CN111325707A - Image processing method and system, and detection method and system - Google Patents
Image processing method and system, and detection method and system Download PDFInfo
- Publication number
- CN111325707A CN111325707A CN201811526089.XA CN201811526089A CN111325707A CN 111325707 A CN111325707 A CN 111325707A CN 201811526089 A CN201811526089 A CN 201811526089A CN 111325707 A CN111325707 A CN 111325707A
- Authority
- CN
- China
- Prior art keywords
- image
- period
- detection
- units
- average value
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
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
- 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
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- 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
-
- 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
- G06T2207/30148—Semiconductor; IC; Wafer
Abstract
The invention provides an image processing method and system, a detection method and a detection system, which comprise the following steps: providing an image, the image having a plurality of repeating structural units; carrying out periodic processing on an image, dividing the image into a plurality of periodic units, wherein each periodic unit comprises a structural unit; dividing each period unit into a plurality of detection units and forming a plurality of intermediate images; acquiring detection areas of the intermediate images, and marking the detection areas; and splicing the intermediate images into a target image, and obtaining a target area according to the mark in the target image. The intermediate images are composed of the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different, so that the intermediate images can be processed in parallel to obtain the detection areas of the intermediate images, and the image processing speed and the detection efficiency can be effectively improved.
Description
Technical Field
The present invention relates to the field of optical detection technologies, and in particular, to an image processing method and system, and a detection method and system.
Background
With the development of semiconductor technology, mobile terminal devices such as mobile phones and tablet computers have become an indispensable part of daily work and life of people.
The glass panel as an important component of the display screen inevitably has some defects such as poor size, scratches, foreign matters and the like in the production, processing and transportation processes, and the defects not only affect the quality and the use value of the display screen, but also affect the acceptance of consumers to the electronic product brand and cause immeasurable loss to enterprises. Based on this, the glass panel is subjected to online detection by adopting a strict detection procedure, and the glass panel with surface defects is screened from a plurality of samples, which is a crucial step in the production process of the glass panel.
The method comprises two processes of image acquisition and image processing, wherein in the image acquisition stage, a machine vision product is adopted to acquire an image of a glass panel, in the image processing stage, the acquired image is divided into a plurality of small units, and each small unit is compared with a corresponding standard image small unit respectively to obtain the defects on the surface of the glass panel. However, this method has a low image processing speed, resulting in a low detection speed, which affects the production efficiency of the glass panel.
Disclosure of Invention
In view of the above, the present invention provides an image processing method and system, and a detection method and system, so as to improve the image processing speed and the detection speed.
In order to achieve the purpose, the invention provides the following technical scheme:
an image processing method comprising:
providing an image having a plurality of repeating structural units;
carrying out periodic processing on the image, dividing the image into a plurality of periodic units, wherein each periodic unit comprises one structural unit;
dividing each period unit into a plurality of detection units, and forming a plurality of intermediate images, wherein the intermediate images are formed by the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different;
acquiring a detection area of each intermediate image, and marking the detection area;
and splicing the intermediate images into a target image, and obtaining a target area according to the mark in the target image.
Optionally, performing a periodic process on the image, and dividing the image into a plurality of periodic units, includes:
acquiring an average value of the gray scale of the image in a first side direction and an average value of the gray scale of the image in a second side direction, wherein the first side is perpendicular to the second side;
performing fourier transform on the average value of the gray scale in the direction of the first side to obtain a first frequency under a peak value, obtaining a first period according to the first frequency and the length of the first side, performing fourier transform on the average value of the gray scale in the direction of the second side to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side;
dividing the image into a plurality of period units according to the first period and the second period.
Optionally, performing a periodic process on the image, and dividing the image into a plurality of periodic units, includes:
acquiring an average value of the gray scale of any one or more columns of pixels of the image in a first side direction and an average value of the gray scale of any one or more rows of pixels of the image in a second side direction, wherein the first side is vertical to the second side, the plurality of columns of pixels are sequentially arranged along the first side direction, and the plurality of rows of pixels are sequentially arranged along the second side direction;
performing fourier transform on an average value of the gray scale of the any one or more columns of pixels in the direction of the first side to obtain a first frequency under a peak value, obtaining a first period according to the first frequency and the length of the first side, performing fourier transform on an average value of the gray scale of the any one or more rows of pixels in the direction of the second side to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side;
dividing the image into a plurality of period units according to the first period and the second period.
Optionally, acquiring the detection region of each intermediate image includes:
and acquiring the detection area of each intermediate image by adopting a threshold segmentation method.
Optionally, the obtaining the detection area of each intermediate image by using a threshold segmentation method includes:
setting a gray threshold;
and acquiring the detection area of each intermediate image according to the gray threshold.
Optionally, setting the grayscale threshold includes:
setting the grayscale threshold to M + -n × sigma;
wherein M is an average value of the gray values of the intermediate image, n is an arbitrary coefficient, and σ is a mean square error of the gray values of the intermediate image.
A method of detection, comprising:
providing an object to be tested, wherein the object to be tested comprises a plurality of repeated periodic structures;
acquiring an image of the object to be detected;
processing the image according to the image processing method as described in any one of the above;
and acquiring the detection information of the object to be detected according to the target area.
