CN104749801A - High-precision automatic optical detection method and high-precision automatic optical detection system - Google Patents
High-precision automatic optical detection method and high-precision automatic optical detection system Download PDFInfo
- Publication number
- CN104749801A CN104749801A CN201310754700.5A CN201310754700A CN104749801A CN 104749801 A CN104749801 A CN 104749801A CN 201310754700 A CN201310754700 A CN 201310754700A CN 104749801 A CN104749801 A CN 104749801A
- Authority
- CN
- China
- Prior art keywords
- image
- effective coverage
- pure color
- lcds
- visual field
- 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
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02F—OPTICAL DEVICES OR ARRANGEMENTS FOR THE CONTROL OF LIGHT BY MODIFICATION OF THE OPTICAL PROPERTIES OF THE MEDIA OF THE ELEMENTS INVOLVED THEREIN; NON-LINEAR OPTICS; FREQUENCY-CHANGING OF LIGHT; OPTICAL LOGIC ELEMENTS; OPTICAL ANALOGUE/DIGITAL CONVERTERS
- G02F1/00—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics
- G02F1/01—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour
- G02F1/13—Devices or arrangements for the control of the intensity, colour, phase, polarisation or direction of light arriving from an independent light source, e.g. switching, gating or modulating; Non-linear optics for the control of the intensity, phase, polarisation or colour based on liquid crystals, e.g. single liquid crystal display cells
- G02F1/1306—Details
- G02F1/1309—Repairing; Testing
Landscapes
- Physics & Mathematics (AREA)
- Nonlinear Science (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Testing Of Optical Devices Or Fibers (AREA)
- Liquid Crystal (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a high-precision automatic optical detection method and a high-precision automatic optical detection system. The high-precision automatic optical detection method includes displaying first pure colors and second pure colors in a plurality of effective areas under the control; acquiring a plurality of view area images and extracting effective area images from the various view area images; interchanging the pure colors of the adjacent effective areas; acquiring a plurality of view area images after the pure colors are interchanged with one another, and extracting effective area images from the various view area images after the pure colors are interchanged with one another; processing the various effective area images, recognizing defects, recording the locations of the defects and counting the quantities of the defects; acquiring the types of defects in liquid crystal display images in first pure color display states and second pure color display states by means of computing, recording the locations of the defects and counting the quantities of the defects. The pure colors of the adjacent effective areas are different from one another. The high-precision automatic optical detection method and the high-precision automatic optical detection system have the advantages that only each single effective image needs to be processed in each procedure, accordingly, the shooting precision is high, the defects can be accurately detected, and the detection accuracy can be improved.
Description
Technical field
The present invention relates to LCDs detection field, particularly relate to a kind of High Precision Automatic optical detecting method and system.
Background technology
In recent years, due to LCD(Liquid Crystal Display, LCDs) there is light, thin superperformance, therefore in the communication product (as auto-navigation system, mobile phone) of the overwhelming majority, consumption electronic products (as LCD TV, video camera), the field such as instrument product and industrial automation product, all use LCD as control panel, its range of application is very extensive.Because the whole technological process of production of LCD is long, and substrate size is increasing, and wire sizes is more and more accurate, therefore, needs to carry out strict quality control in the production run of LCD.Traditional LCD detection method mainly by manual detection, due to the subjective differences of people, can bring a lot of uncontrollable factor to quality testing.Adopt Machine Vision Inspecting System to replace manual operation for this reason, all drawbacks that manual detection is brought can be eliminated, quality and the efficiency of detection can be improved.
But current Machine Vision Inspecting System carries out detection defect, detect whole LCDs, degree of accuracy is not high, and accuracy rate is lower.
Summary of the invention
Based on this, be necessary for traditional High Precision Automatic optical detection degree of accuracy not high, the problem that accuracy rate is low, a kind of High Precision Automatic optical detecting method is provided, the degree of accuracy and accuracy that detect can be improved.
In addition, there is a need to provide a kind of High Precision Automatic Systems for optical inspection, the degree of accuracy and accuracy that detect can be improved.
A kind of High Precision Automatic optical detecting method, comprising:
Control multiple effective coverage and show the first pure color and the second pure color, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance;
Obtain multiple area of visual field image, from each area of visual field image, extract effective coverage image;
Each effective coverage image is processed, defect recognition, recording defect position, and statistical shortcomings quantity;
By the pure color swap between adjacent effective coverage;
Obtain the multiple area of visual field images after pure color swap, from each area of visual field image, extract the effective coverage image after pure color swap;
Effective coverage image after each exchanges is processed, defect recognition, recording defect position, and statistical shortcomings quantity;
The effective coverage image of display the first pure color before exchange is formed the liquid crystal display screen image under the first pure color display state with the effective coverage exchanging rear display the first pure color, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
A kind of High Precision Automatic Systems for optical inspection, comprises display control module, acquisition module, processing module and synthesis module;
Described display control module shows the first pure color and the second pure color for controlling multiple effective coverage, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance;
Described acquisition module, for obtaining multiple area of visual field image, extracts effective coverage image from each area of visual field image;
Described processing module is used for processing each effective coverage image, defect recognition, recording defect position, and statistical shortcomings quantity;
Described display control module is also for by the pure color swap between adjacent effective coverage;
Described acquisition module also for obtaining the multiple area of visual field images after pure color swap, extracts the effective coverage image after pure color swap from each area of visual field image;
Described processing module also for each exchange after effective coverage image process, defect recognition, recording defect position, and statistical shortcomings quantity;
Described synthesis module is used for the liquid crystal display screen image formed with the effective coverage exchanging rear display the first pure color by the effective coverage image of display the first pure color before exchange under the first pure color display state, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
Above-mentioned High Precision Automatic optical detecting method and system, detected LCDs is divided into multiple effective coverage, only need to process single effective coverage image respectively at every turn, shooting precision is high, accurately can detect defect, improve the accuracy of detection, and each process single effective coverage image, calculated amount is low.
