WO2019127708A1 - 基于霍夫变换及高斯拟合的面内mura检测方法及检测系统 - Google Patents

基于霍夫变换及高斯拟合的面内mura检测方法及检测系统 Download PDF

Info

Publication number
WO2019127708A1
WO2019127708A1 PCT/CN2018/073057 CN2018073057W WO2019127708A1 WO 2019127708 A1 WO2019127708 A1 WO 2019127708A1 CN 2018073057 W CN2018073057 W CN 2018073057W WO 2019127708 A1 WO2019127708 A1 WO 2019127708A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
fitting
grayscale image
mura
plane
Prior art date
Application number
PCT/CN2018/073057
Other languages
English (en)
French (fr)
Inventor
王艳雪
Original Assignee
惠州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 惠州市华星光电技术有限公司 filed Critical 惠州市华星光电技术有限公司
Priority to US15/974,382 priority Critical patent/US10740889B2/en
Publication of WO2019127708A1 publication Critical patent/WO2019127708A1/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
    • G06T7/0006—Industrial image inspection using a design-rule based 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
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T7/41—Analysis of texture based on statistical description of texture
    • G06T7/44—Analysis of texture based on statistical description of texture using image operators, e.g. filters, edge density metrics or local histograms
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • 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 relates to the field of image processing technologies, and in particular, to an in-plane mura detection method and detection system based on Hough transform and Gaussian fitting.
  • Display devices may have in-plane unevenness due to their process or driving factors, which may seriously affect the visual effect, that is, the in-plane mura phenomenon.
  • Mura refers to the phenomenon that the brightness of the display is uneven, causing various traces.
  • mura There are many types of mura, and this kind of in-plane mura is generally in the center of the display. Due to its low contrast and irregular shape, the detection has certain difficulty, which in turn affects the opposite side. Detection of the inner mura.
  • an object of the present invention is to provide an in-plane mura detection method and detection system based on Hough transform and Gaussian fitting to realize in-plane mura detection.
  • An in-plane mura detection method based on Hough transform and Gaussian fitting comprising:
  • the method further includes:
  • step S2 is specifically:
  • a polynomial fit is performed on the histogram of the grayscale image, and a threshold demarcation point is obtained according to the minimum value of the fitting curve, thereby obtaining a binarized image.
  • step S3 is specifically:
  • the edge detection is performed by Hough transform, and the starting point of the line in the binarized image is obtained, and the display area is determined according to the starting point of the line and the number of pixels, and then the edge of the grayscale image is cropped.
  • the Gaussian function is a, b, c are free parameters
  • step S4 “the degree of detecting mura in the surface according to the fitting parameter” is specifically as follows:
  • the parameter a is greater than or equal to the first preset threshold a 0 and the parameter c is less than or equal to the second preset threshold c 0 , it is determined that the in-plane mura is slight; otherwise, it is determined that the in-plane mura is severe.
  • An in-plane mura detection system based on Hough transform and Gaussian fitting comprising:
  • An image obtaining unit configured to acquire an original grayscale image
  • An image processing unit configured to acquire a binarized image according to the grayscale image
  • a Hough transform unit for performing edge detection by Hough transform, and then trimming an edge of the grayscale image
  • a Gaussian fitting unit is used for performing histogram statistics on the trimmed grayscale image, fitting the histogram to a Gaussian function, and detecting the extent of the in-plane mura according to the fitting parameter.
  • the detection system further comprises:
  • the low pass filtering unit comprising a Butterworth filter for low pass filtering the gray scale image to eliminate noise in the gray scale image.
  • the image processing unit is further configured to:
  • a polynomial fit is performed on the histogram of the grayscale image, and a threshold demarcation point is obtained according to the minimum value of the fitting curve, thereby obtaining a binarized image.
  • the Hough transform unit is further used for:
  • the edge detection is performed by Hough transform, and the starting point of the line in the binarized image is obtained, and the display area is determined according to the starting point of the line and the number of pixels, and then the edge of the grayscale image is cropped.
  • the Gaussian fitting unit is further used for:
  • the Gaussian function is a, b, and c are free parameters. If the parameter a is greater than or equal to the first preset threshold a 0 and the parameter c is less than or equal to the second preset threshold c 0 , it is determined that the in-plane mura is slight; otherwise, the determination is In-plane mura is serious.
  • the invention determines the display area image by Hough transform, thereby obtaining the in-plane mura detection area, and using the Gaussian function fitting, the parameters obtained by the fitting are used to evaluate the severity of the in-plane mura, and the in-plane mura is quickly detected.
  • FIG. 1 is a schematic flow chart of a method for detecting in-plane mura in the present invention.
  • FIG. 2 is a schematic block diagram of an in-plane mura detection system of the present invention.
  • FIG. 3 is a grayscale image acquired in Embodiment 1 of the present invention.
  • Embodiment 4 is a histogram of a grayscale image after cropping in Embodiment 1 of the present invention.
  • FIG. 5 is a grayscale image acquired in Embodiment 2 of the present invention.
  • FIG. 6 is a histogram of a grayscale image after cropping according to Embodiment 2 of the present invention.
  • the present invention discloses an in-plane mura detection method based on Hough transform and Gaussian fitting, including:
  • step S2 the method further includes:
  • the camera When the original grayscale image is taken in the darkroom, the camera is an electronic device, and noise is inevitably introduced. According to the characteristics of the noise, the Butterworth filter is selected for low-pass filtering.
  • Step S2 is specifically:
  • a polynomial fit is performed on the histogram of the grayscale image, and a threshold demarcation point is obtained according to the minimum value of the fitting curve, thereby obtaining a binarized image.
  • the camera Since the camera will have a non-display screen, it is necessary to automatically detect the portion of the display area to eliminate the influence of the non-display on the calculation.
  • Performing a polynomial fitting on the histogram of the gray-scale image obtaining a threshold demarcation point according to the minimum value of the fitting curve, and setting the gray value of the boundary point larger than the threshold value to 255, and setting the gray value of the demarcation point smaller than the threshold value to 0, thereby A binarized image is obtained, and the binarized image exhibits a distinct black and white visual effect.
  • step S3 edge detection is performed by Hough transform, the starting point of the line in the binarized image is obtained, and the display area is determined according to the starting point of the line and the number of pixels, and then the edge of the grayscale image is cropped.
  • the edge portion is trimmed when the image is cropped, and the middle of the large area is retained. Display the image.
  • step S4 the Gaussian function is a, b, c are free parameters
  • the parameter a is greater than or equal to the first preset threshold a 0 and the parameter c is less than or equal to the second preset threshold c 0 , it is determined that the in-plane mura is slight; otherwise, it is determined that the in-plane mura is severe.
  • Histogram statistics are performed on the grayscale image after clipping. If the image surface mura is slight, the distribution of the histogram will be concentrated and the peak value will be high. If the mura is severe, the distribution of the histogram will be more dispersed and the peak value will be lower.
  • the histogram is fitted with a Gaussian function, and the fitting parameters a and c are used to describe the in-plane mura.
  • the present invention also discloses an in-plane mura detection system based on Hough transform and Gaussian fitting, including:
  • An image obtaining unit 10 configured to acquire an original grayscale image
  • An image processing unit 20 configured to acquire a binarized image according to the grayscale image
  • the Hough transform unit 30 is configured to perform edge detection by Hough transform, and then trim the edge of the grayscale image
  • the Gaussian fitting unit 40 is configured to perform histogram statistics on the trimmed grayscale image, perform a Gaussian function fitting on the histogram, and detect the extent of the in-plane mura according to the fitting parameter.
  • the detection system of the present invention further includes:
  • the low-pass filtering unit 50 and the low-pass filtering unit select a Butterworth filter for low-pass filtering the gray-scale image to eliminate noise in the gray-scale image.
  • Low-pass filtering of grayscale images is used to eliminate noise in grayscale images
  • Edge detection is performed by Hough transform, the starting point of the line in the binarized image is obtained, and the display area is determined according to the starting point of the line and the number of pixels, and then the edge of the grayscale image is trimmed;
  • the histogram statistics are performed on the gray scale image after the cropping, the Gaussian function is fitted to the histogram, and the degree of mura in the plane is detected according to the fitting parameter, and the Gaussian function is a, b, and c are free parameters.
  • the inside mura is slight.
  • the first preset threshold a 0 and the second preset threshold c 0 are related to the display status and specifications.
  • the threshold in this embodiment is only an exemplary threshold, and may be set to other values in other embodiments.
  • Low-pass filtering of grayscale images is used to eliminate noise in grayscale images
  • Edge detection is performed by Hough transform, the starting point of the line in the binarized image is obtained, and the display area is determined according to the starting point of the line and the number of pixels, and then the edge of the grayscale image is trimmed;
  • the histogram statistics are performed on the gray scale image after the cropping, the Gaussian function is fitted to the histogram, and the degree of mura in the plane is detected according to the fitting parameter, and the Gaussian function is a, b, and c are free parameters.
  • the first preset threshold a 0 and the second preset threshold c 0 are related to the display status and specifications.
  • the threshold in this embodiment is only an exemplary threshold, and may be set to other values in other embodiments.
  • An embodiment of the present invention further provides an electronic device.
  • the electronic device includes at least one processor and a memory coupled to the at least one processor, the memory for storing instructions executable by the at least one processor, the instructions being executed by the at least one processor
  • the at least one processor is caused to perform the in-plane mura detection method in the above embodiment.
  • the embodiment of the invention further provides a non-transitory storage medium storing computer executable instructions, the computer executable instructions being configured to perform the in-plane mura detection method described above.
  • Embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, when the program instructions are executed by a computer
  • the computer is caused to perform the above-described in-plane mura detection method.
  • the in-plane mura detection system provided by the embodiment of the present invention can perform the in-plane mura detection method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
  • the in-plane mura detection method provided by any embodiment of the present invention can perform the in-plane mura detection method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
  • the present invention determines the display area image by Hough transform, thereby obtaining the in-plane mura detection area, using a Gaussian function fitting, and estimating the severity of the in-plane mura by fitting the parameters.
  • the rapid detection of in-plane mura is the rapid detection of in-plane mura.
  • Any process or method description in the flowchart of the present application or otherwise described herein may be understood as a module representing code comprising one or more executable instructions for implementing the steps of a particular logical function or process, Fragments or portions, and the scope of the embodiments of the invention includes additional implementations, in which the functions may be performed in a substantially simultaneous manner or in the reverse order, depending on the order in which they are illustrated. This should be understood by those skilled in the art to which the embodiments of the present invention pertain.
  • a "computer readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with such an instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (mobile terminals) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.
  • portions of the invention may be implemented in hardware, software, firmware or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals. Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.
  • each functional unit in each embodiment of the present invention may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

本发明公开了一种基于霍夫变换及高斯拟合的面内mura检测方法及检测系统，检测方法包括：S1、获取原始灰阶图像；S2、根据灰阶图像获取二值化图像；S3、通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；S4、对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。本发明通过霍夫变换确定显示区图像，从而获得面内mura检测区域，使用高斯函数拟合，通过拟合而来的参数，评价面内mura的严重程度，实现了面内mura的快速检测。

Description

基于霍夫变换及高斯拟合的面内mura检测方法及检测系统 技术领域
本发明涉及图像处理技术领域，特别是涉及一种基于霍夫变换及高斯拟合的面内mura检测方法及检测系统。
背景技术
显示器件(例如TFT-LCD等)由于其制程或驱动因素，会产生面内不均，严重的会影响视觉效果，即面内mura现象。mura是指显示器亮度不均匀造成各种痕迹的现象，mura的种类很多，而这种面内mura一般在显示器中心区域，由于其低对比度及形状的不规则，检测存在一定难度，进而影响到对面内mura的检测。
因此，针对上述技术问题，有必要提供一种基于霍夫变换及高斯拟合的面内mura检测方法及检测系统。
发明内容
为克服现有技术的不足，本发明的目的在于提供一种基于霍夫变换及高斯拟合的面内mura检测方法及检测系统，以实现面内mura的检测。
为了实现上述目的，本发明一实施例提供的技术方案如下：
一种基于霍夫变换及高斯拟合的面内mura检测方法，所述检测方法包括：
S1、获取原始灰阶图像；
S2、根据灰阶图像获取二值化图像；
S3、通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
S4、对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。
作为本发明的进一步改进，所述步骤S2前还包括：
对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
作为本发明的进一步改进，所述步骤S2具体为：
对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像。
作为本发明的进一步改进，所述步骤S3具体为：
通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
作为本发明的进一步改进，所述高斯函数为
Figure PCTCN2018073057-appb-000001
a、b、c为自由参数；
步骤S4中“根据拟合参数检测面内mura的程度”具体为：
若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
本发明另一实施例提供的技术方案如下：
一种基于霍夫变换及高斯拟合的面内mura检测系统，所述检测系统包括：
图像获取单元，用于获取原始灰阶图像；
图像处理单元，用于根据灰阶图像获取二值化图像；
霍夫变换单元，用于通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
高斯拟合单元，用于对剪裁后的灰阶图像进行直方图统计，将直方图进行 高斯函数拟合，根据拟合参数检测面内mura的程度。
作为本发明的进一步改进，所述检测系统还包括：
低通滤波单元，所述低通滤波单元包括巴特沃斯滤波器，用于对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
作为本发明的进一步改进，所述图像处理单元还用于：
对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像。
作为本发明的进一步改进，所述霍夫变换单元还用于：
通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
作为本发明的进一步改进，所述高斯拟合单元还用于：
对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，高斯函数为
Figure PCTCN2018073057-appb-000002
a、b、c为自由参数，若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
本发明通过霍夫变换确定显示区图像，从而获得面内mura检测区域，使用高斯函数拟合，通过拟合而来的参数，评价面内mura的严重程度，实现了面内mura的快速检测。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术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发明中记载的一些实施例，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为本发明中面内mura检测方法的流程示意图。
图2为本发明中面内mura检测系统的模块示意图。
图3为本发明实施例1中获取的灰阶图像。
图4为本发明实施例1中剪裁后灰阶图像的直方图。
图5为本发明实施例2中获取的灰阶图像。
图6为本发明实施例2中剪裁后灰阶图像的直方图。
具体实施方式
为了使本技术领域的人员更好地理解本发明中的技术方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
参图1所示，本发明公开了一种基于霍夫变换及高斯拟合的面内mura检测方法，包括：
S1、获取原始灰阶图像；
S2、根据灰阶图像获取二值化图像；
S3、通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
S4、对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。
在本发明一实施例中，步骤S2前还包括：
对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
在暗房内拍摄原始灰阶图像时，因拍摄用的相机为电子器件，不可避免的会有噪声引入，根据噪声的特点，选择巴特沃斯滤波器进行低通滤波。
步骤S2具体为：
对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像。
因相机拍摄会有非显示器的画面存在，故需要能自动检测出显示区域的部分，排除非显示器对计算的影响。对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，大于阈值分界点的灰度值设为255，小于阈值分界点的灰度值设为0，从而得到二值化图像，二值化图像呈现出明显的只有黑和白的视觉效果。
步骤S3中，通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
因在显示器图像的边缘部分，面内mura存在的概率较低，且对人眼视觉观察上影响较小，为了快速计算，则在图像剪裁时，边缘部分会被剪裁掉，保留大面积的中间显示图像。
步骤S4中，高斯函数为
Figure PCTCN2018073057-appb-000003
a、b、c为自由参数；
“根据拟合参数检测面内mura的程度”具体为：
若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
对剪裁后的灰阶图像进行直方图统计，若图像面mura轻微，则直方图的分布会较为集中，峰值较高；若面内mura严重，直方图的分布会较为分散，峰值变低。本发明中将直方图进行高斯函数拟合，使用拟合参数a与c进行面内mura的描述。
参图4所示，本发明还公开了一种基于霍夫变换及高斯拟合的面内mura检测系统，包括：
图像获取单元10，用于获取原始灰阶图像；
图像处理单元20，用于根据灰阶图像获取二值化图像；
霍夫变换单元30，用于通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
高斯拟合单元40，用于对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。
另外，本发明的检测系统中还包括：
低通滤波单元50，低通滤波单元选用巴特沃斯滤波器，用于对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
以下结合具体实施例对本发明作进一步说明。
实施例1：
本实施例中的面内mura检测方法，包括：
获取得到参图3所示的灰阶图像；
采用对灰阶图像进行低通滤波，消除灰阶图像中的噪声；
对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像；
通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁；
对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度，高斯函数为
Figure PCTCN2018073057-appb-000004
a、b、c为自由参数。
本实施例中的剪裁后灰阶图像的直方图参图4所示，根据该直方图分布拟合得到相应的高斯函数，得到a＝2.08e+05、c＝6.559、R 2＝0.964，其中，R 2表示直方图与高斯拟合曲线的匹配度，数值越接近1，表示匹配程度越高，可见，本实施例中的直方图于高斯拟合曲线具有较高的匹配度。
比较得到参数a＝2.08e+05大于第一预设阈值a 0(例如1.5e+05)，参数c＝6.559小于第二预设阈值c 0(例如8)，则判定本实施例中为面内mura轻微。
第一预设阈值a 0和第二预设阈值c 0与显示器状况及规格相关，本实施例中的阈值仅为示例性的阈值，在其他实施例中可以设置为其他数值。
实施例2：
本实施例中的面内mura检测方法，包括：
获取得到参图5所示的灰阶图像；
采用对灰阶图像进行低通滤波，消除灰阶图像中的噪声；
对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像；
通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁；
对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度，高斯函数为
Figure PCTCN2018073057-appb-000005
a、b、c为自由参数。
本实施例中的剪裁后灰阶图像的直方图参图6所示，根据该直方图分布拟合得到相应的高斯函数，得到a＝1.07e+05、c＝13.83、R 2＝0.967，其中，R 2表示直方图与高斯拟合曲线的匹配度，数值越接近1，表示匹配程度越高，可见，本实施例中的直方图于高斯拟合曲线具有较高的匹配度。
比较得到参数a＝1.07e+05小于第一预设阈值a 0(例如1.5e+05)，参数c＝13.83小于第二预设阈值c 0(例如8)，则判定本实施例中为面内mura严重。
第一预设阈值a 0和第二预设阈值c 0与显示器状况及规格相关，本实施例中的阈值仅为示例性的阈值，在其他实施例中可以设置为其他数值。
本发明实施例还提供一种电子设备。所述电子设备包括至少一个处理器和与所述至少一个处理器连接的存储器，所述存储器用于存储可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行时，使所述至少一个处理器执行上述实施例中的面内mura检测方法。
本发明实施例还提供了一种非暂态存储介质，存储有计算机可执行指令，所述计算机可执行指令设置为执行上述的面内mura检测方法。
本发明实施例还提供了一种计算机程序产品，所述计算机程序产品包括存储在非暂态计算机可读存储介质上的计算机程序，所述计算机程序包括程序指令，当所述程序指令被计算机执行时，使所述计算机执行上述的面内mura检测方法。
本发明实施例提供的面内mura检测系统可执行本发明任意实施例所提供的面内mura检测方法，具备执行方法相应的功能模块和有益效果。未在上述实施例中详尽描述的技术细节，可参见本发明任意实施例所提供的面内mura检测方法。
由以上实施方式可以看出，本发明通过霍夫变换确定显示区图像，从而获得面内mura检测区域，使用高斯函数拟合，通过拟合而来的参数，评价面内mura的严重程度，实现了面内mura的快速检测。
本申请的流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指 令的代码的模块、片段或部分，并且本发明的实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本发明的实施例所属技术领域的技术人员所理解。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，″计算机可读介质″可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(移动终端)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合 逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施例方法携带的全部或部分步骤是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，该程序在执行时，包括方法实施例的步骤之一或其组合。此外，在本发明各个实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本发明的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本发明的限制，本领域的普通技术人员在本发明的范围内可以对上述实施例进行变化、修改、替换和变型。
对于本领域技术人员而言，显然本发明不限于上述示范性实施例的细节，而且在不背离本发明的精神或基本特征的情况下，能够以其他的具体形式实现本发明。因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本发明的范围由所附权利要求而不是上述说明限定，因此旨在将落在权利要求的等同要件的含义和范围内的所有变化囊括在本发明内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。
此外，应当理解，虽然本说明书按照实施方式加以描述，但并非每个实施方式仅包含一个独立的技术方案，说明书的这种叙述方式仅仅是为清楚起见，本领域技术人员应当将说明书作为一个整体，各实施例中的技术方案也可以经适当组合，形成本领域技术人员可以理解的其他实施方式。

Claims (16)

  1. 一种基于霍夫变换及高斯拟合的面内mura检测方法，其中，所述检测方法包括：
    S1、获取原始灰阶图像；
    S2、根据灰阶图像获取二值化图像；
    S3、通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
    S4、对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。
  2. 根据权利要求1所述的检测方法，其中，所述步骤S2前还包括：
    对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
  3. 根据权利要求1所述的检测方法，其中，所述步骤S2具体为：
    对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像。
  4. 根据权利要求1所述的检测方法，其中，所述步骤S3具体为：
    通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
  5. 根据权利要求4所述的检测方法，其中，所述高斯函数为
    Figure PCTCN2018073057-appb-100001
    Figure PCTCN2018073057-appb-100002
    a、b、c为自由参数；
    步骤S4中“根据拟合参数检测面内mura的程度”具体为：
    若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
  6. 根据权利要求2所述的检测方法，其中，所述步骤S2具体为：
    对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分 界点，从而得到二值化图像。
  7. 根据权利要求6所述的检测方法，其中，所述步骤S3具体为：
    通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
  8. 根据权利要求7所述的检测方法，其中，所述高斯函数为
    Figure PCTCN2018073057-appb-100003
    Figure PCTCN2018073057-appb-100004
    a、b、c为自由参数；
    步骤S4中“根据拟合参数检测面内mura的程度”具体为：
    若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
  9. 一种基于霍夫变换及高斯拟合的面内mura检测方法，其中，所述检测方法包括：
    S1、获取原始灰阶图像；
    S2、根据灰阶图像获取二值化图像，具体为：
    对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像；
    S3、通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
    S4、对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度；
    所述步骤S2前还包括：
    对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
  10. 一种基于霍夫变换及高斯拟合的面内mura检测系统，其中，所述检测系统包括：
    图像获取单元，用于获取原始灰阶图像；
    图像处理单元，用于根据灰阶图像获取二值化图像；
    霍夫变换单元，用于通过霍夫变换进行边缘检测，进而对灰阶图像的边缘进行剪裁；
    高斯拟合单元，用于对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，根据拟合参数检测面内mura的程度。
  11. 根据权利要求10所述的检测系统，其中，所述检测系统还包括：
    低通滤波单元，所述低通滤波单元包括巴特沃斯滤波器，用于对灰阶图像进行低通滤波，消除灰阶图像中的噪声。
  12. 根据权利要求10所述的检测系统，其中，所述图像处理单元还用于：
    对灰阶图像的直方图进行多项式拟合，根据拟合曲线极小值点得到阈值分界点，从而得到二值化图像。
  13. 根据权利要求10所述的检测系统，其中，所述霍夫变换单元还用于：
    通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
  14. 根据权利要求13所述的检测系统，其中，所述高斯拟合单元还用于：
    对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，高斯函数为
    Figure PCTCN2018073057-appb-100005
    a、b、c为自由参数，若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
  15. 根据权利要求12所述的检测系统，其中，所述霍夫变换单元还用于：
    通过霍夫变换进行边缘检测，获取二值化图像中的直线起始点，根据直线起始点及像素数确定显示区域，进而对灰阶图像的边缘进行剪裁。
  16. 根据权利要求15所述的检测系统，其中，所述高斯拟合单元还用于：
    对剪裁后的灰阶图像进行直方图统计，将直方图进行高斯函数拟合，高斯函数为
    Figure PCTCN2018073057-appb-100006
    a、b、c为自由参数，若检测到参数a大于或等于第一预设阈值a 0，且参数c小于或等于第二预设阈值c 0，则判定面内mura轻微；否则，则判定面内mura严重。
PCT/CN2018/073057 2017-12-29 2018-01-17 基于霍夫变换及高斯拟合的面内mura检测方法及检测系统 WO2019127708A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/974,382 US10740889B2 (en) 2017-12-29 2018-05-08 Method and system for detection of in-panel mura based on hough transform and gaussian fitting

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711479755.4A CN108074238B (zh) 2017-12-29 2017-12-29 基于霍夫变换及高斯拟合的面内mura检测方法及检测系统
CN201711479755.4 2017-12-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/974,382 Continuation US10740889B2 (en) 2017-12-29 2018-05-08 Method and system for detection of in-panel mura based on hough transform and gaussian fitting

Publications (1)

Publication Number Publication Date
WO2019127708A1 true WO2019127708A1 (zh) 2019-07-04

Family

ID=62156387

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/073057 WO2019127708A1 (zh) 2017-12-29 2018-01-17 基于霍夫变换及高斯拟合的面内mura检测方法及检测系统

Country Status (2)

Country Link
CN (1) CN108074238B (zh)
WO (1) WO2019127708A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111145130B (zh) * 2019-12-06 2023-05-30 Oppo广东移动通信有限公司 一种图像处理方法及装置、存储介质
CN111290582B (zh) * 2020-02-29 2021-09-21 华南理工大学 一种基于改进型直线检测的投影交互区域定位方法
CN112950495B (zh) * 2021-02-05 2024-05-14 苏州华兴源创科技股份有限公司 一种显示面板demura补偿方法、装置、设备及介质
CN114266719B (zh) * 2021-10-22 2022-11-25 广州辰创科技发展有限公司 一种基于霍夫变换的产品检测方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103413288A (zh) * 2013-08-27 2013-11-27 南京大学 一种lcd总体检测缺陷方法
WO2016149877A1 (zh) * 2015-03-20 2016-09-29 华为技术有限公司 一种校正显示屏不均匀的方法、装置及系统
CN106650770A (zh) * 2016-09-29 2017-05-10 南京大学 一种基于样本学习和人眼视觉特性的mura缺陷检测方法
CN107194919A (zh) * 2017-05-18 2017-09-22 南京大学 基于规律纹理背景重建的手机屏幕缺陷检测方法
CN107316303A (zh) * 2017-08-25 2017-11-03 深圳市华星光电半导体显示技术有限公司 显示面板Mura区域侦测方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9117440B2 (en) * 2011-05-19 2015-08-25 Dolby International Ab Method, apparatus, and medium for detecting frequency extension coding in the coding history of an audio signal
JP5701181B2 (ja) * 2011-08-18 2015-04-15 株式会社Pfu 画像処理装置、画像処理方法及びコンピュータプログラム
KR101958634B1 (ko) * 2012-12-13 2019-03-15 엘지디스플레이 주식회사 디스플레이 장치의 무라 검출 장치 및 방법
CN103440654B (zh) * 2013-08-27 2016-08-10 南京大学 一种lcd异物缺陷检测方法
CN103792699A (zh) * 2013-09-09 2014-05-14 中华人民共和国四川出入境检验检疫局 基于B样条曲面拟合的TFT-LCD Mura缺陷机器视觉检测方法
US10290092B2 (en) * 2014-05-15 2019-05-14 Applied Materials Israel, Ltd System, a method and a computer program product for fitting based defect detection
KR101635461B1 (ko) * 2014-11-05 2016-07-04 한밭대학교 산학협력단 웨이블릿 변환에서 마스크 필터링을 이용한 얼룩 결함 자동 검출 시스템 및 방법
CN106485702B (zh) * 2016-09-30 2019-11-05 杭州电子科技大学 基于自然图像特征统计的图像模糊检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103413288A (zh) * 2013-08-27 2013-11-27 南京大学 一种lcd总体检测缺陷方法
WO2016149877A1 (zh) * 2015-03-20 2016-09-29 华为技术有限公司 一种校正显示屏不均匀的方法、装置及系统
CN106650770A (zh) * 2016-09-29 2017-05-10 南京大学 一种基于样本学习和人眼视觉特性的mura缺陷检测方法
CN107194919A (zh) * 2017-05-18 2017-09-22 南京大学 基于规律纹理背景重建的手机屏幕缺陷检测方法
CN107316303A (zh) * 2017-08-25 2017-11-03 深圳市华星光电半导体显示技术有限公司 显示面板Mura区域侦测方法

Also Published As

Publication number Publication date
CN108074238A (zh) 2018-05-25
CN108074238B (zh) 2020-07-24

Similar Documents

Publication Publication Date Title
US10740889B2 (en) Method and system for detection of in-panel mura based on hough transform and gaussian fitting
WO2019127708A1 (zh) 基于霍夫变换及高斯拟合的面内mura检测方法及检测系统
US20200334793A1 (en) Method and device for blurring image background, storage medium and electronic apparatus
US20190197735A1 (en) Method and apparatus for image processing, and robot using the same
US20230377158A1 (en) Image segmentation method, apparatus, device, and medium
WO2019223069A1 (zh) 基于直方图的虹膜图像增强方法、装置、设备及存储介质
US8787690B2 (en) Binarizing an image
WO2016206087A1 (zh) 一种低照度图像处理方法和装置
CN106920233B (zh) 基于图像处理的划痕检测方法、装置及电子设备
US9576210B1 (en) Sharpness-based frame selection for OCR
CN107123124B (zh) 视网膜图像分析方法、装置和计算设备
CN108830873A (zh) 深度图像物体边缘提取方法、装置、介质及计算机设备
CN111369523B (zh) 显微图像中细胞堆叠的检测方法、系统、设备及介质
CN105096330A (zh) 一种自动识别纯色边框的图像处理方法、系统及拍摄终端
CN109829510A (zh) 一种产品品质分级的方法、装置和设备
CN104504703A (zh) 一种基于片式元器件smt焊点彩色图像分割方法
CN116309562B (zh) 一种板卡缺陷识别方法及系统
Kumari et al. Real time visibility enhancement for single image haze removal
CN114298985B (zh) 缺陷检测方法、装置、设备及存储介质
CN117197527A (zh) 一种液晶显示屏的玻璃基板的缺陷检测分类方法及装置
Yang et al. Automatic TFT-LCD mura detection based on image reconstruction and processing
CN117372415A (zh) 喉镜图像识别方法、装置、计算机设备及存储介质
US20170140217A1 (en) System and method for quantifying reflection e.g. when analyzing laminated documents
US20110158516A1 (en) Image classification methods and systems
CN111311610A (zh) 图像分割的方法及终端设备

Legal Events

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

Ref document number: 18895527

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: 18895527

Country of ref document: EP

Kind code of ref document: A1