Optionally, the detection region is a region of a defect on the surface of the object to be detected in the image; or, the detection area is an area of a specific structure on the surface of the object to be detected in the image.
An image processing system comprises an input module, a first processing module, a second processing module, a third processing module and a fourth processing module;
the input module is used for providing an image, and the image is provided with a plurality of repeated structural units;
the first processing module is used for carrying out periodic processing on the image, dividing the image into a plurality of periodic units, and each periodic unit comprises one structural unit;
the second processing module is used for dividing each period unit into a plurality of detection units and forming a plurality of intermediate images, each intermediate image is composed of the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different;
the third processing module is used for acquiring the detection area of each intermediate image and marking the detection area;
and the fourth processing module is used for splicing the intermediate images into a target image and acquiring a target area according to the mark in the target image.
Optionally, the first processing module includes a first sub-module, a second sub-module, and a third sub-module;
the first sub-module is used for acquiring an average value of the gray scale of the image in a first side direction and an average value of the gray scale of the image in a second side direction, or acquiring an average value of the gray scale of any row of pixels of the image in the first side direction and an average value of the gray scale of any row of pixels of the image in the second side direction, wherein the first side is perpendicular to the second side, a plurality of rows of pixels are sequentially arranged along the first side direction, and a plurality of rows of pixels are sequentially arranged along the second side direction;
the second sub-module is used for carrying out Fourier transform on the average value of the gray scale in the direction of the first side edge to obtain a first frequency under a peak value and obtain a first period according to the first frequency and the length of the first side edge, fourier transforming the average value of the gray scale in the second side direction to obtain a second frequency at the peak value, and obtaining a second period based on the second frequency and the length of the second side, or, performing Fourier transform on the average value of the gray scale of any column of pixels in the direction of the first side edge to obtain a first frequency under the peak value, and obtaining a first period according to the first frequency and the length of the first side edge, carrying out Fourier transform on the average value of the gray scale of any row of pixels in the direction of the second side edge to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side edge;
the third sub-module is used for dividing the image into a plurality of period units according to the first period and the second period.
Optionally, the third processing module is configured to acquire a detection region of each intermediate image by using a threshold segmentation method.
Optionally, the third processing module is configured to set a grayscale threshold, and obtain a detection region of each intermediate image according to the grayscale threshold, where the grayscale threshold is M ± n × σ, where M is an average of grayscales of the intermediate images, n is an arbitrary coefficient, and σ is a mean square error of the grayscales of the intermediate images.
An inspection system comprising an image processing system as claimed in any one of the preceding claims.
Optionally, the detection region is a region of a defect on the surface of the object to be detected in the image; or, the detection area is an area of a specific structure on the surface of the object to be detected in the image.
Compared with the prior art, the technical scheme provided by the invention has the following advantages:
the image processing method and system, and the detection method and system provided by the invention have the advantages that after the image with a plurality of repeated structural units is provided, the image is subjected to periodic processing, the image is divided into a plurality of periodic units, each periodic unit comprises one structural unit, each periodic unit is divided into a plurality of detection units, and a plurality of intermediate images are formed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of an image processing method according to an embodiment of the present invention;
fig. 2 is an image of an object under test according to an embodiment of the present invention;
FIG. 3 is a spectrum diagram of an image in a first side direction according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a cycle unit according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a detecting unit according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of an image processing system according to an embodiment of the present invention;
FIG. 7 is a flow chart of a detection method according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a detection system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, so that the above is the core idea of the present invention, and the above objects, features and advantages of the present invention can be more clearly understood. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The present invention provides an image processing method, as shown in fig. 1, including:
s101: providing an image having a plurality of repeating structural units;
s102: carrying out periodic processing on an image, dividing the image into a plurality of periodic units, wherein each periodic unit comprises a structural unit;
s103: dividing each period unit into a plurality of detection units and forming a plurality of intermediate images, wherein the intermediate images are composed of the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different;
s104: acquiring detection areas of the intermediate images, and marking the detection areas;
s105: and splicing the intermediate images into a target image, and obtaining a target area according to the mark in the target image.
In the embodiment of the invention, the provided image is an image of an object to be detected, and the object to be detected can be a glass panel or a wafer with a plurality of repeated periodic structures. In the examples of the present invention, only a glass panel is used as an example for explanation. As shown in fig. 2, the glass panel 100 has a plurality of repeating periodic structures, i.e., pixel units 110, and transition regions 120 between adjacent pixel units 110. The structural units in the image of the glass panel 100 are the image of the pixel unit 110.
Since the pixel units 110 are transparent, and the transition region 120 is provided with an opaque black matrix, after the image of the glass panel as the object to be measured is obtained, the gray value of the pixel units 110 is different from that of the transition region 120, so that the number of the pixel units 110 of the image in the first side direction X and the number of the pixel units 110 of the image in the second side direction Y can be obtained according to the average value of the gray values, and the image is divided into a plurality of cycle units according to the number of the pixel units 110 of the image in the first side direction X and the number of the pixel units 110 in the second side direction Y, and each cycle unit has one pixel unit 110.