Accompanying drawing explanation
The schematic diagram of a kind of implementation environment of Fig. 1 involved by High Precision Automatic optical detecting method;
Fig. 2 is the process flow diagram of High Precision Automatic optical detecting method in an embodiment;
Fig. 3 is that in an embodiment, detected LCDs is divided into the schematic diagram of 12 parts;
Fig. 4 is the process flow diagram of High Precision Automatic optical detecting method in another embodiment;
Fig. 5 is the process flow diagram in an embodiment, whole LCDs being divided into multiple effective coverage;
Fig. 6 is the particular flow sheet of step 206 or step 212 in Fig. 2 in an embodiment;
Fig. 7 is the structured flowchart of High Precision Automatic Systems for optical inspection in an embodiment;
Fig. 8 is the inner structure block diagram of processing module in an embodiment;
Fig. 9 is the structured flowchart of High Precision Automatic Systems for optical inspection in another embodiment;
Figure 10 is the inner structure block diagram dividing module in an embodiment.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
The schematic diagram of a kind of implementation environment of Fig. 1 involved by High Precision Automatic optical detecting method.This implementation environment comprises video camera 110, light source 120, image pick-up card 130, computing machine 140, display 150, detected LCDs 160 and drive system 170.Wherein, computing machine 140 is connected with detected LCDs 160 with image pick-up card 130, display 150 respectively, and video camera 110 is connected with image pick-up card 130, and drive system 170 is connected with detected LCDs 160 with computing machine 140 respectively.Computing machine 140 controls detected LCDs 160 and shows image; Drive system 170 controls video camera 110 and moves and choose perform region; Video camera 110 is detected the image (image of display when display screen is lighted) that LCDs 160 shows and the image (image of display when display screen is not lighted) being detected LCDs 160 under light source 120 irradiates by lens shooting; The image that the image pick-up card 130 pairs of video cameras 110 are taken gathers, and is transferred to computing machine 140 and carries out analyzing and processing and obtain defect type, defective locations and defects count, and by display 150 display defect type, defective locations and defects count.This drive system 170 can comprise machinery mount, motor and telecontrol equipment.Detected LCDs 160 is placed on the machinery mount of drive system 170, and described drive system 170 moves for actuated camera 110.
In order to improve precision and the stability of system, need to get defect area more than 3 to 4 pixels, because if corresponding one an of pixel detects defect, then an arbitrary interference pixel all may be mistaken as defect.In order to accuracy of detection reaches RGB(Red-Green-Blue) subpixel accuracy, need ensure that a complete point drops in R component, R sub-pixel at least needs 2 pixels to represent, in like manner G, B component also needs 2 pixels to represent respectively, whole some needs 6 pixels.Consider horizontal and vertical situation, a physical picture element point in LCDs adopts 6*6 camera pixel point to represent.Be the LCDs of 1920*1080 for physical resolution, total 1920*1080=2073600 pixel, then need 1920*1080*(6*6) individual camera pixel point represents, namely video camera complete whole LCDs shooting needed for the summation of pixel be 1920*1080*(6*6) individual.Be only 1920*1080 with physical resolution be example herein, the display screen of other physical resolution can also be detected, as 1024*768 or 800*600 etc.
According to 500W(ten thousand) video camera of pixel shooting (suppose that resolution is: 2588*1940), laterally need to clap 1920*6/2588=5 time, longitudinal direction needs bat 1080*6/1940=4 time.If with 4 this kind of video cameras along shielding longitudinal lay out in parallel to cover whole longitudinal direction, then transversely clapping respectively in conjunction with drive system control shooting unit and can cover whole panel 5 times.
In sum, when LCDs physical resolution is constant, detection number of times and panel size size have nothing to do.Number of times is constant, the panel that size is larger, and the visual field of each shooting is larger.So, when detection faces board size changes, the visual field of video camera shooting only need be adjusted.
Suppose that the visual angle of camera lens X-direction is β, the visual angle of Y-direction is α, and video camera is d to the distance of detected LCDs, X-direction visual field lx, Y-direction visual field ly, and the relation between them is as follows:
Camera lens X-direction visual field lx=2*d*tag (β/2)
Camera lens Y-direction visual field ly=2*d*tag (α/2)
Learnt by above-mentioned relation formula, want to change the visual field, only need adjust the spacing d between camera to thing to be checked.Ly is also in change simultaneously, so also need to adjust the spacing between video camera and video camera.In order to avoid undetected screen region, retain between video camera and video camera and necessarily repeat area of visual field.
After having detected, need to add up detection defect, because each defect is very little, only have 0.06 millimeter or less.For the ease of the statistical shortcomings quantity of precision, adopt domain division method statistical shortcomings quantity.
Fig. 2 is the process flow diagram of High Precision Automatic optical detecting method in an embodiment.This High Precision Automatic optical detecting method can be applicable to the experimental situation in Fig. 1.This High Precision Automatic optical detecting method, comprising:
Step 202, controls multiple effective coverage and shows the first pure color and the second pure color, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance.
Wherein, the first pure color and the second pure color can be selected as required, and if the first pure color is white, the second pure color is black; Or the first pure color is red, the second pure color is green; Or the first pure color is red, the second pure color is blue; First pure color is green, and the second pure color is blueness etc.Whole LCDs can be divided into multiple effective coverage in advance, and the side of non-decile also can be adopted to be divided into multiple effective coverage, and will meet effective coverage when non-decile must be less than area of visual field.
As shown in Figure 3, detected LCDs 300 is divided into 12 effective coverages 310, and area of visual field 320 size equals the inactive area 330 size sum of effective coverage 310 size and this effective coverage 310 surrounding.Area of visual field is the region captured by camera.Inactive area refers to the region that all at least high than the video camera displacement accuracy order of magnitude of its X-direction and Y-direction is formed.
The alternate display detected LCDs 160 being carried out the first pure color and the second pure color is controlled by computing machine 140.
Step 204, obtains multiple area of visual field image, from each area of visual field image, extract effective coverage image.
First the distance between video camera and detected LCDs is adjusted, each camera coverage repeat region is made to equal inactive area size, and then to be coordinated by video camera 110 and image pick-up card 130 and obtain each area of visual field image respectively, effective coverage image is extracted, effective coverage image and area of visual field image bulk portion broad in the middle from area of visual field image.
Extract effective coverage method, comprise (1) to (6), as follows:
(1) image center is got;
(2) computing center's point RGB component;
(3) principal component is determined;
Such as, determine that R component is principal component, then follow-uply remove G component and B component.
(4) other component is removed;
(5) connected region is calculated;
(6) getting the larger connected region in centre is effective coverage.
Step 206, processes each effective coverage image, defect recognition, recording defect position, and statistical shortcomings quantity.
Concrete, this defect comprises bright spot, dim spot, bright line, concealed wire etc.
Step 208, by the pure color swap between adjacent effective coverage.
Concrete, change the effective coverage previously showing the first pure color into display second pure color, previously the effective coverage of display the second pure color changed the pure color of display first into.Such as, the first pure color is red, and the second pure color is blue, first adjacent effective coverage is set to red indigo plant alternate, then by red blue color swap.
Step 210, obtains the multiple area of visual field images after pure color swap, extracts the effective coverage image after pure color swap from each area of visual field image.
Step 212, processes the effective coverage image after each exchanges, defect recognition, recording defect position, and statistical shortcomings quantity.
Step 214, the effective coverage image of display the first pure color before exchange is formed the liquid crystal display screen image under the first pure color display state with the effective coverage exchanging rear display the first pure color, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
Above-mentioned High Precision Automatic optical detecting method, detected LCDs is divided into multiple effective coverage, only need to process single effective coverage image respectively at every turn, shooting precision is high, accurately can detect defect, improve the accuracy of detection, and each process single effective coverage image, calculated amount is low.
Fig. 4 is the process flow diagram of High Precision Automatic optical detecting method in another embodiment.Shown in composition graphs 1, Fig. 3 and Fig. 4, this High Precision Automatic optical detecting method, comprising:
Step 402, is divided into multiple effective coverage by whole LCDs.
Step 404, controls multiple effective coverage and shows the first pure color and the second pure color, and the pure color difference between adjacent effective coverage.
Step 406, obtains multiple area of visual field image, from each area of visual field image, extract effective coverage image.
Step 408, processes each effective coverage image, defect recognition, recording defect position, and statistical shortcomings quantity.
Step 410, by the pure color swap between adjacent effective coverage.
Step 412, obtains the multiple area of visual field images after pure color swap, extracts the effective coverage image after pure color swap from each area of visual field image.
Step 414, processes the effective coverage image after each exchanges, defect recognition, recording defect position, and statistical shortcomings quantity.
Step 416, the effective coverage image of display the first pure color before exchange is formed the liquid crystal display screen image under the first pure color display state with the effective coverage exchanging rear display the first pure color, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
Above-mentioned High Precision Automatic optical detecting method, detected LCDs is divided into multiple effective coverage, only need to process single effective coverage image respectively at every turn, shooting precision is high, accurately can detect defect, improve the accuracy of detection, and each process single effective coverage image, calculated amount is low.
Fig. 5 is the process flow diagram in an embodiment, whole LCDs being divided into multiple effective coverage.As shown in Figure 3 and Figure 5, whole LCDs is divided into multiple effective coverage, comprises:
Step 502, according to the physical resolution of detected LCDs with represent that the camera pixel preset adopted needed for each physical picture element point is counted the total pixel needed for calculating.
Concrete, represent that the required camera pixel preset adopted of each physical picture element point is counted and can be set as required, as can be 3*3,4*4 is individual, 6*6 is individual, 8*8 is individual or 9*9 camera pixel point represents a physical picture element point, if 3*3 or 4*4, each field range is large, but accuracy of detection may be inadequate, if 8*8 or 9*9, each field range is little, then testing time is many.The required camera pixel preset adopted of each physical picture element point of preferred expression is counted as 6*6, and its precision can reach RGB sub-pixel-level, and it is moderate to detect number of times.
Total=detected LCDs X-direction resolution * 6 of total pixel X of video camera imaging X-direction
Total=detected LCDs Y-direction resolution * 6 of total pixel Y of video camera imaging Y-direction
Step 504, calculates according to resolution of video camera the detected LCDs pixel number that each video camera can take.
Such as suppose that resolution of video camera is ResX*ResY, then
Area of visual field X-direction is detected LCDs physical picture element point number nX=ResX/6, and area of visual field Y-direction is detected LCDs physical picture element point number nY=ResY/6.
Step 506, calculates each camera coverage area size.
Suppose that detected LCDs pixel size is xx millimeter * yy millimeter, pixel size refers to the lateral separation * fore-and-aft distance between adjacent two pixels.
Area of visual field is of a size of nSizeX*nSizeY.NSizeX=nX*xx millimeter, nSizeY=nY*yy millimeter.
Step 508, according to area of visual field size and video camera displacement accuracy, calculates effective coverage size.
Effective coverage size X=area of visual field X-inactive area X*2;
Effective coverage size Y=area of visual field Y-inactive area Y*2.All at least high than the video camera displacement accuracy order of magnitude of described inactive area X and described inactive area Y.
Step 510, divides whole detected LCDs according to effective coverage size.
Determine effective area size by the physical picture element of video camera precision and detected LCDs, effective area size can be determined more accurately, improve and detect degree of accuracy and accuracy.
Fig. 6 is the particular flow sheet of step 206 or step 212 in Fig. 2 in an embodiment.Step 206 or step 212 specifically comprise:
Step 602, transfers the effective coverage image of acquisition to pre-set image form.
In the present embodiment, the liquid crystal display screen image gathered by video camera 110 and image pick-up card 130 cooperation is processed into the data structure of specific format through the integrated image processing software of video camera 110 self, in this data structure, image information through overcompression or may change into image data structure, for this reason, need this acquisition liquid crystal display screen image to be converted to the accessible pre-set image form of computing machine.This pre-set image form can be the picture formats such as bmp, gif.
Step 604, carries out pre-service by the effective coverage image transferring pre-set image form to.
This pre-service comprises image denoising and filtering process.Denoising can remove Gaussian noise, salt-pepper noise etc.Filtering process can adopt 1 × 8 template to carry out longitudinal mean filter, removes noise further.The quality of image is improve by pre-service.
Step 606, carries out Threshold segmentation by pretreated effective coverage image, wiping out background information, extracts image information in the image of effective coverage.
Because applied environment is uncertain, automatic threshold segmentation can be adopted.Automatic threshold segmentation is based on grey level histogram, analyzes the characteristic of image information and background information in grey level histogram, gets two peak-to-peak troughs of ripple as segmentation threshold, thus obtain threshold value.In the present embodiment, liquid crystal display screen image is converted to grey level histogram, analyzes the characteristic of image information and background information in grey level histogram, get two peak-to-peak troughs of ripple as segmentation threshold, split by this segmentation threshold, wiping out background information, extract image information.
Step 608, carries out enhancing process to this image information.
Because image information is more weak, by Morphological scale-space, image information is strengthened.
Step 610, extracting image deflects from strengthening this image information after processing, image deflects being carried out Iamge Segmentation and obtains defect block.
Pass through RGB(Red-Green-Blue) component threshold value, from image information, extract image deflects, this RGB component threshold value according to many data experiment statistics obtains.According to the connectedness of image, image deflects are divided into multiple defect block.After being divided into defect block, in image recognition processes, only need calculate the information of each defect block, greatly reduce the operand of image procossing, save the time overhead of defect recognition.
Step 612, identifies defect type in this defect block, recording defect position, and statistical shortcomings quantity.
Concrete, set up defect characteristic database in advance, in this defect characteristic database, record the feature of every class defect.The feature recorded in the defect of detection and defect characteristic database is contrasted, identifies the type of the defect of this detection.
By changing the image gathered, pre-service, enhancing, the process such as segmentation, improve the quality of image, reduce the calculated amount of image recognition.
Fig. 7 is the structured flowchart of High Precision Automatic Systems for optical inspection in an embodiment.The present embodiment is applied to the experimental situation shown in Fig. 1 with High Precision Automatic Systems for optical inspection and is described.This High Precision Automatic Systems for optical inspection, comprises display control module 720, acquisition module 740, processing module 760 and synthesis module 780.Wherein:
Display control module 720 shows the first pure color and the second pure color for controlling multiple effective coverage, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance.Whole LCDs can be divided into multiple effective coverage in advance, and the side of non-decile also can be adopted to be divided into multiple effective coverage, and will meet effective coverage when non-decile must be less than area of visual field.
Wherein, the first pure color and the second pure color can be selected as required, and if the first pure color is white, the second pure color is black; Or the first pure color is red, the second pure color is green; Or the first pure color is red, the second pure color is blue; First pure color is green, and the second pure color is blueness etc.
Acquisition module 740, for obtaining multiple area of visual field image, extracts effective coverage image from each area of visual field image.
Concrete, this defect comprises bright spot, dim spot, bright line, concealed wire etc.
Processing module 760 for processing each effective coverage image, defect recognition, recording defect position, and statistical shortcomings quantity.
Display control module 720 is also for by the pure color swap between adjacent effective coverage.
Acquisition module 740 also for obtaining the multiple area of visual field images after pure color swap, extracts the effective coverage image after pure color swap from each area of visual field image.
Processing module 760 also for each exchange after effective coverage image process, defect recognition, recording defect position, and statistical shortcomings quantity.
Synthesis module 780 is for forming the liquid crystal display screen image under the first pure color display state by the effective coverage of the effective coverage image of display the first pure color before exchange and the rear display of exchange the first pure color, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
Above-mentioned High Precision Automatic Systems for optical inspection, detected LCDs is divided into multiple effective coverage, only need to process single effective coverage image respectively at every turn, shooting precision is high, accurately can detect defect, improve the accuracy of detection, and each process single effective coverage image, calculated amount is low.
Fig. 8 is the inner structure block diagram of processing module in an embodiment.This processing module 740 comprises format conversion unit 741, pretreatment unit 742, Threshold segmentation unit 743, image enhancing unit 744, image segmentation unit 745 and identification record unit 746.
Format conversion unit 741 is for transferring the liquid crystal display screen image of acquisition to pre-set image form.This pre-set image form can be the picture formats such as bmp, gif.
Pretreatment unit 742 is for carrying out pre-service by the liquid crystal display screen image transferring pre-set image form to.
This pre-service comprises image denoising and filtering process.Denoising can remove Gaussian noise, salt-pepper noise etc.Filtering process can adopt 1 × 8 template to carry out longitudinal mean filter, removes noise further.The quality of image is improve by pre-service.
Threshold segmentation unit 743 for pretreated liquid crystal display screen image being carried out Threshold segmentation, wiping out background information, image information in extract crystal display screen image.
Because applied environment is uncertain, automatic threshold segmentation can be adopted.Automatic threshold segmentation is based on grey level histogram, analyzes the characteristic of image information and background information in grey level histogram, gets two peak-to-peak troughs of ripple as segmentation threshold, thus obtain threshold value.In the present embodiment, liquid crystal display screen image is converted to grey level histogram, analyzes the characteristic of image information and background information in grey level histogram, get two peak-to-peak troughs of ripple as segmentation threshold, split by this segmentation threshold, wiping out background information, extract image information.
Image enhancing unit 744 is for carrying out enhancing process to this image information.Because image information is more weak, by Morphological scale-space, image information is strengthened.
Image deflects, for extracting image deflects from strengthening in this image information after processing, being carried out Iamge Segmentation and being obtained defect block by image segmentation unit 745.
Pass through RGB(Red-Green-Blue) component threshold value, from image information, extract image deflects, this RGB component threshold value according to many data experiment statistics obtains.According to the connectedness of image, image deflects are divided into multiple defect block.After being divided into defect block, in image recognition processes, only need calculate the information of each defect block, greatly reduce the operand of image procossing, save the time overhead of defect recognition.
Identification record unit 746 identifies defect type in this defect block, recording defect position, and statistical shortcomings quantity.
Concrete, set up defect characteristic database in advance, in this defect characteristic database, record the feature of every class defect.The feature recorded in the defect of detection and defect characteristic database is contrasted, identifies the type of the defect of this detection.
By changing the image gathered, pre-service, enhancing, the process such as segmentation, improve the quality of image, reduce the calculated amount of image recognition.
Fig. 9 is the structured flowchart of High Precision Automatic Systems for optical inspection in another embodiment.The present embodiment is applied to the experimental situation shown in Fig. 1 with High Precision Automatic Systems for optical inspection and is described.This High Precision Automatic Systems for optical inspection, except comprising display control module 720, acquisition module 740, processing module 760 and synthesis module 780, also comprising and dividing module 710.Wherein: divide module 710 in advance whole LCDs being divided into multiple effective coverage.Whole LCDs can be divided into multiple effective coverage in advance, and the side of non-decile also can be adopted to be divided into multiple effective coverage, and will meet effective coverage when non-decile must be less than area of visual field.
Figure 10 is the inner structure block diagram dividing module in an embodiment.This division module 710 comprises total pixel computing unit 711, pixel number calculating unit 712, area of visual field dimension calculating unit 713, effective coverage dimension calculating unit 714 and division unit 715.
Total pixel computing unit 711 is detected the physical resolution of LCDs for basis and represents total pixel that the camera pixel preset adopted needed for each physical picture element point is counted needed for calculating.
Concrete, represent that the required camera pixel preset adopted of each physical picture element point is counted and can be set as required, as can be 3*3,4*4 is individual, 6*6 is individual, 8*8 is individual or 9*9 camera pixel point represents a physical picture element point, if 3*3 or 4*4, each field range is large, but accuracy of detection may be inadequate, if 8*8 or 9*9, each field range is little, then testing time is many.Preferably 6*6, its precision can reach RGB sub-pixel-level, and it is moderate to detect number of times.
Total=detected LCDs X-direction resolution * 6 of total pixel X of video camera imaging X-direction
Total=detected LCDs Y-direction resolution * 6 of total pixel Y of video camera imaging Y-direction
The detected LCDs pixel number that pixel number calculating unit 712 can be taken for calculating each video camera according to resolution of video camera.
Such as suppose that resolution of video camera is ResX*ResY, then
Area of visual field X-direction is detected LCDs physical picture element point number nX=ResX/6, and area of visual field Y-direction is detected LCDs physical picture element point number nY=ResY/6.
Area of visual field dimension calculating unit 713 is for calculating each camera coverage area size.
Suppose that detected LCDs pixel size is xx millimeter * yy millimeter, pixel size refers to the lateral separation * fore-and-aft distance between adjacent two pixels.
Area of visual field is of a size of nSizeX*nSizeY.NSizeX=nX*xx millimeter, nSizeY=nY*yy millimeter.
Effective coverage dimension calculating unit 714, for according to area of visual field size and video camera displacement accuracy, calculates effective coverage size.
Effective coverage size X=area of visual field X-inactive area X;
Effective coverage size Y=area of visual field Y-inactive area Y.
All at least high than the video camera displacement accuracy order of magnitude of described inactive area X and described inactive area Y.
Division unit 715 divides whole detected LCDs according to effective coverage size.
The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. a High Precision Automatic optical detecting method, comprising:
Control multiple effective coverage and show the first pure color and the second pure color, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance;
Obtain multiple area of visual field image, from each area of visual field image, extract effective coverage image;
Each effective coverage image is processed, defect recognition, recording defect position, and statistical shortcomings quantity;
By the pure color swap between adjacent effective coverage;
Obtain the multiple area of visual field images after pure color swap, from each area of visual field image, extract the effective coverage image after pure color swap;
Effective coverage image after each exchanges is processed, defect recognition, recording defect position, and statistical shortcomings quantity;
The effective coverage image of display the first pure color before exchange is formed the liquid crystal display screen image under the first pure color display state with the effective coverage exchanging rear display the first pure color, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
2. High Precision Automatic optical detecting method according to claim 1, it is characterized in that, described each effective coverage image to be processed, defect recognition, recording defect position, and statistical shortcomings quantity, or the effective coverage image after each exchanges is processed, defect recognition, recording defect position, and the step of statistical shortcomings quantity comprises:
Transfer the effective coverage image of acquisition to pre-set image form;
The effective coverage image transferring pre-set image form to is carried out pre-service;
Pretreated effective coverage image is carried out Threshold segmentation, wiping out background information, extract image information in the image of effective coverage;
Enhancing process is carried out to described image information;
Extracting image deflects from strengthening the described image information after processing, image deflects being carried out Iamge Segmentation and obtains defect block;
Identify defect type in described defect block, recording defect position, and statistical shortcomings quantity.
3. High Precision Automatic optical detecting method according to claim 1, it is characterized in that, the first pure color and the second pure color is shown in the multiple effective coverage of described control, and the pure color difference between adjacent effective coverage, wherein, before whole LCDs is divided into the step of multiple effective coverage in advance, described method also comprises:
In advance whole LCDs is divided into multiple effective coverage.
4. High Precision Automatic optical detecting method according to claim 3, is characterized in that, the described step in advance whole LCDs being divided into multiple effective coverage comprises:
According to the physical resolution of detected LCDs with represent that the camera pixel preset adopted needed for each physical picture element point is counted the total pixel needed for calculating;
The detected LCDs pixel number that each video camera can take is calculated according to resolution of video camera;
Calculate each camera coverage area size;
According to area of visual field size and telecontrol equipment precision, calculate effective coverage size;
According to effective coverage size, whole detected LCDs is divided.
5. High Precision Automatic optical detecting method according to claim 4, is characterized in that, the required camera pixel preset adopted of each physical picture element point of described expression is counted as 6*6.
6. a High Precision Automatic Systems for optical inspection, is characterized in that, comprises display control module, acquisition module, processing module and synthesis module;
Described display control module shows the first pure color and the second pure color for controlling multiple effective coverage, and the pure color difference between adjacent effective coverage, wherein, whole LCDs is divided into multiple effective coverage in advance;
Described acquisition module, for obtaining multiple area of visual field image, extracts effective coverage image from each area of visual field image;
Described processing module is used for processing each effective coverage image, defect recognition, recording defect position, and statistical shortcomings quantity;
Described display control module is also for by the pure color swap between adjacent effective coverage;
Described acquisition module also for obtaining the multiple area of visual field images after pure color swap, extracts the effective coverage image after pure color swap from each area of visual field image;
Described processing module also for each exchange after effective coverage image process, defect recognition, recording defect position, and statistical shortcomings quantity;
Described synthesis module is used for the liquid crystal display screen image formed with the effective coverage exchanging rear display the first pure color by the effective coverage image of display the first pure color before exchange under the first pure color display state, and by the effective coverage image of display the second pure color before exchanging and after exchanging the effective coverage of display the second pure color form liquid crystal display screen image under the second pure color display state, defect type in liquid crystal display screen image under calculating the first pure color display state respectively and under the second pure color display state, recording defect position, and statistical shortcomings quantity.
7. High Precision Automatic Systems for optical inspection according to claim 6, is characterized in that, described processing module comprises:
Format conversion unit, for transferring the liquid crystal display screen image of acquisition to pre-set image form;
Pretreatment unit, for carrying out pre-service by the liquid crystal display screen image transferring pre-set image form to;
Threshold segmentation unit, for pretreated liquid crystal display screen image being carried out Threshold segmentation, wiping out background information, image information in extract crystal display screen image;
Image enhancing unit, for carrying out enhancing process to described image information;
Image deflects, for extracting image deflects from strengthening in the described image information after processing, being carried out Iamge Segmentation and being obtained defect block by image segmentation unit; And
Identification record unit, for identifying defect type in described defect block, recording defect position, and statistical shortcomings quantity.
8. High Precision Automatic Systems for optical inspection according to claim 7, is characterized in that, described system also comprises:
Divide module, in advance whole LCDs being divided into multiple effective coverage.
9. High Precision Automatic Systems for optical inspection according to claim 8, is characterized in that, described division module comprises:
Total pixel computing unit, is detected the physical resolution of LCDs for basis and represents total pixel that the camera pixel preset adopted needed for each physical picture element point is counted needed for calculating;
Pixel number calculating unit, for calculating the detected LCDs pixel number that each video camera can be taken according to resolution of video camera;
Area of visual field dimension calculating unit, for calculating each camera coverage area size;
Effective coverage dimension calculating unit, for according to area of visual field size and video camera displacement accuracy, calculates effective coverage size;
Division unit, for dividing whole detected LCDs according to effective coverage size.
10. High Precision Automatic Systems for optical inspection according to claim 9, is characterized in that, the required camera pixel preset adopted of each physical picture element point of described expression is counted as 6*6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310754700.5A CN104749801B (en) | 2013-12-31 | 2013-12-31 | High Precision Automatic optical detecting method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310754700.5A CN104749801B (en) | 2013-12-31 | 2013-12-31 | High Precision Automatic optical detecting method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104749801A true CN104749801A (en) | 2015-07-01 |
| CN104749801B CN104749801B (en) | 2019-09-20 |
Family
ID=53589737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310754700.5A Active CN104749801B (en) | 2013-12-31 | 2013-12-31 | High Precision Automatic optical detecting method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104749801B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108072443A (en) * | 2018-01-26 | 2018-05-25 | 上海康斐信息技术有限公司 | A kind of Weight-detecting device test method and system |
| CN109445134A (en) * | 2018-11-28 | 2019-03-08 | 深圳同兴达科技股份有限公司 | A kind of system and method for auto-measuring liquid crystal display die set optical data |
| CN110796974A (en) * | 2019-11-01 | 2020-02-14 | 深圳同兴达科技股份有限公司 | Display screen detection method |
| CN112394064A (en) * | 2020-10-22 | 2021-02-23 | 惠州高视科技有限公司 | Point-line measuring method for screen defect detection |
| CN114114734A (en) * | 2021-12-06 | 2022-03-01 | 苏州华兴源创科技股份有限公司 | Screen layering detection method, device, equipment, storage medium and system |
| CN114279683A (en) * | 2021-11-22 | 2022-04-05 | 浙江大华技术股份有限公司 | Automatic detection platform for display screen, method thereof and LED display screen |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1837900A (en) * | 2005-03-23 | 2006-09-27 | 三星电子株式会社 | Visual inspection apparatus and method of inspecting display panel using the visual inspection apparatus |
| CN102279189A (en) * | 2010-03-17 | 2011-12-14 | 夏普株式会社 | Defect detecting device, defect restoring device and defect detecting method |
| CN102514771A (en) * | 2011-10-27 | 2012-06-27 | 广东工业大学 | Industrial explosive roll transmission attitude identification and diagnosis system and method thereof |
| CN102520537A (en) * | 2011-12-02 | 2012-06-27 | 深圳市华星光电技术有限公司 | Automatic optical inspection method and automatic optical inspection equipment |
| US20130010239A1 (en) * | 2011-07-05 | 2013-01-10 | Jae-Hyun Park | Display substrate and method of repairing the display substrate |
-
2013
- 2013-12-31 CN CN201310754700.5A patent/CN104749801B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1837900A (en) * | 2005-03-23 | 2006-09-27 | 三星电子株式会社 | Visual inspection apparatus and method of inspecting display panel using the visual inspection apparatus |
| CN102279189A (en) * | 2010-03-17 | 2011-12-14 | 夏普株式会社 | Defect detecting device, defect restoring device and defect detecting method |
| US20130010239A1 (en) * | 2011-07-05 | 2013-01-10 | Jae-Hyun Park | Display substrate and method of repairing the display substrate |
| CN102514771A (en) * | 2011-10-27 | 2012-06-27 | 广东工业大学 | Industrial explosive roll transmission attitude identification and diagnosis system and method thereof |
| CN102520537A (en) * | 2011-12-02 | 2012-06-27 | 深圳市华星光电技术有限公司 | Automatic optical inspection method and automatic optical inspection equipment |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108072443A (en) * | 2018-01-26 | 2018-05-25 | 上海康斐信息技术有限公司 | A kind of Weight-detecting device test method and system |
| CN109445134A (en) * | 2018-11-28 | 2019-03-08 | 深圳同兴达科技股份有限公司 | A kind of system and method for auto-measuring liquid crystal display die set optical data |
| CN110796974A (en) * | 2019-11-01 | 2020-02-14 | 深圳同兴达科技股份有限公司 | Display screen detection method |
| CN112394064A (en) * | 2020-10-22 | 2021-02-23 | 惠州高视科技有限公司 | Point-line measuring method for screen defect detection |
| CN114279683A (en) * | 2021-11-22 | 2022-04-05 | 浙江大华技术股份有限公司 | Automatic detection platform for display screen, method thereof and LED display screen |
| CN114114734A (en) * | 2021-12-06 | 2022-03-01 | 苏州华兴源创科技股份有限公司 | Screen layering detection method, device, equipment, storage medium and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104749801B (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104749184A (en) | Automatic optical detection method and automatic optical detection system | |
| CN111612763B (en) | Mobile phone screen defect detection method, device and system, computer equipment and medium | |
| CN105913093B (en) | A kind of template matching method for Text region processing | |
| CN104749801A (en) | High-precision automatic optical detection method and high-precision automatic optical detection system | |
| CN107133969B (en) | A kind of mobile platform moving target detecting method based on background back projection | |
| CN107843599B (en) | AOI-based panel detection and judgment method and device | |
| CN102103753B (en) | Use method and the terminal of real time camera estimation detect and track Moving Objects | |
| CN107014819A (en) | A kind of solar panel surface defects detection system and method | |
| CN104616275A (en) | Defect detecting method and defect detecting device | |
| CN104484659A (en) | Method for automatically identifying and calibrating medical color images and medical gray scale images | |
| TWI549482B (en) | Detection Method for Camera | |
| CN107067595B (en) | State identification method and device of indicator light and electronic equipment | |
| CN107613229A (en) | A kind of dead pixels of image sensor surveys means for correcting and method | |
| CN116990993B (en) | LCD display panel quality detection method | |
| CN110567976A (en) | mobile phone cover plate silk-screen defect detection device and detection method based on machine vision | |
| KR102009740B1 (en) | Apparatus for inspecting of display panel and method thereof | |
| CN106228541A (en) | The method and device of screen location in vision-based detection | |
| CN105223208A (en) | A kind of circuit board detecting template and preparation method thereof, circuit board detecting method | |
| CN110648330A (en) | Defect detection method for camera glass | |
| CN110740314B (en) | Method and system for correcting defective pixel of color line array camera | |
| CN207283689U (en) | A kind of dead pixels of image sensor surveys means for correcting | |
| CN106093052B (en) | A kind of broken yarn detection method | |
| CN116503324A (en) | Black glue module appearance detection method and system based on machine vision | |
| CN106645174A (en) | Automatic online visual apparent-defect inspection system for general purpose engine | |
| CN114125435A (en) | Intelligent image defect detection method for network camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230710 Address after: 518000 1701, Yanxiang Science and Technology Building, No. 31, High-tech Middle 4th Road, Maling Community, Yuehai Street, Nanshan District, Shenzhen, Guangdong Province Patentee after: Yanxiang smart IOT Technology Co.,Ltd. Address before: 518057 Guangdong city of Shenzhen province Nanshan District high in the four EVOC Technology Building No. 31 Patentee before: EVOC INTELLIGENT TECHNOLOGY Co.,Ltd. |