺陷位于指定区域外时，即使该缺陷的类型为指定类型的缺陷，该缺陷对显示面板的正常显示造成影响的概率也较小。当缺陷的类型不为指定类型的缺陷时，即使该缺陷位于指定区域内，该缺陷对显示面板的正常显示造成影响的概率也较小。
为此，在本申请中，当缺陷的维修指示信息包括：缺陷的位置信息和缺陷类型信息时，可以将位于指定区域内，且缺陷类型为指定类型的缺陷作为目标缺陷，这样不仅保证了筛选出的目标缺陷的个数可以进一步的降低，以进一步 的提高了AOI系统对显示面板的检测效率。还可以保证筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，以保证显示面板的良品率较高。
需要说明的是，上述实施例是以先基于缺陷的位置信息进行了初步筛选，再基于缺陷的缺陷类型信息进行了二次筛选为例进行示意性说明的。在其他可选的实现方式中，还可以先基于缺陷的缺陷类型信息进行初步筛选，再基于缺陷的位置信息进而二次筛选。本申请实施例对此不做限定。
步骤305、获取包含所述显示面板的第二图像。
在本申请实施例中，处理设备可以通过图像采集设备获取包含显示面板的第二图像。为了便于后续操作人员可以基于维修参考图像对缺陷进行维修，该维修参考图像通常为彩色图像。而维修参考图像是在第二图像中进行截取得到的，为此，该第二图像也通常为彩色图像。
示例的，该图像采集设备还可以包括：用于获取彩色图像的光学相机。处理设备可以通过该光学相机获取第二图像。需要说明的是，包含显示面板的第一图像和第二图像可以同时获取，也即是，上述步骤305与步骤301可以是同时执行的。
步骤306、基于目标缺陷的位置信息，在第二图像中截取包含目标缺陷的图像，以得到维修参考图像。
在本申请实施例中，在处理设备获取到第二图像和目标缺陷后，可以基于目标缺陷的位置信息，在第二图像中截取包含目标缺陷的图像。其中，该包含目标缺陷的图像即为目标缺陷的维修参考图像。后续，维修人员可以基于该维修参考图像对目标缺陷进行维修，以提高显示面板的良品率。
需要说明的是，本申请实施例提供的图像获取方法的步骤的先后顺序可以进行适当调整，步骤也可以根据情况进行相应增减，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化的方法，都应涵盖在本申请的保护范围之内，因此不再赘述。
综上所述，本申请实施例提供的图像获取方法，通过获取显示面板的多个缺陷维修指示信息，并基于该维修指示信息在多个缺陷中筛选目标缺陷，后续获取包含该目标缺陷的维修参考图像即可。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率较高，以保证显示面板的产能 较高。并且，在通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，有效的提高了AOI系统获取维修参考图像的准确度，保证该显示面板的良品率较高。
本申请实施例还提供了一种图像获取装置。如图6所示，图6是本申请实施例提供的一种图像获取装置的结构框图。该图像获取装置400可以集成在图1示出的AOI系统100中的处理设备102。该图像获取装置400可以包括：
获取模块401，用于获取包含显示面板的第一图像。
确定模块402，用于确定第一图像中的显示面板的多个缺陷的维修指示信息。该维修指示信息用于反映对缺陷的维修需求的程度。
筛选模块403，用于基于多个缺陷的维修指示信息，在多个缺陷中筛选目标缺陷。
拍摄模块404，用于对显示面板的目标缺陷进行拍摄，以得到包含目标缺陷的维修参考图像。
综上所述，本申请实施例提供的图像获取装置，通过获取显示面板的多个缺陷维修指示信息，并基于该维修指示信息在多个缺陷中筛选目标缺陷，后续获取包含该目标缺陷的维修参考图像即可。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率较高，以保证显示面板的产能较高。并且，在通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，有效的提高了AOI系统获取维修参考图像的准确度，保证该显示面板的良品率较高。
可选的，维修指示信息包括：位置信息。该筛选模块403，用于：基于多个缺陷的位置信息，将位于显示面板的指定区域内的缺陷确定为目标缺陷。
可选的，该指定区域包括：所述显示面板中的晶体管所在区域、两种信号线的交叉区域和相连的两个电极的连接区域中的至少一种。
可选的，维修指示信息包括：缺陷类型信息。该筛选模块403，用于：基于多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为目标缺陷。
可选的，该筛选模块403，用于：对第一图像中的多个缺陷进行特征提取，得到多个缺陷的第一缺陷特征，第一缺陷特征用于表征缺陷的缺陷类型信息；将提取得到第一缺陷特征与第二缺陷特征进行比对，以在多个缺陷中筛选目标 缺陷，第二缺陷特征用于表征指定类型的缺陷类型信息。
可选的，该筛选模块403，用于：确定各个第一缺陷特征与第二缺陷特征的相似度；将相似度大于相似度阈值的第一缺陷特征所对应的缺陷确定为目标缺陷。
可选的，维修指示信息包括：位置信息和缺陷类型信息。该筛选模块403，用于：基于多个缺陷的位置信息，在多个缺陷中，将位于显示面板的指定区域内的缺陷确定为待选缺陷；基于待选缺陷的缺陷类型信息，在多个待选缺陷中，将缺陷类型为指定类型的缺陷确定为目标缺陷。
可选的，拍摄模块404用于：获取包含显示面板的第二图像；基于目标缺陷的位置信息，在第二图像中截取包含目标缺陷的图像，以得到维修参考图。
综上所述，本申请实施例提供的图像获取装置，通过获取显示面板的多个缺陷维修指示信息，并基于该维修指示信息在多个缺陷中筛选目标缺陷，后续获取包含该目标缺陷的维修参考图像即可。无需获取显示面板全部缺陷的维修参考图像，保证AOI系统对显示面板的检测效率较高，以保证显示面板的产能较高。并且，在通过维修指示信息筛选出的目标缺陷需要被维修的概率较高，未被筛选出的其他缺陷需要被维修的概率较低，有效的提高了AOI系统获取维修参考图像的准确度，保证该显示面板的良品率较高。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置、模块和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
本申请实施例还提供了一种图像获取设备。该图像获取设备可以包括：处理器，以及用于存储该处理器的可执行指令的存储器。其中，该处理器被配置为执行图2或图3示出的图像获取方法。
本申请实施例还提供了一种计算机可读存储介质。该计算机可读存储介质中存储有指令，当该计算机可读存储介质在处理组件上运行时，使得该处理组件执行图2或图3示出的图像获取方法。
在本申请中，术语“第一”和“第二”仅用于描述目的，而不能理解为指 示或暗示相对重要性。术语“多个”指两个或两个以上，除非另有明确的限定。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本申请的可选的实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。
Claims (15)
- 一种图像获取方法，其特征在于，所述方法包括：获取包含显示面板的第一图像；确定所述第一图像中的所述显示面板的多个缺陷的维修指示信息，所述维修指示信息用于反映对缺陷的维修需求的程度；基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷；对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像。
- 根据权利要求1所述的方法，其特征在于，所述维修指示信息包括：位置信息和缺陷类型信息中的至少一种。
- 根据权利要求2所述的方法，其特征在于，所述维修指示信息包括：所述位置信息，基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷，包括：基于所述多个缺陷的位置信息，将位于所述显示面板的指定区域内的缺陷确定为所述目标缺陷。
- 根据权利要求3所述的方法，其特征在于，所述指定区域包括：所述显示面板中的晶体管所在区域、两种信号线的交叉区域和相连的两个电极的连接区域中的至少一种。
- 根据权利要求2所述的方法，其特征在于，所述维修指示信息包括：所述缺陷类型信息，基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷，包括：基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
- 根据权利要求5所述的方法，其特征在于，基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷，包括：对所述第一图像中的所述多个缺陷进行特征提取，得到所述多个缺陷的第一缺陷特征，所述第一缺陷特征用于表征所述缺陷的缺陷类型信息；将提取得到所述第一缺陷特征与第二缺陷特征进行比对，以在所述多个缺陷中筛选目标缺陷，所述第二缺陷特征用于表征所述指定类型的缺陷类型信息。
- 根据权利要求6所述的方法，其特征在于，将各个所述第一缺陷特征与第二缺陷特征进行比对，以在多个所述缺陷中筛选目标缺陷，包括：确定各个所述第一缺陷特征与所述第二缺陷特征的相似度；将相似度大于相似度阈值的第一缺陷特征所对应的缺陷确定为所述目标缺陷。
- 根据权利要求2所述的方法，其特征在于，所述维修指示信息包括：所述位置信息和所述缺陷类型信息，基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷，包括：基于所述多个缺陷的位置信息，在所述多个缺陷中，将位于所述显示面板的指定区域内的缺陷确定为待选缺陷；基于所述待选缺陷的缺陷类型信息，在多个所述待选缺陷中，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
- 根据权利要求1至8任一所述的方法，其特征在于，对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像，包括：获取包含所述显示面板的第二图像；基于所述目标缺陷的位置信息，在所述第二图像中截取包含所述目标缺陷的图像，以得到所述维修参考图像。