In the embodiment of the present invention, the first and second substrates,
after providing the image of the object to be measured, the image is periodically processed, the image is divided into a plurality of periodic units, and the method comprises the following steps: acquiring an average value of the gray scale of the image in a first side direction X and an average value of the gray scale of the image in a second side direction Y, wherein the first side direction X is vertical to the second side direction Y;
performing fourier transform on the average value of the gray scale in the first side direction X to obtain a spectrogram of the image along the first side direction X, which is shown in fig. 3, from which a first frequency F1 under the peak value can be obtained, and obtaining a first period D1 according to the first frequency F1 and the length L1 of the first side, where D1 is L1/F1; performing fourier transform on the average value of the gray scale in the second side direction Y to obtain a spectrogram of the image along the second side direction Y, obtaining a second frequency F2 at the peak from the spectrogram, and obtaining a second period D2 according to the second frequency F2 and the length L2 of the second side, wherein D2 is L2/F2; here, the first period D1 indicates the number of pixel cells 110 of the image in the first side direction X, and the second period D2 indicates the number of pixel cells 110 of the image in the second side direction Y.
The image is divided into a plurality of periodic units 130 according to the first and second periods D1 and D2, each periodic unit 130 having one pixel unit 110 therein.
Of course, the invention is not limited thereto, and in other embodiments, the image is periodically processed to divide the image into a plurality of periodic units 130, including:
acquiring an average value of the gray scale of any one or more columns of pixels of the image in a first side direction X and an average value of the gray scale of any one or more rows of pixels of the image in a second side direction Y, wherein the first side direction X is vertical to the second side direction Y, the multiple columns of pixels are sequentially arranged along the first side direction X, and the multiple rows of pixels are sequentially arranged along the second side direction Y;
performing fourier transform on an average value of grayscales of any one or more columns of pixels in the first side direction X to obtain a first frequency F1 at a peak value, obtaining a first period D1 according to the first frequency F1 and the length L1 of the first side, performing fourier transform on an average value of grayscales of any one or more rows of pixels in the second side direction Y to obtain a second frequency F2 at the peak value, and obtaining a second period D2 according to the second frequency F2 and the length L2 of the second side;
the image is divided into a plurality of period units 130 according to the first and second periods D1 and D2.
That is, in the present invention, the first period D1 may be obtained from an average value of the gradations of the entire image in the first side direction X, the first period D1 may be obtained from an average value of the gradations of the partial image in the first side direction X, the second period D2 may be obtained from an average value of the gradations of the entire image in the second side direction Y, and the second period D2 may be obtained from an average value of the gradations of the partial image in the second side direction Y. Of course, the accuracy of the manner of obtaining the first period D1 and the second period D2 from the entire image is higher, while the efficiency of the manner of obtaining the first period D1 and the second period D2 from the partial image is higher, and in practical applications, different manners may be selected for the period processing according to actual situations.
It should be noted that, since the fourier transform of the discrete periodic signal is a discrete signal, the peak with the largest peak value in the spectrogram represents the dc signal, and the peak with the second largest peak value in the spectrogram represents the period, the first frequency F1 is the frequency of the peak with the second largest peak value, and the second frequency F2 is also the frequency of the peak with the second largest peak value in the corresponding spectrogram.
After dividing the image into a plurality of period units 130, as shown in fig. 5, each period unit 130 may be divided into a plurality of detection units 140. Optionally, each detection unit 140 is a pixel of a camera for acquiring an image of the object to be detected, i.e., the glass panel 100, so as to prevent the pixel unit 110 and the transition area 120 in the image from being divided into the same detection unit 140 and affecting the accuracy of the subsequent detection area acquisition.
After each period unit 130 is divided into a plurality of detection units 140, the detection units 140 at the same position of each period unit 130 form an intermediate image, and the detection units 140 at different positions form different intermediate images, as shown in fig. 5, the detection unit 140 in the first row and the first column in each period unit 130 forms an intermediate image, the detection unit 140 in the first row and the second column in each period unit 130 forms an intermediate image, the detection unit 140 in the first row and the third column in each period unit 130 forms an intermediate image, and so on. In this way, by splitting the periodic unit 130, the texture of the background, such as the transition area 120, can be eliminated, and the accuracy of image processing can be improved.
Since the detecting units 140 in the same intermediate image are from the same position of each detecting area, the gray values of the detecting units 140 in the same intermediate image are the same, and therefore, the detecting areas in the intermediate image can be acquired by threshold segmentation. That is, after a plurality of intermediate images are formed, the detection areas of the respective intermediate images are acquired at the same time, and the detection units 140 in any one of the cycle units 130 are independent of each other, so that the plurality of intermediate images can be operated in parallel, and the image processing speed and the optical detection efficiency can be improved.
It should be noted that the detection area in the embodiment of the present invention is an area of a defect on the surface of the object in the image, or the detection area is an area of a specific structure on the surface of the object in the image, which is not limited thereto.
In this embodiment, acquiring the detection area of each intermediate image includes: and acquiring the defect detection area of each intermediate image by adopting a threshold segmentation method. Of course, in the embodiment of the present invention, other methods may also be used to obtain the detection area of the intermediate image, and the present invention is not limited thereto.
Specifically, the step of acquiring the detection area of each intermediate image by using a threshold segmentation method includes: setting a gray threshold; and acquiring the detection area of each intermediate image according to the gray threshold.
Optionally, the setting of the gray threshold comprises setting the gray threshold to M + -n × sigma, wherein M is an average value of the gray values of the intermediate image, n is an arbitrary coefficient, and sigma is a mean square error of the gray values of the intermediate image.
It should be noted that, since the detection units 140 in different intermediate images are from different positions of the period unit 130, the average values of the grayscales of different intermediate images are different, that is, after each intermediate image is formed, the average value of the grayscales of each intermediate image and the mean square error of the grayscales of each intermediate image are obtained, then the grayscale threshold of each intermediate image is calculated according to the preset coefficient n and the formula M ± n × σ, and then the defect region of the corresponding intermediate image is obtained according to the grayscale threshold.
It should be noted that, when the detection area is an area with a defect on the surface of the object to be detected, if the image of the object to be detected is obtained, bright-field illumination is adopted, the area with the gray value greater than the gray threshold is the detection area, that is, whether the gray value of the current area is greater than the gray threshold is determined, if yes, the current area is the detection area, the area is marked, and if not, the current area is not the detection area and the marking is not performed.
If the dark field illumination is adopted when the image of the object to be detected is obtained, the area with the gray value smaller than the gray threshold value is a detection area, namely whether the gray value of the current area is smaller than the gray threshold value is judged, if so, the area is the detection area, the area is marked, and if not, the area is not a defect area and the marking is not carried out.
After the detection areas of the intermediate images are marked, the intermediate images are spliced into a target image according to the position of the detection unit 140 in the intermediate images in the images, the target image is the same as the original image, only the marks of the detection areas are added on the basis of the original image, and based on the marks, the target area can be obtained according to the marks in the target image.
The image processing method provided by the invention has the advantages that after the image is provided, the image is subjected to periodic processing, the image is divided into a plurality of periodic units, each periodic unit is divided into a plurality of detection units, and a plurality of intermediate images are formed.
An embodiment of the present invention further provides an image processing system, as shown in fig. 6, including an input module 200, a first processing module 210, a second processing module 211, a third processing module 212, and a fourth processing module 213.
The input module 200 is used to provide an image having a plurality of repeating structural units. The first processing module 210 is configured to perform periodic processing on an image, and divide the image into a plurality of periodic units, where each periodic unit includes one structural unit; the second processing module 211 is configured to divide each period unit into a plurality of detection units and form a plurality of intermediate images, each intermediate image is composed of detection units at the same position of each period unit, and the positions of the detection units belonging to the same period unit in different intermediate images are different; the third processing module 212 is configured to obtain a detection area of each intermediate image, and mark the detection area; the fourth processing module 213 is configured to splice the intermediate images into a target image, and obtain a target region according to a mark in the target image.
Optionally, the first processing module 210 includes a first sub-module, a second sub-module, and a third sub-module;
the first sub-module is used for acquiring the average value of the gray scale of the image in the first side edge direction and the average value of the gray scale of the image in the second side edge direction, or acquiring the average value of the gray scale of any row of pixels of the image in the first side edge direction and the average value of the gray scale of any row of pixels of the image in the second side edge direction, wherein the first side edge is vertical to the second side edge, a plurality of rows of pixels are sequentially arranged along the first side edge direction, and a plurality of rows of pixels are sequentially arranged along the second side edge;
the second sub-module is used for performing Fourier transform on the average value of the gray scale in the direction of the first side to obtain a first frequency under the peak value, obtaining a first period according to the first frequency and the length of the first side, performing Fourier transform on the average value of the gray scale in the direction of the second side to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side, or performing Fourier transform on the average value of the gray scale of any column of pixels in the direction of the first side to obtain the first frequency under the peak value, obtaining the first period according to the first frequency and the length of the first side, performing Fourier transform on the average value of the gray scale of any row of pixels in the direction of the second side to obtain the second frequency under the peak value, and obtaining the second period according to the second frequency and the length of the;
the third sub-module is used for dividing the image into a plurality of period units according to the first period and the second period.
Optionally, the third processing module 212 is configured to obtain the detection area of each intermediate image by using a threshold segmentation method.
Since the detection units in the same intermediate image are from the same position of the detection areas, the gray values of the detection units in the same intermediate image are the same, and therefore the detection areas in the intermediate image can be obtained by means of threshold segmentation. That is, after a plurality of intermediate images are formed, the detection areas of the respective intermediate images are acquired at the same time, and since the respective detection units are independent of each other in any cycle unit, the plurality of intermediate images can be operated in parallel, and thus the speed of image processing can be increased, and the optical detection efficiency can be improved.
Further optionally, the third processing module 212 is configured to set a gray threshold, and obtain the defect region of each intermediate image according to the gray threshold, where the gray threshold is M ± n × σ, where M is an average value of the gray levels of the intermediate images, n is an arbitrary coefficient, and σ is a mean square error of the gray levels of the intermediate images.
The method sets the gray value, can realize the automation of image processing, and improves the robustness of the algorithm. In this embodiment, n is 3, but in other embodiments, n may have other values.
According to the image processing system provided by the invention, after the input module provides the image, the first processing module performs periodic processing on the image, the image is divided into a plurality of periodic units, the second processing module divides each periodic unit into a plurality of detection units and forms a plurality of intermediate images, and the intermediate images consist of the detection units at the same positions of the periodic units, and the positions of the detection units belonging to the same periodic unit in different intermediate images are different, so that the third processing module can perform parallel processing on the intermediate images to acquire the detection areas of the intermediate images, thereby effectively improving the speed of image processing and improving the detection efficiency.
An embodiment of the present invention provides a detection method, as shown in fig. 7, including:
s201: providing an object to be tested, wherein the object to be tested comprises a plurality of repeated periodic structures;
s202: acquiring an image of the object to be detected;
s203: processing the image by adopting an image processing method, wherein the image processing method is the image processing method provided by any one of the embodiments;
s204: and acquiring the detection information of the object to be detected according to the target area.
In the embodiment of the invention, the object to be measured can be a glass panel or a wafer with a plurality of repeating periodic structures. The periodic structure is imaged to form a structural unit. In the examples of the present invention, only a glass panel is used as an example for explanation.
In the embodiment of the present invention, an image of a glass panel, which is an object to be detected, is obtained first, and since the field of view of the detection device is small and the region to be detected of the glass panel 100 is large, the step of obtaining the image of the glass substrate 100 includes: scanning the glass panel 100 to obtain a plurality of images; several images are stitched to obtain an image of the entire glass panel 100. Thereafter, the image of the entire glass panel 100 is subjected to periodic processing. Of course, the invention is not limited thereto, and in other embodiments, after obtaining the plurality of images, the plurality of images may not be stitched, but each image is processed separately, and finally the images marked with the defective area are stitched to obtain the defect information of the entire glass panel 100.
Then, according to the image processing method, processing the image of the object to be detected, namely, periodically processing the image, dividing the image into a plurality of periodic units, wherein each periodic unit comprises a structural unit; dividing each period unit into a plurality of detection units and forming a plurality of intermediate images, wherein the intermediate images are composed of the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different; acquiring detection areas of the intermediate images, and marking the detection areas; splicing the intermediate images into a target image, and obtaining a target area according to a mark in the target image; and acquiring the detection information of the object to be detected according to the target area.
Optionally, the detection area is an area of the defect on the surface of the object to be detected in the image; or the detection area is an area of a specific structure on the surface of the object to be detected in the image. The detection information may be position and size information of a defective region or a specific structure region, etc.
An embodiment of the present invention further provides a detection system, as shown in fig. 8, the detection system includes an image acquisition system 20 and an image processing system 21, where the image acquisition system 20 includes a CCD image sensor or a CMOS image sensor, and the like. The image processing system 21 comprises an image processing system as provided in any of the above embodiments, i.e. the image processing system 21 comprises an input module 200, a first processing module 210, a second processing module 211, a third processing module 212 and a fourth processing module 213. The functions of the input module 200, the first processing module 210, the second processing module 211, the third processing module 212 and the fourth processing module 213 are as described above, and are not described herein again.
According to the detection method and the detection system provided by the invention, after the image of the object to be detected is obtained, the image is subjected to periodic processing, the image is divided into a plurality of periodic units, each periodic unit is divided into a plurality of detection units, and a plurality of intermediate images are formed.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (14)
1. An image processing method, comprising:
providing an image having a plurality of repeating structural units;
carrying out periodic processing on the image, dividing the image into a plurality of periodic units, wherein each periodic unit comprises one structural unit;
dividing each period unit into a plurality of detection units, and forming a plurality of intermediate images, wherein the intermediate images are formed by the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different;
acquiring a detection area of each intermediate image, and marking the detection area;
and splicing the intermediate images into a target image, and obtaining a target area according to the mark in the target image.
2. The method of claim 1, wherein the periodically processing the image, dividing the image into a plurality of periodic units, comprises:
acquiring an average value of the gray scale of the image in a first side direction and an average value of the gray scale of the image in a second side direction, wherein the first side is perpendicular to the second side;
performing fourier transform on the average value of the gray scale in the direction of the first side to obtain a first frequency under a peak value, obtaining a first period according to the first frequency and the length of the first side, performing fourier transform on the average value of the gray scale in the direction of the second side to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side;
dividing the image into a plurality of period units according to the first period and the second period.
3. The method of claim 1, wherein the periodically processing the image, dividing the image into a plurality of periodic units, comprises:
acquiring an average value of the gray scale of any one or more columns of pixels of the image in a first side direction and an average value of the gray scale of any one or more rows of pixels of the image in a second side direction, wherein the first side is vertical to the second side, the plurality of columns of pixels are sequentially arranged along the first side direction, and the plurality of rows of pixels are sequentially arranged along the second side direction;
performing fourier transform on an average value of the gray scale of the any one or more columns of pixels in the direction of the first side to obtain a first frequency under a peak value, obtaining a first period according to the first frequency and the length of the first side, performing fourier transform on an average value of the gray scale of the any one or more rows of pixels in the direction of the second side to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side;
dividing the image into a plurality of period units according to the first period and the second period.
4. The method of claim 1, wherein acquiring the detection region of each of the intermediate images comprises:
and acquiring the detection area of each intermediate image by adopting a threshold segmentation method.
5. The method of claim 4, wherein the step of obtaining the detection region of each intermediate image by using a threshold segmentation method comprises:
setting a gray threshold;
and acquiring the detection area of each intermediate image according to the gray threshold.
6. The method of claim 5, wherein setting a grayscale threshold comprises:
setting the grayscale threshold to M + -n × sigma;
wherein M is an average value of the gray values of the intermediate image, n is an arbitrary coefficient, and σ is a mean square error of the gray values of the intermediate image.
7. A method of detection, comprising:
providing an object to be tested, wherein the object to be tested comprises a plurality of repeated periodic structures;
acquiring an image of the object to be detected;
the image processing method according to any one of claims 1 to 6, processing the image;
and acquiring the detection information of the object to be detected according to the target area.
8. The method according to claim 7, wherein the detection area is an area of the image of the defect on the surface of the object; or, the detection area is an area of a specific structure on the surface of the object to be detected in the image.
9. An image processing system is characterized by comprising an input module, a first processing module, a second processing module, a third processing module and a fourth processing module;
the input module is used for providing an image, and the image is provided with a plurality of repeated structural units;
the first processing module is used for carrying out periodic processing on the image, dividing the image into a plurality of periodic units, and each periodic unit comprises one structural unit;
the second processing module is used for dividing each period unit into a plurality of detection units and forming a plurality of intermediate images, each intermediate image is composed of the detection units at the same positions of the period units, and the positions of the detection units belonging to the same period unit in different intermediate images are different;
the third processing module is used for acquiring the detection area of each intermediate image and marking the detection area;
and the fourth processing module is used for splicing the intermediate images into a target image and acquiring a target area according to the mark in the target image.
10. The system of claim 9, wherein the first processing module comprises a first sub-module, a second sub-module, and a third sub-module;
the first sub-module is used for acquiring an average value of the gray scale of the image in a first side direction and an average value of the gray scale of the image in a second side direction, or acquiring an average value of the gray scale of any row of pixels of the image in the first side direction and an average value of the gray scale of any row of pixels of the image in the second side direction, wherein the first side is perpendicular to the second side, a plurality of rows of pixels are sequentially arranged along the first side direction, and a plurality of rows of pixels are sequentially arranged along the second side direction;
the second sub-module is used for carrying out Fourier transform on the average value of the gray scale in the direction of the first side edge to obtain a first frequency under a peak value and obtain a first period according to the first frequency and the length of the first side edge, fourier transforming the average value of the gray scale in the second side direction to obtain a second frequency at the peak value, and obtaining a second period based on the second frequency and the length of the second side, or, performing Fourier transform on the average value of the gray scale of any column of pixels in the direction of the first side edge to obtain a first frequency under the peak value, and obtaining a first period according to the first frequency and the length of the first side edge, carrying out Fourier transform on the average value of the gray scale of any row of pixels in the direction of the second side edge to obtain a second frequency under the peak value, and obtaining a second period according to the second frequency and the length of the second side edge;
the third sub-module is used for dividing the image into a plurality of period units according to the first period and the second period.
11. The system of claim 9, wherein the third processing module is configured to obtain the detection region of each of the intermediate images by using a threshold segmentation method.
12. The system according to claim 11, wherein the third processing module is configured to set a gray threshold, and obtain the detection region of each of the intermediate images according to the gray threshold, the gray threshold is M ± n × σ, where M is an average value of gray values of the intermediate images, n is an arbitrary coefficient, and σ is a mean square error of gray values of the intermediate images.
13. An inspection system comprising the image processing system of any one of claims 9 to 12.
14. The system of claim 13, wherein the detection region is a region of the surface of the test object in the image where a defect is located; or, the detection area is an area of a specific structure on the surface of the object to be detected in the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526089.XA CN111325707B (en) | 2018-12-13 | 2018-12-13 | Image processing method and system, and detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526089.XA CN111325707B (en) | 2018-12-13 | 2018-12-13 | Image processing method and system, and detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111325707A true CN111325707A (en) | 2020-06-23 |
| CN111325707B CN111325707B (en) | 2021-11-30 |
Family
ID=71170086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811526089.XA Active CN111325707B (en) | 2018-12-13 | 2018-12-13 | Image processing method and system, and detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111325707B (en) |
Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070139715A1 (en) * | 2005-12-20 | 2007-06-21 | Fuji Xerox Co., Ltd. | Image forming apparatus, image forming method, and storage medium storing program for forming image |
| CN101115136A (en) * | 2006-07-26 | 2008-01-30 | 佳能株式会社 | Image processing apparatus, control method therefor, image processing method, and program |
| CN101405766A (en) * | 2006-01-07 | 2009-04-08 | 伊斯拉视像系统股份公司 | Method and system for the optical inspection of a periodic structure |
| US20130113772A1 (en) * | 2011-11-04 | 2013-05-09 | Samsung Electronics Co., Ltd. | Display panel |
| CN103196368A (en) * | 2013-03-18 | 2013-07-10 | 华东师范大学 | Automatic estimation method for single tree three-dimensional green quantity based on vehicle-mounted laser scanning data |
| CN103729856A (en) * | 2014-01-24 | 2014-04-16 | 浙江师范大学 | Method for cloth defect detection by utilizing S transformation signal extraction |
| CN103782164A (en) * | 2011-06-23 | 2014-05-07 | 沙克拉德有限责任公司 | Apparatus and method for detecting defects in metallic surface |
| CN103885217A (en) * | 2014-03-10 | 2014-06-25 | 京东方科技集团股份有限公司 | Method and device for detecting liquid crystal display panel column-shaped shock insulator defects |
| CN103985364A (en) * | 2013-02-08 | 2014-08-13 | Nlt科技股份有限公司 | Display device and checking method thereof |
| CN105004795A (en) * | 2015-08-03 | 2015-10-28 | 中国人民解放军海军工程大学 | Pseudo-flaw signal recognition method and method for improving pipeline nondestructive testing precision through same |
| CN105301810A (en) * | 2015-11-24 | 2016-02-03 | 上海斐讯数据通信技术有限公司 | Screen defect detecting method and screen defect detecting device |
| CN105486687A (en) * | 2014-10-07 | 2016-04-13 | 东友精细化工有限公司 | Touch panel inspection device and method |
| CN105787949A (en) * | 2016-03-24 | 2016-07-20 | 上海工程技术大学 | Colored woven fabric warp and weft yarn density measurement method |
| CN106198542A (en) * | 2016-07-05 | 2016-12-07 | 江南大学 | A kind of WARP-KNITTING technique fast analyser based on smart mobile phone and method |
| CN106204517A (en) * | 2015-04-17 | 2016-12-07 | 铭传大学 | Automatic optical detection method for periodic pattern |
| CN106447607A (en) * | 2016-08-25 | 2017-02-22 | 中国科学院长春光学精密机械与物理研究所 | Image stitching method and apparatus |
| CN107037050A (en) * | 2017-04-05 | 2017-08-11 | 东华大学 | A kind of method for automatic measurement of textile image Texture-period |
| CN107104730A (en) * | 2017-03-07 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | The fringe counting method and infomation detection algorithm positioned for visual light imaging |
| CN107194919A (en) * | 2017-05-18 | 2017-09-22 | 南京大学 | The mobile phone screen defect inspection method rebuild based on rule grain background |
| CN107895363A (en) * | 2017-10-31 | 2018-04-10 | 常州大学 | Textile flaw detection method based on template characteristic |
| CN108280822A (en) * | 2017-12-20 | 2018-07-13 | 歌尔科技有限公司 | The detection method and device of screen cut |
| CN108510472A (en) * | 2018-03-08 | 2018-09-07 | 北京百度网讯科技有限公司 | Method and apparatus for handling image |
| US10089763B2 (en) * | 2016-12-29 | 2018-10-02 | BioMech Sensor LLC | Systems and methods for real-time data quantification, acquisition, analysis and feedback |
| CN108830909A (en) * | 2018-05-24 | 2018-11-16 | 武汉精测电子集团股份有限公司 | Promote the image preprocessing system and method for period texture image compression ratio |
| CN108986069A (en) * | 2018-05-30 | 2018-12-11 | 哈尔滨工业大学深圳研究生院 | A kind of AMOLED display screen defect inspection method, system and storage medium |
-
2018
- 2018-12-13 CN CN201811526089.XA patent/CN111325707B/en active Active
Patent Citations (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070139715A1 (en) * | 2005-12-20 | 2007-06-21 | Fuji Xerox Co., Ltd. | Image forming apparatus, image forming method, and storage medium storing program for forming image |
| CN101405766A (en) * | 2006-01-07 | 2009-04-08 | 伊斯拉视像系统股份公司 | Method and system for the optical inspection of a periodic structure |
| CN101115136A (en) * | 2006-07-26 | 2008-01-30 | 佳能株式会社 | Image processing apparatus, control method therefor, image processing method, and program |
| CN103782164A (en) * | 2011-06-23 | 2014-05-07 | 沙克拉德有限责任公司 | Apparatus and method for detecting defects in metallic surface |
| US20130113772A1 (en) * | 2011-11-04 | 2013-05-09 | Samsung Electronics Co., Ltd. | Display panel |
| CN103985364A (en) * | 2013-02-08 | 2014-08-13 | Nlt科技股份有限公司 | Display device and checking method thereof |
| CN103196368A (en) * | 2013-03-18 | 2013-07-10 | 华东师范大学 | Automatic estimation method for single tree three-dimensional green quantity based on vehicle-mounted laser scanning data |
| CN103729856A (en) * | 2014-01-24 | 2014-04-16 | 浙江师范大学 | Method for cloth defect detection by utilizing S transformation signal extraction |
| CN103885217A (en) * | 2014-03-10 | 2014-06-25 | 京东方科技集团股份有限公司 | Method and device for detecting liquid crystal display panel column-shaped shock insulator defects |
| CN105486687A (en) * | 2014-10-07 | 2016-04-13 | 东友精细化工有限公司 | Touch panel inspection device and method |
| CN106204517A (en) * | 2015-04-17 | 2016-12-07 | 铭传大学 | Automatic optical detection method for periodic pattern |
| CN105004795A (en) * | 2015-08-03 | 2015-10-28 | 中国人民解放军海军工程大学 | Pseudo-flaw signal recognition method and method for improving pipeline nondestructive testing precision through same |
| CN105301810A (en) * | 2015-11-24 | 2016-02-03 | 上海斐讯数据通信技术有限公司 | Screen defect detecting method and screen defect detecting device |
| CN105787949A (en) * | 2016-03-24 | 2016-07-20 | 上海工程技术大学 | Colored woven fabric warp and weft yarn density measurement method |
| CN106198542A (en) * | 2016-07-05 | 2016-12-07 | 江南大学 | A kind of WARP-KNITTING technique fast analyser based on smart mobile phone and method |
| CN106447607A (en) * | 2016-08-25 | 2017-02-22 | 中国科学院长春光学精密机械与物理研究所 | Image stitching method and apparatus |
| US10089763B2 (en) * | 2016-12-29 | 2018-10-02 | BioMech Sensor LLC | Systems and methods for real-time data quantification, acquisition, analysis and feedback |
| CN107104730A (en) * | 2017-03-07 | 2017-08-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | The fringe counting method and infomation detection algorithm positioned for visual light imaging |
| CN107037050A (en) * | 2017-04-05 | 2017-08-11 | 东华大学 | A kind of method for automatic measurement of textile image Texture-period |
| CN107194919A (en) * | 2017-05-18 | 2017-09-22 | 南京大学 | The mobile phone screen defect inspection method rebuild based on rule grain background |
| CN107895363A (en) * | 2017-10-31 | 2018-04-10 | 常州大学 | Textile flaw detection method based on template characteristic |
| CN108280822A (en) * | 2017-12-20 | 2018-07-13 | 歌尔科技有限公司 | The detection method and device of screen cut |
| CN108510472A (en) * | 2018-03-08 | 2018-09-07 | 北京百度网讯科技有限公司 | Method and apparatus for handling image |
| CN108830909A (en) * | 2018-05-24 | 2018-11-16 | 武汉精测电子集团股份有限公司 | Promote the image preprocessing system and method for period texture image compression ratio |
| CN108986069A (en) * | 2018-05-30 | 2018-12-11 | 哈尔滨工业大学深圳研究生院 | A kind of AMOLED display screen defect inspection method, system and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| YI-HUNG LIU等: ""Automatic Defect Detection for TFT-LCD Array Process Using Quasiconformal Kernel Support Vector Data Description"", 《INT J MOL SCI》 * |
| 简川霞 等: ""手机玻璃屏表面缺陷视觉检测方法研究"", 《包装工程》 * |
| 董明利 等: ""基于光学傅里叶变换的周期性微结构缺陷检测"", 《光学精密工程》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111325707B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20070047801A1 (en) | Defect detecting method and defect detecting device | |
| CN109932369B (en) | Method and device for detecting special-shaped display panel | |
| JP2010164487A (en) | Defect inspecting apparatus and defect inspecting method | |
| CN113538603B (en) | Optical detection method and system based on array product and readable storage medium | |
| JP2003085558A (en) | Method of pattern inspection and inspection device | |
| CN104331695A (en) | Robust round identifier shape quality detection method | |
| CN104749801B (en) | High Precision Automatic optical detecting method and system | |
| CN114757913A (en) | Display screen defect detection method | |
| JP2005172559A (en) | Method and device for detecting line defect on panel | |
| WO2004083901A2 (en) | Detection of macro-defects using micro-inspection inputs | |
| JP3907874B2 (en) | Defect inspection method | |
| CN112986259B (en) | Defect detection method and device for manufacturing process of intelligent terminal OLED panel | |
| JP2006138708A (en) | Image flaw inspection method, image flaw inspecting device and visual inspection device | |
| CN111325707B (en) | Image processing method and system, and detection method and system | |
| JP2005345290A (en) | Streak-like flaw detecting method and streak-like flaw detector | |
| CN111598851A (en) | Solar cell fragment detection method based on morphological image processing | |
| KR101702752B1 (en) | Method of inspecting electronic components | |
| JP2009097959A (en) | Defect detecting device and defect detection method | |
| KR20140082333A (en) | Method and apparatus of inspecting mura of flat display | |
| US8059886B2 (en) | Adaptive signature detection | |
| CN108303640B (en) | Line detection method and device | |
| CN111640122B (en) | LED special-shaped display screen facula positioning and dividing method based on connected domain | |
| JP5239275B2 (en) | Defect detection method and defect detection apparatus | |
| CN114565558A (en) | CMOS chip surface defect detection method and system | |
| JP3127598B2 (en) | Method for extracting density-varying constituent pixels in image and method for determining density-fluctuation block |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 518110 101, 201, 301, No.2, Shanghenglang fourth industrial zone, Tongsheng community, Dalang street, Longhua District, Shenzhen City, Guangdong Province Applicant after: Shenzhen Zhongke feice Technology Co.,Ltd. Address before: 518110 101, 201, 301, No.2, Shanghenglang fourth industrial zone, Tongsheng community, Dalang street, Longhua District, Shenzhen City, Guangdong Province Applicant before: Shenzhen Zhongke Flying Test Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |