WO2016095318A1 - 显示面板缺陷的自动检测方法 - Google Patents

显示面板缺陷的自动检测方法 Download PDF

Info

Publication number
WO2016095318A1
WO2016095318A1 PCT/CN2015/071129 CN2015071129W WO2016095318A1 WO 2016095318 A1 WO2016095318 A1 WO 2016095318A1 CN 2015071129 W CN2015071129 W CN 2015071129W WO 2016095318 A1 WO2016095318 A1 WO 2016095318A1
Authority
WO
WIPO (PCT)
Prior art keywords
value
image
mark
original
original image
Prior art date
Application number
PCT/CN2015/071129
Other languages
English (en)
French (fr)
Inventor
胡厚亮
朱立伟
Original Assignee
深圳市华星光电技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市华星光电技术有限公司 filed Critical 深圳市华星光电技术有限公司
Priority to US14/425,048 priority Critical patent/US9646370B2/en
Publication of WO2016095318A1 publication Critical patent/WO2016095318A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • 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
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • 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
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • 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
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G01N2021/889—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques providing a bare video image, i.e. without visual measurement aids
    • 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
    • G01N2021/9513—Liquid crystal panels
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20132—Image cropping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30121—CRT, LCD or plasma display

Definitions

  • the present invention belongs to the technical field of manufacturing display panels, and in particular, to an automatic detection method for defects of a display panel in a manufacturing process of a display panel.
  • a camera is usually used to acquire an original image generated by the display panel, and then the acquired original image is transmitted to a computer, and the original image is analyzed and processed by a computer to obtain a defect on the display panel.
  • the existing method of detecting a display panel defect cannot accurately acquire the position of the defect and the difference between the defect and the normal area, so that the defect on the display panel cannot be accurately quantified and judged.
  • an object of the present invention is to provide an automatic detection method for a defect of a display panel, comprising: acquiring a mark image, mapping an original image, and mapping a mark image; and dividing the mapped original image into several sub-objects Mapping the original image, and dividing the mapping mark image into a plurality of sub-map mark images; acquiring a normal area and a defect area of the original image of the sub-map; and combining the original images of the sub-maps to distinguish between a normal area and a defect area Mapping the original image; correcting the mapped original image of the normal region and the defective region by using the mapping mark image and the mark image to obtain a defect position of the display panel.
  • the method for acquiring the mark image, the map original image, and the map mark image includes: setting a plurality of mark points in an original image displayed by the display panel to obtain the mark image; and the original image and The mark image is sampled to obtain the original sample image and mark Cutting the original sample image and the mark sample image by using a plurality of corner points to obtain an original cut image and a mark cut image; respectively rotating the original cut image and the mark cut image to Obtaining an original rotated image and a mark rotating image; respectively cutting and stretching the original rotated image and the mark rotated image to obtain an original stretched image and a mark stretched image; and the original stretched image and the The marker stretched image is respectively blackened by pixels, and the black pixel is filled by linear interpolation to acquire the mapped original image and the map mark image.
  • mapping original image has the same resolution as the original image
  • mapping mark image has the same resolution as the original image
  • the method for determining the corner point includes: establishing a plurality of mask matrices, wherein the number of the mask matrices is the same as the number of the mark points; and convolving the mark image by using the mask matrix Obtaining a convolution value; determining whether a Euclidean distance between a point at which the convolution value is greater than a predetermined threshold and a vertex of the marked sample image is minimum; a convolution value greater than the predetermined threshold and a vertex of the sampled image with the marker The smallest point between the Euclidean distances is used as the corner point.
  • the method for acquiring the normal area and the defect area of the original image of the sub-map includes: acquiring a luminance histogram of the original image of the sub-map; determining a maximum brightness value and a minimum brightness value based on the brightness histogram; determining the Whether the brightness value of the pixel in the sub-map original image is greater than the minimum brightness value and less than the maximum brightness value; if the brightness value of the pixel in the sub-map original image is greater than the minimum brightness value and less than the maximum brightness value, the child The pixel in the original image is mapped to the normal region.
  • the pixel in the sub-map original image is a defect area if the brightness value of the pixel in the sub-map original image is not greater than the minimum brightness value or not less than the maximum brightness value, the pixel in the sub-map original image is a defect area.
  • the determining method of the maximum brightness value and the minimum brightness value includes: determining, in the brightness histogram, a maximum brightness initial value and a minimum brightness value initial value of a pixel whose ratio is not less than a predetermined number of proportions Calculating a first weighted average value of luminance values of pixels in the sub-map original image that is greater than the initial value of the maximum luminance, and calculating a luminance value of a pixel in the sub-map original image that is smaller than an initial value of the minimum luminance value a second weighted average value; when a difference between the first weighted average value and the maximum brightness initial value is greater than a threshold value, and a difference between the minimum brightness value initial value and the second weighted average value is greater than In the threshold value, the initial value of the maximum brightness and the initial value of the minimum brightness value are respectively taken as the maximum brightness value and the minimum brightness value.
  • the threshold value when a difference between the first weighted average value and the maximum brightness initial value is not greater than the threshold value and/or the difference between the minimum value of the minimum brightness value and the second weighted average value is not greater than
  • the threshold value is set, the predetermined number of proportions is increased by 0.01, and the maximum brightness initial value and the minimum brightness initial value are re-determined.
  • the method for determining the maximum brightness value and the minimum brightness value further includes: when the predetermined number of ratios is 1, the maximum brightness initial value and the minimum brightness initial value are respectively taken as the maximum A brightness value and the minimum brightness value.
  • the method for correcting the mapped original image that distinguishes the normal region and the defect region includes: calculating a plurality of corresponding coordinate deviation values of the plurality of marker points in the marker stretched image and the plurality of marker points of the marker image And correcting the mapped original image that distinguishes the normal region and the defective region by using the plurality of corresponding coordinate deviation values.
  • the automatic detection method for the defect of the display panel of the present invention can accurately acquire the position of the defect and the difference between the defect and the normal region, thereby accurately quantizing and judging the defect on the display panel.
  • FIG. 1 is a flow chart of an automatic detection method for display panel defects in accordance with an embodiment of the present invention.
  • FIG. 1 is a flow chart of an automatic detection method for display panel defects in accordance with an embodiment of the present invention.
  • step 110 a mark image, a map original image, and a map mark image are acquired.
  • the method for acquiring the mark image, mapping the original image, and mapping the mark image includes:
  • Step S111 setting a plurality of mark points in the original image displayed by the display panel to obtain the mark image.
  • four mark points may be marked in the upper left corner, the lower left corner, the upper right corner, and the lower right corner of the original image to form a mark image. It should be understood that the number of marked points in the original image is not limited to four.
  • Step S112 sampling the original image and the mark image to obtain the original sample image and the mark sample image.
  • the display panel displaying the original image and the marked image may be separately sampled by setting parameters such as camera aperture, sensitivity (ISO), shutter, and image processing function, and converted into brightness of the display panel by RGB-YCbCr color space.
  • the value that is, the brightness value of the original sampled image and the brightness value of the marked sample image.
  • Step S113 Determine a plurality of corner points, and use the plurality of corner points to respectively cut the original sample image and the mark sample image to obtain the original cut image and the mark cut image.
  • the method for determining the corner point comprises:
  • S1131 Establish a mask matrix corresponding to the number of points. Here, four mask matrices are created.
  • S1133 Determine whether the Euclidean distance between the point where the convolution value is greater than a predetermined threshold and the vertex of the marked sample image is the smallest.
  • the predetermined threshold is k*gray*16, where k is a luminance mapping coefficient which is set according to a difference in luminance between the marker image and the marker sample image, and gray is a luminance value of the marker sample image.
  • S1134 A point where the convolution value is greater than a predetermined threshold and the Euclidean distance between the vertices of the marked sample image is the smallest is used as a corner point.
  • Step S114 respectively rotating the original cut image and the mark cut image to obtain the original rotated image and the mark rotated image.
  • the position where the camera and the display panel are placed is not strictly horizontal (or vertical) in the horizontal plane (or vertical plane)
  • a gap occurs in the upper right corner and the lower right corner of the original cut image and the mark cut image, and the gap follows the camera and
  • the placement position of the display panel also appears in the upper left corner and the lower left corner.
  • the original cut image and the mark cut image are respectively rotated by the rotation operator.
  • Step S115 Perform cutting and stretching on the original rotated image and the mark rotated image respectively to obtain the original stretched image and the mark stretched image.
  • Step S116 performing pixel insertion blackning on the original stretched image and the mark stretched image respectively, and filling the black pixel by linear interpolation to obtain the mapped original image and the mapped mark image.
  • the mapping original image has the same resolution as the original image
  • the mapping mark image has the same resolution as the original image.
  • the mapped original image is divided into a number of sub-mapped original images
  • the mapped mark image is divided into a number of sub-mapped mark images.
  • the mapped original image and the map mark image are divided into regions.
  • step 130 a normal area and a defective area of the original image of the sub-map are acquired.
  • the method for obtaining the normal area and the defective area of the sub-map original image includes:
  • Step S131 Acquire a luminance histogram of the original image of the sub-map.
  • Step S132 Determine a maximum brightness value and a minimum brightness value based on the luminance histogram.
  • Step S133 determining whether the brightness value of the pixel in the original image of the sub-map is greater than the minimum brightness value and less than the maximum brightness value, wherein if the determination is yes, the brightness value of the pixel in the original image of the sub-map is greater than the minimum brightness value and less than the maximum brightness value. Then, the pixel in the original image of the sub-map is a normal region; if the determination is no, that is, the luminance value of the pixel in the original image of the sub-map is not greater than the minimum luminance value or not less than the maximum luminance value, the pixel in the original image of the sub-map is the defect region.
  • step S132 the determining method of determining the maximum brightness value and the minimum brightness value based on the luminance histogram comprises:
  • Step S1321 determining, in the luminance histogram, a maximum brightness initial value and a minimum brightness initial value of the pixels whose number is not less than a predetermined number of proportions. For example, the maximum brightness initial value and the minimum brightness initial value of the first 80% of the pixels in the luminance histogram.
  • Step S1322 Calculate a first weighted average value of pixel brightness in the original image of the sub-map larger than the initial value of the maximum brightness, and calculate a second increase of the brightness of the pixel in the original image of the sub-map smaller than the initial value of the minimum brightness Average weight.
  • Step S1323 When the difference between the first weighted average value and the maximum brightness initial value is greater than a threshold value, and the difference between the minimum brightness initial value and the second weighted average value is greater than a threshold value, the maximum brightness initial value and the minimum brightness initial value are obtained.
  • the values are respectively taken as the maximum brightness value and the minimum brightness value; otherwise, the difference between the first weighted average value and the maximum brightness initial value is not greater than the threshold value and/or the difference between the minimum brightness initial value and the second weighted average value is not greater than
  • the threshold is thresholded
  • the predetermined amount is added to the ratio of 0.01, and the maximum luminance initial value and the minimum luminance initial value are re-determined.
  • the determining method of determining the maximum brightness value and the minimum brightness value based on the brightness histogram further comprises: when the predetermined quantity ratio is 100%, the maximum brightness initial value and the minimum brightness initial value are respectively taken as the maximum brightness Value and minimum brightness value.
  • each of the sub-map original images is merged into a mapped original image that resolves the normal and defective regions.
  • step 150 the mapped original image of the normal area and the defective area is corrected by using the mapping mark image and the mark image to obtain the defect position of the display panel.
  • the method for correcting the mapped original image that distinguishes the normal region from the defective region includes:
  • Step S151 Calculate a corresponding coordinate deviation value of the marker point in the marker stretched image and the marker point of the marker image.
  • Step S152 Correcting the mapped original image that distinguishes the normal area and the defective area by using a plurality of corresponding coordinate deviation values.
  • the automatic detection method for the defect of the display panel can accurately acquire the position of the defect and the difference between the defect and the normal region, thereby accurately quantizing and judging the defect on the display panel.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Quality & Reliability (AREA)
  • Signal Processing (AREA)
  • Image Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

一种显示面板缺陷的自动检测方法，包括：获取标记图像、映射原始图像及映射标记图像(110)；将所述映射原始图像划分为若干子映射原始图像，且将所述映射标记图像划分为若干子映射标记图像(120)；获取所述子映射原始图像的正常区与缺陷区(130)；将各所述子映射原始图像合并为分辨出正常区和缺陷区的映射原始图像(140)；利用所述映射标记图像与所述标记图像对所述分辨出正常区和缺陷区的映射原始图像进行校正，以获取显示面板的缺陷位置(150)。该显示面板缺陷的自动检测方法，能够准确地获取缺陷的位置以及缺陷与正常区域的差异，从而能够准确地量化和评判显示面板上的缺陷。

Description

显示面板缺陷的自动检测方法 技术领域
本发明属于显示面板的制造技术领域，具体地讲，涉及一种在显示面板的制造过程中的显示面板缺陷的自动检测方法。
背景技术
随着显示技术的快速发展，能够实现高品质图像显示的显示面板(诸如，液晶显示面板或者有机发光显示面板)已成为市场的主流。然而，针对现有的显示面板制造技术，完全避免显示缺陷的发生是非常困难的，因此在显示面板的制造工序中，对显示面板进行显示缺陷的检查工序是十分必要的。
在现有技术的检测显示面板缺陷的方法中，通常采用照相机获取显示面板产生的原始图像，然后将获取的原始图像传递至计算机，通过计算机对原始图像进行分析处理，从而获得显示面板上的缺陷。然而，现有的检测显示面板缺陷的方法无法准确地获取缺陷的位置以及缺陷与正常区域的差异，从而无法准确地量化和评判显示面板上的缺陷。
发明内容
为了解决上述现有技术存在的问题，本发明的目的在于提供一种显示面板缺陷的自动检测方法，包括：获取标记图像、映射原始图像及映射标记图像；将所述映射原始图像划分为若干子映射原始图像，且将所述映射标记图像划分为若干子映射标记图像；获取所述子映射原始图像的正常区与缺陷区；将各所述子映射原始图像合并为分辨出正常区和缺陷区的映射原始图像；利用所述映射标记图像与所述标记图像对所述分辨出正常区和缺陷区的映射原始图像进行校正，以获取显示面板的缺陷位置。
进一步地，获取所述标记图像、所述映射原始图像及所述映射标记图像的方法包括：在显示面板显示的原始图像中设置若干标记点，以获取所述标记图像；对所述原始图像及所述标记图像进行采样，以获取原始采样图像及标记采 样图像；利用若干角点对所述原始采样图像及所述标记采样图像分别进行切割，以获取原始切割图像及标记切割图像；对所述原始切割图像及所述标记切割图像分别进行旋转，以获取原始旋转图像及标记旋转图像；对所述原始旋转图像及所述标记旋转图像分别进行再次切割与拉伸，以获取原始拉伸图像及标记拉伸图像；对所述原始拉伸图像及所述标记拉伸图像分别进行像素插黑，并利用线性插值方式填补插黑像素，以获取所述映射原始图像及所述映射标记图像。
进一步地，所述映射原始图像与所述原始图像的解析度相同，且所述映射标记图像与所述原始图像的解析度相同。
进一步地，所述角点的确定方法包括：建立若干掩膜矩阵，其中，所述掩膜矩阵的数量与所述标记点的数量相同；利用所述掩膜矩阵对所述标记图像进行卷积，以获取卷积值；判断卷积值大于预定阈值的点与所述标记采样图像的顶点之间的欧式距离是否最小；将卷积值大于所述预定阈值且与所述标记采样图像的顶点之间的欧式距离最小的点作为角点。
进一步地，获取所述子映射原始图像的正常区与缺陷区的方法包括：获取所述子映射原始图像的亮度直方图；基于所述亮度直方图确定最大亮度值和最小亮度值；判断所述子映射原始图像中像素的亮度值是否大于所述最小亮度值且小于最大亮度值；如果所述子映射原始图像中像素的亮度值大于所述最小亮度值且小于最大亮度值，则所述子映射原始图像中该像素为正常区。
进一步地，如果所述子映射原始图像中像素的亮度值不大于所述最小亮度值或不小于最大亮度值，则所述子映射原始图像中该像素为缺陷区。
进一步地，所述最大亮度值和所述最小亮度值的确定方法包括：在所述亮度直方图中确定数量占比不小于一预定数量占比的像素的最大亮度初值和最小亮度值初值；计算大于所述最大亮度初值的所述子映射原始图像中像素的亮度值的第一加权平均值，且计算小于所述最小亮度值初值的所述子映射原始图像中像素的亮度值的第二加权平均值；当所述第一加权平均值与所述最大亮度初值的差值大于一门槛值，且所述最小亮度值初值与所述第二加权平均值的差值大于所述门槛值时，将所述最大亮度初值和所述最小亮度值初值分别作为所述最大亮度值和所述最小亮度值。
进一步地，当所述第一加权平均值与所述最大亮度初值的差值不大于所述门槛值和/或所述最小亮度值初值与所述第二加权平均值的差值不大于所述门槛值时，将所述预定数量占比加0.01，且重新确定最大亮度初值和最小亮度初值。
进一步地，所述最大亮度值和所述最小亮度值的确定方法还包括：当所述预定数量占比为1时，将所述最大亮度初值和所述最小亮度初值分别作为所述最大亮度值和所述最小亮度值。
进一步地，对所述分辨出正常区和缺陷区的映射原始图像进行校正的方法包括：计算所述标记拉伸图像中的若干标记点与所述标记图像的若干标记点的若干对应坐标偏差值；利用所述若干对应坐标偏差值对所述分辨出正常区和缺陷区的映射原始图像进行校正。
本发明的显示面板缺陷的自动检测方法，能够准确地获取缺陷的位置以及缺陷与正常区域的差异，从而能够准确地量化和评判显示面板上的缺陷。
附图说明
通过结合附图进行的以下描述，本发明的实施例的上述和其它方面、特点和优点将变得更加清楚，附图中：
图1是根据本发明的实施例的显示面板缺陷的自动检测方法的流程图。
具体实施方式
以下，将参照附图来详细描述本发明的实施例。然而，可以以许多不同的形式来实施本发明，并且本发明不应该被解释为限制于这里阐述的具体实施例。相反，提供这些实施例是为了解释本发明的原理及其实际应用，从而使本领域的其他技术人员能够理解本发明的各种实施例和适合于特定预期应用的各种修改。
图1是根据本发明的实施例的显示面板缺陷的自动检测方法的流程图。
参照图1，在步骤110中，获取标记图像、映射原始图像及映射标记图像。
这里，获取标记图像、映射原始图像及映射标记图像的方法包括：
步骤S111：在显示面板显示的原始图像中设置若干标记点，以获取标记图像。其中，可在原始图像的左上角、左下角、右上角和右下角分别标记四个标记点，以形成标记图像。应当理解的是，原始图像中的标记点的数量不限于四个。
步骤S112：对原始图像及标记图像进行采样，以获取原始采样图像及标记采样图像。其中，可通过设置照相机光圈、感光度(ISO)、快门、关闭图像处理功能等参数，分别对显示原始图像及标记图像的显示面板进行采样，并通过RGB-YCbCr颜色空间转换为显示面板的亮度值，即原始采样图像的亮度值及标记采样图像的亮度值。
步骤S113：确定若干角点，并利用该若干角点对原始采样图像及标记采样图像分别进行切割，以获取原始切割图像及标记切割图像。
其中，所述角点的确定方法包括：
S1131：建立数量与标记点数量相应的掩膜矩阵。这里，建立四个掩膜矩阵。
S1132：利用该掩膜矩阵分别对标记采样图像进行卷积，以获取卷积值。
S1133：判断卷积值大于一预定阈值的点与标记采样图像的顶点之间的欧式距离是否最小。这里，所述预定阈值为k*gray*16，其中，k为亮度映射系数，其根据标记图像与标记采样图像的亮度差异进行设置，gray为标记采样图像的亮度值。
S1134：将卷积值大于预定阈值且与标记采样图像的顶点之间的欧式距离最小的点作为角点。
步骤S114：对原始切割图像及标记切割图像分别进行旋转，以获取原始旋转图像及标记旋转图像。这里，由于照相机与显示面板放置的位置在水平面(或垂直面)不严格水平(或垂直)，会造成原始切割图像和标记切割图像的右上角、右下角出现空隙，这种空隙随着照相机与显示面板的放置位置也会出现在左上角、左下角，为了校正这种放置位置带来的误差，通过旋转算子对原始切割图像及标记切割图像分别进行旋转。
步骤S115：对原始旋转图像及标记旋转图像分别进行再次切割与拉伸，以获取原始拉伸图像及标记拉伸图像。
步骤S116：对原始拉伸图像及标记拉伸图像分别进行像素插黑，并利用线性插值方式填补插黑像素，以获取映射原始图像及映射标记图像。这里，映射原始图像与原始图像的解析度相同，且映射标记图像与原始图像的解析度相同。
在步骤120中，将映射原始图像划分为若干子映射原始图像，且将映射标记图像划分为若干子映射标记图像。这里，为了避免由于背光造成的显示面板(例如，LCD面板)的亮度的不均匀和照相机采样过程造成的亮度不均匀带来对缺陷的误判，将映射原始图像与映射标记图像进行划分区域。
在步骤130中，获取子映射原始图像的正常区与缺陷区。
这里，获取子映射原始图像的正常区与缺陷区的方法包括：
步骤S131：获取子映射原始图像的亮度直方图。
步骤S132：基于亮度直方图确定最大亮度值和最小亮度值。
步骤S133：判断子映射原始图像中像素的亮度值是否大于最小亮度值且小于最大亮度值，其中，如果判断为是，即子映射原始图像中像素的亮度值大于最小亮度值且小于最大亮度值，则子映射原始图像中像素为正常区；如果判断为否，即子映射原始图像中像素的亮度值不大于最小亮度值或不小于最大亮度值，则子映射原始图像中像素为缺陷区。
进一步地，在步骤S132中，基于亮度直方图确定最大亮度值和最小亮度值的确定方法包括：
步骤S1321：在亮度直方图中确定数量占比不小于一预定数量占比的像素的最大亮度初值和最小亮度初值。例如，亮度直方图中数量前80％的像素的最大亮度初值和最小亮度初值。
步骤S1322：计算大于最大亮度初值的子映射原始图像中像素亮度的第一加权平均值，且计算小于最小亮度初值的子映射原始图像中像素亮度的第二加 权平均值。
步骤S1323：当第一加权平均值与最大亮度初值的差值大于一门槛值，且最小亮度初值与第二加权平均值的差值大于门槛值时，将最大亮度初值和最小亮度初值分别作为最大亮度值和最小亮度值；否则，即当第一加权平均值与最大亮度初值的差值不大于门槛值和/或最小亮度初值与第二加权平均值的差值不大于门槛值时，将预定数量占比加0.01，且重新确定最大亮度初值和最小亮度初值。
进一步地，在步骤S132中，基于亮度直方图确定最大亮度值和最小亮度值的确定方法还包括：当预定数量占比为100％时，将最大亮度初值和最小亮度初值分别作为最大亮度值和最小亮度值。
在步骤140中，将各子映射原始图像合并为分辨出正常区和缺陷区的映射原始图像。
在步骤150中，利用映射标记图像与标记图像对分辨出正常区和缺陷区的映射原始图像进行校正，以获取显示面板的缺陷位置。
这里，对分辨出正常区和缺陷区的映射原始图像进行校正的方法包括：
步骤S151：计算标记拉伸图像中的标记点与标记图像的标记点的对应坐标偏差值。
步骤S152：利用若干对应坐标偏差值对分辨出正常区和缺陷区的映射原始图像进行校正。
综上，根据本发明的实施例的显示面板缺陷的自动检测方法，能够准确地获取缺陷的位置以及缺陷与正常区域的差异，从而能够准确地量化和评判显示面板上的缺陷。
虽然已经参照特定实施例示出并描述了本发明，但是本领域的技术人员将理解：在不脱离由权利要求及其等同物限定的本发明的精神和范围的情况下，可在此进行形式和细节上的各种变化。

Claims (14)

  1. 一种显示面板缺陷的自动检测方法，其中，包括：
    获取标记图像、映射原始图像及映射标记图像；
    将所述映射原始图像划分为若干子映射原始图像，且将所述映射标记图像划分为若干子映射标记图像；
    获取所述子映射原始图像的正常区与缺陷区；
    将各所述子映射原始图像合并为分辨出正常区和缺陷区的映射原始图像；
    利用所述映射标记图像与所述标记图像对所述分辨出正常区和缺陷区的映射原始图像进行校正，以获取显示面板的缺陷位置。
  2. 根据权利要求1所述的自动检测方法，其中，获取所述标记图像、所述映射原始图像及所述映射标记图像的方法包括：
    在显示面板显示的原始图像中设置若干标记点，以获取所述标记图像；
    对所述原始图像及所述标记图像进行采样，以获取原始采样图像及标记采样图像；
    利用若干角点对所述原始采样图像及所述标记采样图像分别进行切割，以获取原始切割图像及标记切割图像；
    对所述原始切割图像及所述标记切割图像分别进行旋转，以获取原始旋转图像及标记旋转图像；
    对所述原始旋转图像及所述标记旋转图像分别进行再次切割与拉伸，以获取原始拉伸图像及标记拉伸图像；
    对所述原始拉伸图像及所述标记拉伸图像分别进行像素插黑，并利用线性插值方式填补插黑像素，以获取所述映射原始图像及所述映射标记图像。
  3. 根据权利要求2所述的自动检测方法，其中，所述映射原始图像与所述原始图像的解析度相同，且所述映射标记图像与所述原始图像的解析度相同。
  4. 根据权利要求2所述的自动检测方法，其中，所述角点的确定方法包括：
    建立若干掩膜矩阵，其中，所述掩膜矩阵的数量与所述标记点的数量相同；
    利用所述掩膜矩阵对所述标记图像进行卷积，以获取卷积值；
    判断卷积值大于预定阈值的点与所述标记采样图像的顶点之间的欧式距离是否最小；
    将卷积值大于所述预定阈值且与所述标记采样图像的顶点之间的欧式距离最小的点作为角点。
  5. 根据权利要求3所述的自动检测方法，其中，所述角点的确定方法包括：
    建立若干掩膜矩阵，其中，所述掩膜矩阵的数量与所述标记点的数量相同；
    利用所述掩膜矩阵对所述标记图像进行卷积，以获取卷积值；
    判断卷积值大于预定阈值的点与所述标记采样图像的顶点之间的欧式距离是否最小；
    将卷积值大于所述预定阈值且与所述标记采样图像的顶点之间的欧式距离最小的点作为角点。
  6. 根据权利要求1所述的自动检测方法，其中，获取所述子映射原始图像的正常区与缺陷区的方法包括：
    获取所述子映射原始图像的亮度直方图；
    基于所述亮度直方图确定最大亮度值和最小亮度值；
    判断所述子映射原始图像中像素的亮度值是否大于所述最小亮度值且小 于最大亮度值；
    如果所述子映射原始图像中像素的亮度值大于所述最小亮度值且小于最大亮度值，则所述子映射原始图像中该像素为正常区。
  7. 根据权利要求6所述的自动检测方法，其中，如果所述子映射原始图像中像素的亮度值不大于所述最小亮度值或不小于最大亮度值，则所述子映射原始图像中该像素为缺陷区。
  8. 根据权利要求6所述的自动检测方法，其中，所述最大亮度值和所述最小亮度值的确定方法包括：
    在所述亮度直方图中确定数量占比不小于预定数量占比的像素的最大亮度初值和最小亮度值初值；
    计算大于所述最大亮度初值的所述子映射原始图像中像素的亮度值的第一加权平均值，且计算小于所述最小亮度值初值的所述子映射原始图像中像素的亮度值的第二加权平均值；
    当所述第一加权平均值与所述最大亮度初值的差值大于一门槛值，且所述最小亮度值初值与所述第二加权平均值的差值大于所述门槛值时，将所述最大亮度初值和所述最小亮度值初值分别作为所述最大亮度值和所述最小亮度值。
  9. 根据权利要求7所述的自动检测方法，其中，所述最大亮度值和所述最小亮度值的确定方法包括：
    在所述亮度直方图中确定数量占比不小于预定数量占比的像素的最大亮度初值和最小亮度值初值；
    计算大于所述最大亮度初值的所述子映射原始图像中像素的亮度值的第一加权平均值，且计算小于所述最小亮度值初值的所述子映射原始图像中像素的亮度值的第二加权平均值；
    当所述第一加权平均值与所述最大亮度初值的差值大于一门槛值，且所述最小亮度值初值与所述第二加权平均值的差值大于所述门槛值时，将所述最大亮度初值和所述最小亮度值初值分别作为所述最大亮度值和所述最小亮度值。
  10. 根据权利要求8所述的自动检测方法，其中，当所述第一加权平均值与所述最大亮度初值的差值不大于所述门槛值和/或所述最小亮度值初值与所述第二加权平均值的差值不大于所述门槛值时，将所述预定数量占比加0.01，且重新确定最大亮度初值和最小亮度初值。
  11. 根据权利要求9所述的自动检测方法，其中，当所述第一加权平均值与所述最大亮度初值的差值不大于所述门槛值和/或所述最小亮度值初值与所述第二加权平均值的差值不大于所述门槛值时，将所述预定数量占比加0.01，且重新确定最大亮度初值和最小亮度初值。
  12. 根据权利要求8所述的自动检测方法，其中，所述最大亮度值和所述最小亮度值的确定方法还包括：
    当所述预定数量占比为1时，将所述最大亮度初值和所述最小亮度初值分别作为所述最大亮度值和所述最小亮度值。
  13. 根据权利要求9所述的自动检测方法，其中，所述最大亮度值和所述最小亮度值的确定方法还包括：
    当所述预定数量占比为1时，将所述最大亮度初值和所述最小亮度初值分别作为所述最大亮度值和所述最小亮度值。
  14. 根据权利要求1所述的自动检测方法，其中，对所述分辨出正常区和缺陷区的映射原始图像进行校正的方法包括：
    计算所述标记拉伸图像中的若干标记点与所述标记图像的若干标记点的若干对应坐标偏差值；
    利用所述若干对应坐标偏差值对所述分辨出正常区和缺陷区的映射原始图像进行校正。
PCT/CN2015/071129 2014-12-16 2015-01-20 显示面板缺陷的自动检测方法 WO2016095318A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US14/425,048 US9646370B2 (en) 2014-12-16 2015-01-20 Automatic detection method for defects of a display panel

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410782236.5 2014-12-16
CN201410782236.5A CN104484878B (zh) 2014-12-16 2014-12-16 显示面板缺陷的自动检测方法

Publications (1)

Publication Number Publication Date
WO2016095318A1 true WO2016095318A1 (zh) 2016-06-23

Family

ID=52759418

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/071129 WO2016095318A1 (zh) 2014-12-16 2015-01-20 显示面板缺陷的自动检测方法

Country Status (3)

Country Link
US (1) US9646370B2 (zh)
CN (1) CN104484878B (zh)
WO (1) WO2016095318A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11650166B2 (en) * 2017-05-31 2023-05-16 Nipro Corporation Method for evaluation of glass container

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104765173B (zh) * 2015-04-29 2018-01-16 京东方科技集团股份有限公司 一种检测装置及其检测方法
KR101668039B1 (ko) * 2015-08-17 2016-10-20 주식회사 홍익기술 디스플레이 패널 점등 검사방법
CN105405137B (zh) * 2015-11-09 2018-10-26 长沙慧联智能科技有限公司 基于机器视觉的轴质量检测方法
WO2018088552A1 (ja) * 2016-11-14 2018-05-17 日本碍子株式会社 セラミックス体の欠陥検査装置および欠陥検査方法
CN106500969B (zh) * 2016-11-17 2019-07-26 深圳Tcl新技术有限公司 显示屏均匀性测试方法及显示屏均匀性测试系统
WO2018170657A1 (zh) * 2017-03-20 2018-09-27 深圳配天智能技术研究院有限公司 一种视觉检测系统及其光场校正方法
CN107179181B (zh) * 2017-06-19 2020-07-14 深圳Tcl新技术有限公司 显示屏均匀性的测试方法、终端及计算机可读存储介质
CN107328791B (zh) * 2017-07-31 2020-06-30 京东方科技集团股份有限公司 一种缺陷检测方法及装置
CN108333805B (zh) * 2018-02-26 2020-11-03 京东方科技集团股份有限公司 不良点坐标自动检测方法和装置、设备和存储介质
CN111712869A (zh) * 2018-09-25 2020-09-25 西安诺瓦星云科技股份有限公司 显示设备故障检测方法、设备、系统及计算机可读介质
CN109799630B (zh) * 2019-03-07 2021-09-03 Tcl华星光电技术有限公司 玻璃基板重测重修系统及其方法
US11961239B2 (en) 2020-12-21 2024-04-16 Seichi Technologies Shenzhen Limited Method and device for marking image position of sub-pixel of display screen, and storage medium
CN112258429B (zh) * 2020-12-21 2021-03-26 深圳精智达技术股份有限公司 显示屏子像素的图像位置标记方法、设备以及存储介质
CN113538603B (zh) * 2021-09-16 2021-12-24 深圳市光明顶照明科技有限公司 一种基于阵列产品的光学检测方法、系统和可读存储介质

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1031730A (ja) * 1996-07-16 1998-02-03 Advantest Corp キャリブレーション方法
JP2003149160A (ja) * 2001-11-13 2003-05-21 Nec Corp 外観検査方法及び外観検査装置
US6597805B1 (en) * 1998-11-27 2003-07-22 Nec Corporation Visual inspection method for electronic device, visual inspecting apparatus for electronic device, and record medium for recording program which causes computer to perform visual inspecting method for electronic device
TW200821572A (en) * 2006-11-15 2008-05-16 Wintek Corp Image detection device with high-positioning precision applied in a flat display panel and method thereof
TW200921087A (en) * 2007-11-06 2009-05-16 Ites Co Ltd Apparatus for determining defect position of panel
CN102202226A (zh) * 2010-03-23 2011-09-28 夏普株式会社 校准装置、缺陷检测装置、缺陷修复装置、校准方法
CN103440654A (zh) * 2013-08-27 2013-12-11 南京大学 一种lcd异物缺陷检测方法
CN103559857A (zh) * 2013-10-31 2014-02-05 桂林机床电器有限公司 一种面向oled屏像素缺陷检测的方法及装置
CN103760165A (zh) * 2013-12-31 2014-04-30 深圳市华星光电技术有限公司 显示面板的缺陷检测方法及缺陷检测装置

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3297950B2 (ja) * 1993-07-13 2002-07-02 シャープ株式会社 平面型表示パネル検査装置
TW351898B (en) * 1995-12-05 1999-02-01 Advantest Corp Portrait processing method
US5754678A (en) * 1996-01-17 1998-05-19 Photon Dynamics, Inc. Substrate inspection apparatus and method
JPH09329527A (ja) * 1996-04-08 1997-12-22 Advantest Corp 画像処理方法および画像処理装置
US6122397A (en) * 1997-07-03 2000-09-19 Tri Path Imaging, Inc. Method and apparatus for maskless semiconductor and liquid crystal display inspection
TWI324326B (en) * 2006-11-03 2010-05-01 Univ Nat Taipei Technology A mura defect detection algorithm for flat panel displays
JP4537467B2 (ja) * 2008-03-18 2010-09-01 アドバンスド・マスク・インスペクション・テクノロジー株式会社 試料検査装置及び試料検査方法
CN202253135U (zh) 2011-08-05 2012-05-30 深圳市华星光电技术有限公司 背光模组及液晶显示器
US20140270347A1 (en) * 2013-03-13 2014-09-18 Sharp Laboratories Of America, Inc. Hierarchical image classification system
KR101565748B1 (ko) * 2013-05-31 2015-11-05 삼성에스디에스 주식회사 이미지에서 반복 패턴을 검출하는 방법 및 장치

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1031730A (ja) * 1996-07-16 1998-02-03 Advantest Corp キャリブレーション方法
US6597805B1 (en) * 1998-11-27 2003-07-22 Nec Corporation Visual inspection method for electronic device, visual inspecting apparatus for electronic device, and record medium for recording program which causes computer to perform visual inspecting method for electronic device
JP2003149160A (ja) * 2001-11-13 2003-05-21 Nec Corp 外観検査方法及び外観検査装置
TW200821572A (en) * 2006-11-15 2008-05-16 Wintek Corp Image detection device with high-positioning precision applied in a flat display panel and method thereof
TW200921087A (en) * 2007-11-06 2009-05-16 Ites Co Ltd Apparatus for determining defect position of panel
CN102202226A (zh) * 2010-03-23 2011-09-28 夏普株式会社 校准装置、缺陷检测装置、缺陷修复装置、校准方法
CN103440654A (zh) * 2013-08-27 2013-12-11 南京大学 一种lcd异物缺陷检测方法
CN103559857A (zh) * 2013-10-31 2014-02-05 桂林机床电器有限公司 一种面向oled屏像素缺陷检测的方法及装置
CN103760165A (zh) * 2013-12-31 2014-04-30 深圳市华星光电技术有限公司 显示面板的缺陷检测方法及缺陷检测装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JIANG, JIAMING ET AL.: "Polymer Light-Emitting Diode Defect Inspection System", 16TH IPPR CONFERENCE ON COMPUTER VISION, GRAPHICS AND IMAGE PROCESSING (CVGIP 2003, 19 August 2003 (2003-08-19), pages 748 - 755 *
PENG, SHAOHU, THE RESEARCH OF THE DEFECTS DETECTION TECHNOLOGY OF THE TFT-LCD, 15 November 2007 (2007-11-15), pages 9, 10, - and 14, ISSN: 1674-0246 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11650166B2 (en) * 2017-05-31 2023-05-16 Nipro Corporation Method for evaluation of glass container

Also Published As

Publication number Publication date
CN104484878A (zh) 2015-04-01
US9646370B2 (en) 2017-05-09
CN104484878B (zh) 2017-10-17
US20160343121A1 (en) 2016-11-24

Similar Documents

Publication Publication Date Title
WO2016095318A1 (zh) 显示面板缺陷的自动检测方法
CN108776140B (zh) 一种基于机器视觉的印刷品瑕疵检测方法及系统
US11120565B2 (en) Image registration method, image registration device and storage medium
US11562480B2 (en) System and method for set up of production line inspection
CN103885217B (zh) 一种检测液晶面板柱状隔垫物缺陷的方法及装置
WO2016106852A1 (zh) 一种显示面板的缺陷的灰阶补偿方法及灰阶补偿系统
JP5659623B2 (ja) 露出属性の設定方法およびコンピューター読み取り可能な記憶媒体
CN112033965B (zh) 基于差分图像分析的3d弧形表面缺陷检测方法
CN109471276B (zh) 一种液晶屏色偏缺陷检测方法及装置
JP2012032369A (ja) 欠陥識別方法、欠陥識別装置、プログラム、及び記録媒体
US20220020136A1 (en) Optimizing a set-up stage in an automatic visual inspection process
CN105069453A (zh) 一种图像校正方法及装置
JPWO2018179559A1 (ja) 航空機用部品の欠陥検出システム及び航空機用部品の欠陥検出方法
WO2020079694A1 (en) Optimizing defect detection in an automatic visual inspection process
WO2023109557A1 (zh) 连接器件检测方法、电子设备及存储介质
CN114299070A (zh) 一种显示屏mura缺陷的检测方法及相关装置
CN111563867A (zh) 一种用于提升图像清晰度的图像融合方法
TW201512649A (zh) 偵測晶片影像瑕疵方法及其系統與電腦程式產品
US20220335585A1 (en) Set up of a visual inspection process
CN113160223A (zh) 轮廓的确定方法、确定装置、检测设备及存储介质
US20150294452A1 (en) Image processing method, image processing device and automated optical inspection machine
CN110288662B (zh) 显示器检测方法与系统
CA2939304A1 (en) A pixel-based universal image info extraction models and process flow
US20220148152A1 (en) System and method for adjustable production line inspection
US20220044379A1 (en) Streamlining an automatic visual inspection process

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 14425048

Country of ref document: US

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15868835

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15868835

Country of ref document: EP

Kind code of ref document: A1