- 一种图像获取装置，其特征在于，所述装置包括：获取模块，用于获取包含显示面板的第一图像；确定模块，用于确定所述第一图像中的所述显示面板的多个缺陷的维修指 示信息，所述维修指示信息用于反映对缺陷的维修需求的程度；筛选模块，用于基于所述多个缺陷的维修指示信息，在所述多个缺陷中筛选目标缺陷；拍摄模块，用于对所述显示面板的所述目标缺陷进行拍摄，以得到包含所述目标缺陷的维修参考图像。
- 根据权利要求10所述的装置，其特征在于，所述维修指示信息包括：位置信息，所述筛选模块，用于：基于所述多个缺陷的位置信息，将位于所述显示面板的指定区域内的缺陷确定为所述目标缺陷。
- 根据权利要求10所述的装置，其特征在于，所述维修指示信息包括：缺陷类型信息，所述筛选模块，用于：基于所述多个缺陷的缺陷类型信息，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
- 根据权利要求10所述的装置，其特征在于，所述维修指示信息包括：位置信息和缺陷类型信息，所述筛选模块，用于：基于所述多个缺陷的位置信息，在所述多个缺陷中，将位于所述显示面板的指定区域内的缺陷确定为待选缺陷；基于所述待选缺陷的缺陷类型信息，在多个所述待选缺陷中，将缺陷类型为指定类型的缺陷确定为所述目标缺陷。
- 根据权利要求10至13任一所述的装置，其特征在于，所述拍摄模块，用于：获取包含所述显示面板的第二图像；基于所述目标缺陷的位置信息，在所述第二图像中截取包含所述目标缺陷的图像，以得到所述维修参考图像。
- 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有指令，当所述可读存储介质在处理组件上运行时，使得所述处理组件执行如权利要求1至9任一所述的图像获取方法。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2021/095789 WO2022246643A1 (zh) | 2021-05-25 | 2021-05-25 | 图像获取方法及装置、存储介质 |
| CN202180001255.8A CN115803610A (zh) | 2021-05-25 | 2021-05-25 | 图像获取方法及装置、存储介质 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2021/095789 WO2022246643A1 (zh) | 2021-05-25 | 2021-05-25 | 图像获取方法及装置、存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022246643A1 true WO2022246643A1 (zh) | 2022-12-01 |
Family
ID=84229157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2021/095789 WO2022246643A1 (zh) | 2021-05-25 | 2021-05-25 | 图像获取方法及装置、存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115803610A (zh) |
| WO (1) | WO2022246643A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058090A (zh) * | 2023-08-11 | 2023-11-14 | 曲靖阳光新能源股份有限公司 | 用于检测单晶硅片缺陷的方法和装置 |
| CN117058090B (zh) * | 2023-08-11 | 2024-05-17 | 曲靖阳光新能源股份有限公司 | 用于检测单晶硅片缺陷的方法和装置 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010109856A1 (ja) * | 2009-03-26 | 2010-09-30 | シャープ株式会社 | 表示パネル点灯検査装置、及び表示パネル点灯検査方法 |
| CN109001208A (zh) * | 2018-05-28 | 2018-12-14 | 南京中电熊猫平板显示科技有限公司 | 一种显示面板的缺陷定位装置及缺陷定位方法 |
| CN109118482A (zh) * | 2018-08-07 | 2019-01-01 | 腾讯科技（深圳）有限公司 | 一种面板缺陷分析方法、装置及存储介质 |
| CN111563889A (zh) * | 2020-05-06 | 2020-08-21 | 深圳市斑马视觉科技有限公司 | 基于计算机视觉的液晶屏幕Mura缺陷检测方法 |
| CN112184667A (zh) * | 2020-09-28 | 2021-01-05 | 京东方科技集团股份有限公司 | 缺陷检测、修复方法、装置以及存储介质 |
-
2021
- 2021-05-25 CN CN202180001255.8A patent/CN115803610A/zh active Pending
- 2021-05-25 WO PCT/CN2021/095789 patent/WO2022246643A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010109856A1 (ja) * | 2009-03-26 | 2010-09-30 | シャープ株式会社 | 表示パネル点灯検査装置、及び表示パネル点灯検査方法 |
| CN109001208A (zh) * | 2018-05-28 | 2018-12-14 | 南京中电熊猫平板显示科技有限公司 | 一种显示面板的缺陷定位装置及缺陷定位方法 |
| CN109118482A (zh) * | 2018-08-07 | 2019-01-01 | 腾讯科技（深圳）有限公司 | 一种面板缺陷分析方法、装置及存储介质 |
| CN111563889A (zh) * | 2020-05-06 | 2020-08-21 | 深圳市斑马视觉科技有限公司 | 基于计算机视觉的液晶屏幕Mura缺陷检测方法 |
| CN112184667A (zh) * | 2020-09-28 | 2021-01-05 | 京东方科技集团股份有限公司 | 缺陷检测、修复方法、装置以及存储介质 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058090A (zh) * | 2023-08-11 | 2023-11-14 | 曲靖阳光新能源股份有限公司 | 用于检测单晶硅片缺陷的方法和装置 |
| CN117058090B (zh) * | 2023-08-11 | 2024-05-17 | 曲靖阳光新能源股份有限公司 | 用于检测单晶硅片缺陷的方法和装置 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115803610A (zh) | 2023-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10388010B2 (en) | Image acquiring device and method of correcting coordinates thereof | |
| WO2019227762A1 (zh) | 光学模组像素缺陷检测方法、装置及设备 | |
| CN110261410A (zh) | 一种玻璃透镜缺陷的检测装置及方法 | |
| WO2020007118A1 (zh) | 显示屏外围电路检测方法、装置、电子设备及存储介质 | |
| WO2020007119A1 (zh) | 显示屏外围电路检测方法、装置、电子设备及存储介质 | |
| CN113240673B (zh) | 缺陷检测方法、装置、电子设备及存储介质 | |
| CN104749801B (zh) | 高精度自动光学检测方法和系统 | |
| JP5088165B2 (ja) | 欠陥検出方法および欠陥検出装置 | |
| CN116559177A (zh) | 一种缺陷检测方法、装置、设备以及存储介质 | |
| CN108107611B (zh) | 一种自适应缺陷检测方法、装置及电子设备 | |
| CN117392042A (zh) | 缺陷检测方法、缺陷检测设备及存储介质 | |
| WO2022134305A1 (zh) | 晶圆缺陷检测方法和装置、设备及存储介质 | |
| KR20120105149A (ko) | 평판패널 기판의 자동광학검사 방법 및 그 장치 | |
| CN113870204A (zh) | 一种光学玻璃的异常检测方法及装置 | |
| WO2022246643A1 (zh) | 图像获取方法及装置、存储介质 | |
| JP2014126445A (ja) | 位置合せ装置、欠陥検査装置、位置合せ方法、及び制御プログラム | |
| CN115690077A (zh) | 一种泡罩板背面缺陷检测方法及系统 | |
| CN115035168A (zh) | 基于多约束的光伏面板多源图像配准方法、装置及系统 | |
| US11508143B2 (en) | Automated salience assessment of pixel anomalies | |
| CN112052727B (zh) | 一种基于大数据的人像识别分析系统及方法 | |
| WO2020158630A1 (ja) | 検出装置、学習器、コンピュータプログラム、検出方法及び学習器の生成方法 | |
| TWI747257B (zh) | 一種檢查鏤空晶圓的方法 | |
| JP5239275B2 (ja) | 欠陥検出方法および欠陥検出装置 | |
| CN114549414A (zh) | 一种针对轨道数据的异常变化检测方法及系统 | |
| EP4075384A1 (en) | Methods and systems for determining pre-determined points in an input image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 17771764 Country of ref document: US |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21942230 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |