CN109087281A - Display screen peripheral circuit detection method, device, electronic equipment and storage medium - Google Patents
Display screen peripheral circuit detection method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109087281A CN109087281A CN201810709408.4A CN201810709408A CN109087281A CN 109087281 A CN109087281 A CN 109087281A CN 201810709408 A CN201810709408 A CN 201810709408A CN 109087281 A CN109087281 A CN 109087281A
- Authority
- CN
- China
- Prior art keywords
- display screen
- peripheral circuit
- screen peripheral
- image
- defects detection
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 298
- 230000002093 peripheral effect Effects 0.000 title claims abstract description 252
- 238000003860 storage Methods 0.000 title claims abstract description 24
- 230000007547 defect Effects 0.000 claims abstract description 293
- 238000004519 manufacturing process Methods 0.000 claims abstract description 88
- 238000012360 testing method Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000012372 quality testing Methods 0.000 claims abstract description 24
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims description 55
- 238000012545 processing Methods 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 13
- 230000006378 damage Effects 0.000 claims description 10
- 230000002950 deficient Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000005520 cutting process Methods 0.000 claims description 7
- 238000010008 shearing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000003321 amplification Effects 0.000 claims 1
- 238000003199 nucleic acid amplification method Methods 0.000 claims 1
- 230000011218 segmentation Effects 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 13
- 238000007689 inspection Methods 0.000 description 12
- 238000013527 convolutional neural network Methods 0.000 description 6
- 230000005611 electricity Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000033772 system development Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Controls And Circuits For Display Device (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of display screen peripheral circuit detection method, device, electronic equipment and storage medium, it is requested by receiving the quality testing that the console being deployed on display screen peripheral circuit production line is sent, the display screen peripheral circuit image comprising the image capture device acquisition on the display screen peripheral circuit production line in the quality testing request；The display screen peripheral circuit image is zoomed in or out, obtain the consistent testing image of input size requirements of size Yu defects detection model, wherein, the defects detection model is to carry out semantic segmentation algorithm FCN training to historic defects display screen peripheral circuit image to obtain；The testing image is input in defects detection model and obtains defects detection result；The quality of the corresponding display screen peripheral circuit of the display screen peripheral circuit image is determined according to the defects detection result.The defects detection accuracy of the technical solution is high, system performance is good, and service expansion capability is high.
Description
Technical field
The present invention relates to defect detecting techniques more particularly to a kind of display screen peripheral circuit detection method, device, electronics to set
Standby and storage medium.
Background technique
With the development of science and technology, the effect of information display technology in people's lives is growing day by day, display screen is also because of its body
Product is small, light-weight, power is low, high resolution, brightness is high and is widely used without geometry deformation various features.But in display screen
In production process, it may cause display screen peripheral circuit existing defects due to technique and environment, for example, point class defect,
Foreign matter class defect and scratch class defect etc..Thus, the detection of display screen peripheral circuit is the important link in production process.
In the prior art, the artificial detection side that the detection of display screen peripheral circuit is mainly assisted using artificial detection or machine
Method.Sentence specifically, artificial detection method refers to rely on industry specialists and visually observe the collected picture from production environment and provide
It is disconnected；Machine auxiliary artificial detection method refer to first with solidify have the quality inspection system of industry specialists experience to it is to be detected show
Display screen peripheral circuit image is detected, and preliminary screening goes out the picture of doubtful existing defects, then by industry specialists to doubtful presence
The picture of defect carries out artificial detection judgement.
However, either artificial detection method or machine auxiliary artificial detection method by the subjective impact of people because
Element is larger, and accuracy in detection is low, system performance is poor, and service expansion capability is low.
Summary of the invention
The present invention provides a kind of display screen peripheral circuit detection method, device, electronic equipment and storage medium, existing to overcome
Have in display screen peripheral circuit defect inspection method since the Subjective Factors by people are larger, causes that accuracy in detection is low, is
The system problem that performance is poor, service expansion capability is low.
According to the first aspect of the invention, a kind of display screen peripheral circuit detection method is provided, comprising:
Receive the quality testing request that the console being deployed on display screen peripheral circuit production line is sent, the quality inspection
Survey the display screen peripheral circuit image comprising the image capture device acquisition on the display screen peripheral circuit production line in request；
The display screen peripheral circuit image is zoomed in or out, size is obtained and the input size of defects detection model is wanted
Seek consistent testing image, wherein the defects detection model is rolled up entirely with historic defects display screen peripheral circuit image
Product network FCN algorithm training obtains；
The testing image is input in defects detection model and obtains defects detection result；
The corresponding display screen peripheral circuit of the display screen peripheral circuit image is determined according to the defects detection result
Quality.
Optionally, in a kind of possible implementation of first aspect, described by the display screen peripheral circuit image
It is input in defects detection model before obtaining defects detection result, further includes:
With the actual pixels classification of historic defects display screen peripheral circuit image to described in defects detection model progress
FCN algorithm training, so that the prediction picture that the defects detection model exports the historic defects display screen peripheral circuit image
Penalty values between plain classification, with the actual pixels classification are lower than default loss threshold value.
Optionally, in the alternatively possible implementation of first aspect, described by the display screen peripheral circuit diagram
As before zooming in or out, further includes:
Image preprocessing is carried out to the display screen peripheral circuit image, wherein described image pretreatment includes following places
It is one or more in reason:
Cutting edge, shearing, rotation.
Optionally, described that the testing image is input to defect in another possible implementation of first aspect
Defects detection result is obtained in detection model, comprising:
According to load balancing, the detection model server of carrying process resource is determined；
The testing image is input to operate in the defects detection model on the detection model server and is obtained
To defects detection result.
Optionally, in another possible implementation of first aspect, the defects detection result, comprising: defect class
Other and/or defective patterns position；
It is described that the corresponding display screen periphery electricity of the display screen peripheral circuit image is determined according to the defects detection result
The quality on road, comprising:
According to production phase information and the defects detection as a result, determining that the display screen peripheral circuit image is corresponding
The quality of display screen peripheral circuit.
Optionally, in another possible implementation of first aspect, described true according to the defects detection result
After the quality for determining the corresponding display screen peripheral circuit of the display screen peripheral circuit image, further includes:
If it is determined that the display screen peripheral circuit is damage circuit, then following one or more operations are executed:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
By the display screen peripheral circuit image and the defects detection result be input in the defects detection model with
Just optimize the defects detection model.
Second aspect of the present invention provides a kind of display screen peripheral circuit detection device, comprising:
Receiving module is asked for receiving the quality testing that the console being deployed on display screen peripheral circuit production line is sent
It asks, the display screen comprising the image capture device acquisition on the display screen peripheral circuit production line in the quality testing request
Peripheral circuit image；
Preprocessing module obtains size and defects detection for zooming in or out the display screen peripheral circuit image
The consistent testing image of input size requirements of model, wherein the defects detection model is with historic defects display screen periphery
Circuit image carries out what full convolutional network FCN algorithm training obtained；
Processing module obtains defects detection result for the testing image to be input in defects detection model；
Determining module, for determining the corresponding display of the display screen peripheral circuit image according to the defects detection result
Shield the quality of peripheral circuit.
Optionally, in a kind of possible implementation of second aspect, the processing module, be also used to it is described will be described
Display screen peripheral circuit image is input to obtain defects detection result in defects detection model before, with historic defects display screen outside
The actual pixels classification for enclosing circuit image carries out the FCN algorithm training to the defects detection model, so that the defect is examined
Survey the prediction pixel classification that exports to the historic defects display screen peripheral circuit image of model, with the actual pixels classification it
Between penalty values lower than default loss threshold value.
Optionally, in the alternatively possible implementation of second aspect, the preprocessing module is also used to incite somebody to action described
The display screen peripheral circuit image carries out image preprocessing to the display screen peripheral circuit image before zooming in or out,
In, described image pretreatment includes one or more in following processing: cutting edge, shearing, rotation.
Optionally, in another possible implementation of second aspect, the processing module is specifically used for according to load
Balance policy determines the detection model server of carrying process resource；The testing image is input to and operates in the detection
Defects detection result is obtained in the defects detection model on model server.
Optionally, in another possible implementation of second aspect, the defects detection result, comprising: defect class
Other and/or defective patterns position；
The determining module is specifically used for according to production phase information and the defects detection as a result, determining described aobvious
The quality of the corresponding display screen peripheral circuit of display screen peripheral circuit image.
Optionally, in another possible implementation of second aspect, the processing module is also used in the basis
After the defects detection result determines the quality of the corresponding display screen peripheral circuit of the display screen peripheral circuit image,
If it is determined that the display screen peripheral circuit is damage circuit, then following one or more operations are executed:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
By the display screen peripheral circuit image and the defects detection result be input in the defects detection model with
Just optimize the defects detection model.
Third aspect present invention provides a kind of electronic equipment, including processor, memory and is stored on the memory
And the computer program that can be run on a processor, the processor realize when executing described program such as above-mentioned first aspect and
Method described in any one of various possible implementations of first aspect.
Fourth aspect present invention provides a kind of storage medium, and instruction is stored in the storage medium, when it is in computer
When upper operation, so that computer executes the side as described in any one of first aspect and the various possible implementations of first aspect
Method.
Display screen peripheral circuit detection method, device, electronic equipment and storage medium provided by the invention, pass through receiving unit
The quality testing request that the console on display screen peripheral circuit production line is sent is affixed one's name to, includes institute in the quality testing request
State the display screen peripheral circuit image of the image capture device acquisition on display screen peripheral circuit production line；It will be outside the display screen
It encloses circuit image to zoom in or out, obtains the consistent testing image of input size requirements of size Yu defects detection model, wherein
The defects detection model is to carry out semantic segmentation algorithm FCN training to historic defects display screen peripheral circuit image to obtain；
The testing image is input in defects detection model and obtains defects detection result；Institute is determined according to the defects detection result
State the quality of the corresponding display screen peripheral circuit of display screen peripheral circuit image.Since drawbacks described above detection model is to going through
History defect display screen peripheral circuit image carries out what FCN training obtained, thus, it is examined using the defect that the defects detection model obtains
The nicety of grading for surveying result is high, and intelligent ability is strong, and system performance increases, and business expandability is high, solves existing
Since the Subjective Factors by people are larger in display screen peripheral circuit defect inspection method, cause that accuracy in detection is low, system
The problem that performance is poor, service expansion capability is low.
Detailed description of the invention
Fig. 1 is a kind of structural schematic diagram of display screen peripheral circuit detection system provided in an embodiment of the present invention；
Fig. 2 is the flow diagram of display screen peripheral circuit detection method embodiment one provided in an embodiment of the present invention；
Fig. 3 is the flow diagram of display screen peripheral circuit detection method embodiment two provided in an embodiment of the present invention；
Fig. 4 is the structural schematic diagram of display screen peripheral circuit detection device embodiment provided in an embodiment of the present invention；
Fig. 5 is the structural schematic diagram of a kind of electronic equipment embodiment provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill
Personnel's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
It should be appreciated that in various embodiments of the present invention, the size of the serial number of each process is not meant to execute sequence
It is successive, the execution of each process sequence should be determined by its function and internal logic, the implementation without coping with the embodiment of the present invention
Journey constitutes any restriction.
It should be appreciated that in the present invention, " comprising " and " having " and their any deformation, it is intended that covering is not arranged
His includes, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to clearly
Those of list step or unit, but may include be not clearly listed or for these process, methods, product or equipment
Intrinsic other step or units.
It should be appreciated that in the present invention, " multiple " refer to two or more."and/or" is only a kind of description pass
Join object incidence relation, indicate may exist three kinds of relationships, for example, and/or B, can indicate: individualism A is existed simultaneously
These three situations of A and B, individualism B.Character "/" typicallys represent the relationship that forward-backward correlation object is a kind of "or".
It should be appreciated that in the present invention, " B corresponding with A ", " A and B are corresponding " or " B and A are corresponding " indicate B
It is associated with A, B can be determined according to A.Determine that B is not meant to determine B only according to A according to A, can also according to A and/or
Other information determines B.The matching of A and B is that the similarity of A and B is greater than or equal to preset threshold value.
Depending on context, as used in this " if " can be construed to " ... when " or " when ... " or
" in response to determination " or " in response to detection ".
At this stage, (3C industry refers to answers 3C industry in conjunction with the integration of computer, communication and the big sci-tech product of consumer electronics three
Information electrical appliance industry) overall intelligence the degree of automation it is lower, by the display screens peripheral circuit industry such as Mobile phone screen
Investigation and analysis is it is found that most of manufacturer can be divided into two kinds to the detection mode that Mobile phone screen uses, it may be assumed that artificial detection method
With the artificial detection method of machine auxiliary.
Wherein, artificial detection method refer to dependent on industry specialists visually observe from production environment acquired image into
Row judgement, this method by people Subjective Factors are larger, detection efficiency is lower and larger to the injury of human eye, further, since
The generation workshop of display screen peripheral circuit is generally dustfree environment, and staff needs to carry out cleaning preparation before entering, and dresses nothing
Dirt clothes is also possible to have an adverse effect to the health and safety of staff.
The artificial detection method of machine auxiliary is referred to as based on liquid crystal module detection device detection method, concrete principle
Are as follows: the image there is no defect is filtered out by the quality inspection system with certain judgement first, then by industry specialists to doubtful
The image of existing defects carries out detection judgement.Machine auxiliary artificial detection method in, quality inspection system be mostly expert system and
Feature Engineering System Development refers to that experience is solidificated in quality inspection system by expert, makes it have certain automatic capability.
Therefore, not only accuracy rate is low for the artificial detection method of machine auxiliary, and system performance is poor, can not cover all detection marks of manufacturer
Standard, and this method also low efficiency, are easy erroneous judgement of failing to judge, and the image data after detection is difficult to carry out secondary use excavation.This
Outside, in above-mentioned quality inspection system, the experience that feature and decision rule are all based on industry specialists is cured in machine, it is difficult to
The development iteration of business, leads to the development with production technology, and the detection accuracy of quality inspection system is lower and lower, in some instances it may even be possible to reduce
To complete unusable state.In addition, the feature of quality inspection system is all solidified by third-party vendor within hardware in advance, when upgrading
Not only need to carry out production line key technological transformation, but also it is expensive, safety, standardization, in terms of all
There is obvious deficiencies, are unfavorable for the optimization and upgrading of display screen peripheral circuit production line, and service expansion capability is low.
In conclusion the artificial detection method of artificial detection method and machine auxiliary has the following disadvantages: not only efficiency
Lowly, it is easy to appear erroneous judgement, and the industrial data that generates of both methods is not easy to store, manage and mining again recycles.
Latest development of the embodiment of the present invention based on artificial intelligence technology in computer vision, research and develop it is a kind of automation,
In high precision, the display screen peripheral circuit detection method of adaptive correction upgrading, using image capture device in display screen periphery electricity
The display screen peripheral circuit image acquired in real time on the production line of road, in real time detects the surface quality of display screen peripheral circuit
Judgement, if detecting that present image acquires the collected display screen peripheral circuit existing defects of equipment, it is determined that go out the defect
The classification of position and defect in picture.
Optionally, defect described in the embodiment of the present invention may include, but be not limited to include point class defect, foreign matter
The different classes of defect problem such as class defect and scratch class defect.Herein without introducing one by one.
It should be appreciated that in the present invention, full convolutional network (Fully Convolutional Network, referred to as: FCN)
Algorithm is by the full articulamentum in traditional convolutional neural networks (Convolutional Neural Network, referred to as: CNN)
It is converted to convolutional layer, compared with classical CNN network, last three layers full articulamentum are converted into three-layer coil lamination by FCN.It is passing through
In allusion quotation CNN structure, first 5 layers are convolutional layers, the 6th layer and the 7th layer be respectively a length be 4096 one-dimensional vector, the 8th layer is
The one-dimensional vector that length is 1000, respectively corresponds 1000 different classes of probability.FCN is converted to convolutional layer, convolution for this 3 layers
The size (port number, wide, height) of core is respectively (4096,1,1), (4096,1,1), (1000,1,1).It seems numerically not
There is any difference, but convolution is different concept and calculating process with connecting entirely, CNN has been trained before using
Weight and biasing, but it is different be weight and biasing be the range for having oneself, an one's own convolution kernel.Cause
All layers are all convolutional layers in this FCN network, therefore referred to as full convolutional network.
Technical solution of the present invention is described in detail with specifically embodiment below.These specific implementations below
Example can be combined with each other, and the same or similar concept or process may be repeated no more in some embodiments.
It is briefly described first below for the application scenarios that the embodiment of the present invention is applicable in.Referring to Fig. 1, for the present invention
A kind of structural schematic diagram for display screen peripheral circuit detection system that embodiment provides.Apply this in the system shown in figure 1
The display screen peripheral circuit detection method provided is provided, defects detection is carried out to display screen peripheral circuit.As shown in Figure 1, the display
Screen peripheral circuit detection system specifically includes that console 12, server group 13, controller 14, database 15, training aids 16 and portion
Affix one's name to the image capture device 11 on display screen peripheral circuit production line.
Wherein, image capture device 11 acquires the display screen peripheral circuit image on display screen peripheral circuit production line, control
Platform 12 processed receives the display screen peripheral circuit image that image capture device 11 acquires, and the display screen peripheral circuit image is sent
To the detection model server 130 in server group 13, display screen peripheral circuit diagram that detection model server 130 will receive
Defects detection is obtained as a result, controller 14 receives detection model server in defects detection model as being input to operation itself
130 defects detection as a result, and provide service response in conjunction with production phase information, controller 14 can also be by defects detection result
As log storage into database 15.In addition, the collected display screen peripheral circuit image of image capture device 11 can be with
It is directly stored in database 15, the initial data as defects detection model training.Training aids 16 extracts going through in database
History defect display screen peripheral circuit image is based on full convolutional network (Fully Convolutional Network, referred to as: FCN)
Algorithm training obtains defects detection model.
Optionally, above-mentioned database 15 may include Production database 151 and tranining database 152, Production database 151
It can receive and save defects detection result and the collected display screen of image capture device 11 periphery of the transmission of controller 14
Circuit image, tranining database 152 can store the historic defects display screen peripheral circuit image extracted from Production database 151
With corresponding original display screen peripheral circuit image so that training aids 16 training obtain the high defects detection mould of Detection accuracy
Type.
Optionally, the training aids 16 in the embodiment of the present invention can be is drawn by the training of hardware and/or software function realization
It holds up, the training tool as defects detection model.It can also be wrapped in the display screen peripheral circuit detection system of the embodiment of the present invention
Other entity modules such as processor, memory are included, the present embodiment is without being limited thereto.
Referring to fig. 2, illustrate for the process of display screen peripheral circuit detection method embodiment one provided in an embodiment of the present invention
Figure, the executing subject of method shown in Fig. 2 can be software service, be also possible to hardware device or software and mutually tie with hardware
The device of conjunction.It is specific as follows including step S101 to step S104:
S101 receives the quality testing request that the console being deployed on display screen peripheral circuit production line is sent, described
Include the display screen periphery electricity of the image capture device acquisition on the display screen peripheral circuit production line in quality testing request
Road image.
Optionally, in embodiments of the present invention, image capture device, control are deployed on display screen peripheral circuit production line
Multiple and different equipment such as platform, server group, controller, database.Image capture device can be high precision image acquisition and take the photograph
As head, in the production process of display screen peripheral circuit, by adjusting the angle of image capture device, light, filter, times mirror,
Focus etc., the corresponding display screen peripheral circuit diagram of display screen peripheral circuit that multiple are in production process can be collected
Picture.
After the image capture device on display screen peripheral circuit production line collects display screen peripheral circuit image, portion
Administration then can dispose defective inspection on display screen peripheral circuit production line in the console on display screen peripheral circuit production line
The server group for surveying model sends quality testing request, includes the aobvious of above-mentioned image capture device acquisition in quality testing request
Display screen peripheral circuit image so that received in server group the quality testing request server to the display screen received outside
Circuit image is enclosed to be handled.
The display screen peripheral circuit image is zoomed in or out, obtains the input of size Yu defects detection model by S102
The consistent testing image of size requirements.
The testing image is input in defects detection model and obtains defects detection result by S103.
Wherein, the defects detection model is to carry out full convolutional network FCN with historic defects display screen peripheral circuit image
Algorithm training obtains.There are size requirements to the image of input by the defects detection model that the training of FCN algorithm obtains, once input
The size of image requires not of uniform size with mode input, and defects detection model will be unable to handle it.To display screen
When peripheral circuit is detected, line alignment that global image is illustrated, coiling shape can more show its defect that may be present
Problem, therefore processing first is zoomed in and out to display screen peripheral circuit figure before inputting defects detection model in the present embodiment,
So that the size of testing image is consistent with the input size requirements of defects detection model.
Display screen peripheral circuit image is zoomed in or out, it can be understood as pixel is constant to be zoomed in or out, can also be with
It is interpreted as zooming in or out for pixel reduction.Pixel is excessively high to exceed the processing capacity of defects detection model, therefore is showing
In the case that screen peripheral circuit image pixel is excessively high, drop processes pixel first can also be carried out to display screen peripheral circuit image,
This is without limitation.
Optionally, receive quality testing request server by quality testing request in display screen peripheral circuit image
Acquisition comes out, and the pretreatment zoomed in or out, obtains size and the input size requirements of defects detection model are consistent
Testing image.Then testing image is input on server in running defects detection model, is held by defects detection model
Row defects detection, and then obtain defects detection result.
In one implementation, it is described the display screen peripheral circuit image is zoomed in or out before, can be with
Image preprocessing is carried out to the display screen peripheral circuit image, wherein described image pretreatment includes one in following processing
Item is multinomial: cutting edge, shearing, rotation.It is to be understood that being deployed in the image capture device on display screen peripheral circuit production line
Usually high-precision camera, thus, it may size using the collected display screen peripheral circuit image of the image capture device
Larger or pixel is higher or position is improper etc..Thus, when receive console transmission asked included in quality testing
After the display screen peripheral circuit image asked, need according to the actual situation to pre-process display screen peripheral circuit image.
For example, if the fringe region of display screen peripheral circuit image is larger, at this point it is possible to carry out cutting edge to display screen peripheral circuit image
Processing retains the useful part of display screen peripheral circuit image.
It is worth noting that the defects detection model run on server is to historic defects display screen peripheral circuit image
Carry out what full convolutional network FCN algorithm training obtained.Specifically, the present embodiment is to carry out image, semantic segmentation with FCN algorithm.
Image, semantic segmentation, which refers to, allows computer to be split according to the semanteme of image.Semanteme refers to image in image domains
Content, the different types of object for being meant that be partitioned into from the angle of pixel in picture of segmentation, to each picture in original image
Element has all carried out type mark.Such as each pixel is noted as normal pixel, first kind defect pixel, the second class defect pixel
In one.Thus, in embodiments of the present invention, defects detection model is by a large amount of historic defects display screen peripheral circuits
Image carries out image, semantic segmentation training and obtains.Specifically, i.e., to normal pixel in historic defects display screen peripheral circuit image
It is all marked with defect pixel, and the label of defect pixel also distinguishes its defect type.Then it is marked with pixel each in this way
The historic defects display screen peripheral circuit image of note carries out the identification of FCN algorithm as training set and test set, according to training mould
The output of type and the type of pixel of label obtain loss function, and loss function is used to carry out backpropagation to model, to model into
Row training and study preset loss threshold value until the corresponding penalty values of loss function are lower than, complete training, thus obtain defect and examine
Survey model.
In embodiments of the present invention, defects detection model uses FCN structure.Specifically, display screen peripheral circuit production line
On input of the display screen peripheral circuit image as defects detection model, it is aobvious using the FCN structure recognition of defects detection model
In display screen peripheral circuit image the feature of each pixel to get into display screen peripheral circuit image which pixel be normal
Pixel, which pixel are defect pixel points.
As an example, it is lacked in the described display screen peripheral circuit image is input in defects detection model
It falls into before testing result, can also include model training process.Specifically, it can be with historic defects display screen peripheral circuit diagram
The actual pixels classification of picture carries out the FCN algorithm training to the defects detection model, so that the defects detection model pair
The prediction pixel classification of the historic defects display screen peripheral circuit image output, with the loss between the actual pixels classification
Value is lower than default loss threshold value.
In FCN model, all layers are realized with convolution, it can be understood as all layers are all convolutional layers.The present invention
Embodiment can use this model operated with convolution, Chi Hua, deconvolution up-sampling, to display screen peripheral circuit production line
On had by features such as deformation, the fuzzy, illumination variations of the collected display screen peripheral circuit image of image capture device it is higher
Robustness, for classification task have it is higher can generalization.
It is worth noting that in embodiments of the present invention, for different production scenes and display screen peripheral circuit image
The characteristics of, train the organizational form of FCN model required for drawbacks described above detection model may be different, it can be according to reality
Situation is determined, and the present embodiment is simultaneously not limited thereof.
S104 determines the corresponding display screen periphery electricity of the display screen peripheral circuit image according to the defects detection result
The quality on road.
It optionally, in an embodiment of the present invention, can be with after obtaining defects detection result according to defects detection model
The quality of the corresponding display screen peripheral circuit of above-mentioned display screen peripheral circuit image is determined according to the defects detection result.
Optionally, in one embodiment of this invention, drawbacks described above testing result may include: defect classification and/or lack
Fall into graph position.For example, when existing defects in display screen peripheral circuit image, the defects detection model it can be concluded that defect
It may include defect classification (a few class defects are co-existed on display screen peripheral circuit), defective locations (each defect in testing result
Specific location).The presentation mode of defects detection result it is to be understood that defects detection model exports segmentation figure, segmentation figure with
First color identifier is normal pixel, the second color identifier first kind defect, second color identifier the second class defect.It is detecting
In the defects detection result of two class defects, such as can be a white is background color, and includes blue color lump and green color lump
Segmentation figure, wherein white represents the pixel of normal region, blue represents the pixel of point class defect area, and green represents foreign matter class and lacks
Sunken pixel.FCN model is pixel identification, therefore from the available all kinds of defective patterns of defects detection result, it is possible to understand that
For chamfered shape and its pixel position in display screen peripheral circuit image for being all kinds of defects.
Correspondingly, S104 (determines the corresponding display of the display screen peripheral circuit image according to the defects detection result
Shield the quality of peripheral circuit) it could alternatively be: according to production phase information and the defects detection as a result, described in determining
The quality of the corresponding display screen peripheral circuit of display screen peripheral circuit image.
Specifically, a variety of different production phases such as the manufacturer of display screen peripheral circuit, production environment and type
Information may obtain different defects detection results in display screen peripheral circuit detection process.For different types of display
Shield peripheral circuit, the production phase of its experience of institute is different, thus, when analyzing defects detection result obtained above, need
The quality to determine display screen peripheral circuit is carried out in conjunction with the production phase information of each display screen peripheral circuit.
It is worth noting that the defects detection model of the embodiment of the present invention can detecte out in display screen peripheral circuit image
There are a few class defect types, and for the specific number of every class defect, the embodiment of the present invention is not limited, that is, using FCN algorithm
Obtained defects detection model does not distinguish the different defects individual for belonging to the same category.
Display screen peripheral circuit detection method provided in an embodiment of the present invention is deployed in display screen peripheral circuit by receiving
The quality testing that console on production line is sent is requested, comprising on display screen peripheral circuit production line in quality testing request
Image capture device acquisition display screen peripheral circuit image, display screen peripheral circuit image is input to defects detection model
In obtain defects detection as a result, and determining display screen peripheral circuit image corresponding display screen periphery according to the defects detection result
The quality of circuit.Since drawbacks described above detection model is to carry out FCN algorithm to historic defects display screen peripheral circuit image
What training obtained, thus, utilize the nicety of grading height for the defects detection result that the defects detection model obtains, intelligent ability
By force, system performance increases, and business expandability is high.
It is that the process of display screen peripheral circuit detection method embodiment two provided in an embodiment of the present invention is illustrated referring to Fig. 3
Figure.On the basis of the above embodiments, in embodiment illustrated in fig. 3, the testing image (is input to defects detection by above-mentioned S104
Defects detection result is obtained in model) it can be realized by step S301-S302, specific as follows:
S301 determines the detection model server of carrying process resource according to load balancing.
Optionally, in an embodiment of the present invention, a server group is deployed on display screen peripheral circuit production line, it should
Number of servers in server group can be it is multiple, defects detection model is run on each server.Optionally, each
The defects detection model run on server be it is identical, therefore, each server can receive console send quality
Detection request, and then the defects detection model that can use itself carrying carries out quality testing to display screen peripheral circuit image.
As an example, it is shown since the image capture device being deployed on display screen peripheral circuit production line acquires in real time
Display screen peripheral circuit image, thus, any server from console to server group can also send quality testing and ask in real time
It asks.
Optionally, due to the defects detection model run on each server in server group be it is identical, thus, in order to
The detection efficiency of the defects detection model on server is improved, guarantees the load balancing of defects detection model, it can be according to preparatory
The load balancing of setting determines detection model service of the server as carrying process resource from server group
Device, i.e., according to the deployment scenario real-time perfoming load balancing and scheduling of defects detection model on display screen peripheral circuit production line.
The testing image is input to the defects detection model operated on the detection model server by S302
In obtain defects detection result.
Optionally, in embodiments of the present invention, when the detection model clothes for determining carrying process resource from server group
It is engaged in after device, above-mentioned display screen peripheral circuit image can be input to the defects detection run on the detection model server
In model, detected using defect of the defects detection model to display screen peripheral circuit image, and then obtain defects detection
As a result.Optionally, which is by training module to the default picture in historic defects display screen peripheral circuit image
What plain classification and actual pixels classification were trained.
Display screen peripheral circuit detection method provided in an embodiment of the present invention, by the way that according to load balancing, determination is held
The detection model server of process resource is carried, and above-mentioned testing image is input to and is operated on above-mentioned detection model server
Defects detection model in obtain defects detection as a result, it is possible to realize the load balancing on server, improve display screen periphery electricity
The detection efficiency of road image promotes the performance of display screen peripheral circuit detection system.
In one implementation, in above-mentioned steps S302 (testing image is input to and operates in the detection mould
Defects detection result is obtained in the defects detection model on type server) after, it can also include the following steps:
If it is determined that the display screen peripheral circuit is damage circuit, then following one or more operations are executed:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
By the display screen peripheral circuit image and the defects detection result be input in the defects detection model with
Just optimize the defects detection model.
Optionally, in embodiments of the present invention, tester can be according to the production scene of display screen peripheral circuit and life
Session information is produced, the solution when determining that display screen peripheral circuit is bad screen is preset, for example, passing through controller to life
It produces manager and sends warning message, and/or, it is stored drawbacks described above testing result as log to creation data by controller
In library, and/or, production control instruction is sent to console to eliminate defect by controller, and/or, by above-mentioned display screen
Peripheral circuit image and drawbacks described above testing result are input in drawbacks described above detection model to optimize drawbacks described above detection mould
Type etc..
Specifically, as an example, when determining display screen peripheral circuit image pair according to drawbacks described above testing result
The display screen peripheral circuit answered is damage circuit, i.e., in display screen peripheral circuit when existing defects, can with alert, with
So that production manager is positioned classification and the position of defect in time, and provides solution.
As another example, when determining existing defects in display screen peripheral circuit according to drawbacks described above testing result,
It can be stored drawbacks described above testing result as log into Production database by controller, i.e., by display screen peripheral circuit
Defect classification, and/or, defective patterns position is stored as log into Production database, and then can be screened instruction
Practice in database, by training module (can be the software programs such as trained engine) according to the display screen peripheral circuit diagram of existing defects
As updating drawbacks described above detection model.
As another example, when determining existing defects in display screen peripheral circuit according to drawbacks described above testing result,
Production control instruction can also be sent to console to eliminate defect by controller.That is, the inspection of carrying defects detection model
The reason of defect occurs can be determined by controller by surveying model server, and then adjust production procedure according to corresponding,
Occur on display screen peripheral circuit that is, detection model server sends production control instruction to console by controller to eliminate
Defect, with reduce damage circuit occur probability.
As another example, when determining existing defects in display screen peripheral circuit according to drawbacks described above testing result,
Directly above-mentioned display screen peripheral circuit image and drawbacks described above testing result can also be input in drawbacks described above detection model
To optimize drawbacks described above detection model, i.e., the corresponding display screen peripheral circuit image of circuit will be directly damaged as defects detection
The training set of model to optimize the defects detection model, and then improves the accuracy in detection of defects detection model.
It is worth noting that detection when the embodiment of the present invention is not limited to determining display screen peripheral circuit to damage circuit
The executable above-mentioned one or more operations of model server, can be determined, details are not described herein again according to the actual situation.
Optionally, for disposed on display screen peripheral circuit production line image capture device, console, server group,
Multiple and different equipment such as controller, database, can also be by the corresponding operating procedure point of display screen peripheral circuit detection method
Above-mentioned multiple and different equipment is scattered to execute.For example, image capture device acquires display screen peripheral circuit image, console root
According to load balancing, the collected display screen peripheral circuit image of image capture device is sent to the detection in server group
Model server carries out display screen peripheral circuit image by the defects detection model run on detection model server preset
Defects detection is carried out after pretreatment, and provides defects detection result.Detection model server can send out defects detection result
Controller is given, on the one hand by controller combination practical business scene, and according to business demand, according to drawbacks described above testing result
Make the response for meeting practical business scene requirement, such as alarm, storage log, control production control instruction, on the other hand, control
Device processed can also be stored the processing behavior of defects detection result and above-mentioned response as log into Production database, so that instruction
Practice module according in tranining database display screen peripheral circuit image and defects detection result update defect obtained above and examine
Model is surveyed, what is stored in the tranining database is the defective display screen peripheral circuit image of tool screened from Production database
With the data such as corresponding defects detection result.
It is worth noting that can gradually be taken by the online mode of small flow for the defects detection model of each suboptimization
The defects detection model that generation is just running on the server, it is dynamic with business scenario and production phase information to reach defects detection model
State extends extensive purpose.When in the embodiment of the present invention display screen peripheral circuit detection method in display screen peripheral circuit production line
After upper operation a period of time, it can manually pass through the information in Production database, the detection of check drawbacks described above and defect location
Then accuracy rate updates above-mentioned tranining database, re -training defects detection model, to improve defects detection accuracy rate.
Following is apparatus of the present invention embodiment, can be used for executing embodiment of the present invention method.For apparatus of the present invention reality
Undisclosed details in example is applied, embodiment of the present invention method is please referred to.
It referring to fig. 4, is the structural schematic diagram of display screen peripheral circuit detection device embodiment provided in an embodiment of the present invention.
As shown in figure 4, display screen peripheral circuit detection device provided in an embodiment of the present invention can specifically include that receiving module 41, pre-
Processing module 42, processing module 43 and determining module 44.
Wherein, receiving module 41, the matter sent for receiving the console being deployed on display screen peripheral circuit production line
Amount detection is requested, and the image capture device acquisition on the display screen peripheral circuit production line is included in the quality testing request
Display screen peripheral circuit image.
Preprocessing module 42 obtains size and defect is examined for zooming in or out the display screen peripheral circuit image
Survey the consistent testing image of input size requirements of model, wherein the defects detection model is with outside historic defects display screen
It encloses circuit image and carries out what full convolutional network FCN algorithm training obtained.
Processing module 43 obtains defects detection result for the testing image to be input in defects detection model.
Determining module 44, for determining that the display screen peripheral circuit image is corresponding aobvious according to the defects detection result
The quality of display screen peripheral circuit.
The display screen peripheral circuit detection device of embodiment illustrated in fig. 4 accordingly can be used for executing the implementation of method shown in Fig. 2
Step in example, it is similar that the realization principle and technical effect are similar, and details are not described herein again.
Optionally, the processing module 43 is also used to that the display screen peripheral circuit image is input to defect described
Before obtaining defects detection result in detection model, with the actual pixels classification of historic defects display screen peripheral circuit image to institute
State defects detection model and carry out FCN algorithm training so that the defects detection model to the historic defects display screen outside
The prediction pixel classification for enclosing circuit image output, the penalty values between the actual pixels classification are lower than default loss threshold value.
Optionally, the preprocessing module 42 is also used to that the display screen peripheral circuit image is amplified or contracted described
Image preprocessing is carried out to the display screen peripheral circuit image before small, wherein described image pretreatment includes following processing
In it is one or more: cutting edge, shearing, rotation.
Optionally, the processing module 43 is specifically used for determining the detection of carrying process resource according to load balancing
Model server；The testing image is input to and is operated in the defects detection model on the detection model server
Obtain defects detection result.
Optionally, the defects detection result, comprising: defect classification and/or defective patterns position.
The determining module 44 is specifically used for according to production phase information and the defects detection as a result, described in determining
The quality of the corresponding display screen peripheral circuit of display screen peripheral circuit image.
Optionally, the processing module 43 is also used to determine the display screen according to the defects detection result described
After the quality of the corresponding display screen peripheral circuit of peripheral circuit image, however, it is determined that the display screen peripheral circuit is damage
Circuit then executes following one or more operations:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
By the display screen peripheral circuit image and the defects detection result be input in the defects detection model with
Just optimize the defects detection model.
The display screen peripheral circuit detection device of above-mentioned apparatus embodiment can be used for executing the implementation of method shown in Fig. 2 to Fig. 3
The implementation of example, specific implementation is similar with technical effect, and which is not described herein again.
It is the structural schematic diagram of a kind of electronic equipment embodiment provided in an embodiment of the present invention, the electronic equipment referring to Fig. 5
It include: processor 51, memory 52 and computer program；Wherein
Memory 52, for storing the computer program, which can also be flash memory (flash).The calculating
Machine program is, for example, to realize application program, the functional module etc. of the above method.
Processor 51, for executing the computer program of the memory storage, to realize electronic equipment in the above method
The each step executed.It specifically may refer to the associated description in previous methods embodiment.
Optionally, memory 52 can also be integrated with processor 51 either independent.
When the memory 52 is independently of the device except processor 51, the electronic equipment can also include:
Bus 53, for connecting the memory 52 and processor 51.
The present invention also provides a kind of storage medium, instruction is stored in the storage medium, when it runs on computers
When, so that the method that computer executes embodiment of the method as shown in Figure 2 to Figure 3.
Wherein, storage medium can be computer storage medium, be also possible to communication media.Communication media include convenient for from
Any medium of one place to another place transmission computer program.Computer storage medium can be general or specialized meter
Any usable medium that calculation machine can access.For example, storage medium is coupled to processor, to enable a processor to readable from this
Read information, and information can be written to the storage medium.Certainly, storage medium is also possible to the composition portion of processor
Point.Pocessor and storage media can be located at specific integrated circuit (Application Specific Integrated
Circuits, referred to as: ASIC) in.In addition, the ASIC can be located in user equipment.Certainly, pocessor and storage media can also
To be present in communication equipment as discrete assembly.
The present invention also provides a kind of program product, which includes computer program, which is stored in
In storage medium.At least one processor of display screen peripheral circuit detection device can read the computer journey from storage medium
Sequence, at least one processor execute the computer program and display screen peripheral circuit detection device are made to execute side shown in Fig. 2 to Fig. 3
The method of method embodiment.
In the embodiment of above-mentioned electronic equipment, it should be appreciated that processor can be central processing unit (English: Central
Processing Unit, referred to as: CPU), it can also be other general processors, digital signal processor (English: Digital
Signal Processor, referred to as: DSP), specific integrated circuit (English: Application Specific Integrated
Circuit, referred to as: ASIC) etc..General processor can be microprocessor or the processor is also possible to any conventional place
Manage device etc..It can be embodied directly in hardware processor in conjunction with the step of the method disclosed in the present and execute completion or use
Hardware and software module combination in reason device execute completion.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (14)
1. a kind of display screen peripheral circuit detection method characterized by comprising
The quality testing request that the console being deployed on display screen peripheral circuit production line is sent is received, the quality testing is asked
Seek the display screen peripheral circuit image comprising the image capture device acquisition on the display screen peripheral circuit production line；
The display screen peripheral circuit image is zoomed in or out, the input size requirements one of size Yu defects detection model are obtained
The testing image of cause, wherein the defects detection model is to carry out full convolution net with historic defects display screen peripheral circuit image
The training of network FCN algorithm obtains；
The testing image is input in defects detection model and obtains defects detection result；
The quality of the corresponding display screen peripheral circuit of the display screen peripheral circuit image is determined according to the defects detection result
Quality.
2. the method according to claim 1, wherein the display screen peripheral circuit image is input to described
Before obtaining defects detection result in defects detection model, further includes:
The FCN calculation is carried out to the defects detection model with the actual pixels classification of historic defects display screen peripheral circuit image
Method training, so that the prediction pixel class that the defects detection model exports the historic defects display screen peripheral circuit image
Not, the penalty values between the actual pixels classification are lower than default loss threshold value.
3. the method according to claim 1, wherein it is described by the display screen peripheral circuit image amplification or
Before diminution, further includes:
Image preprocessing is carried out to the display screen peripheral circuit image, wherein described image pretreatment includes in following processing
It is one or more:
Cutting edge, shearing, rotation.
4. method according to claim 1-3, which is characterized in that described that the testing image is input to defect
Defects detection result is obtained in detection model, comprising:
According to load balancing, the detection model server of carrying process resource is determined；
The testing image is input to operate in the defects detection model on the detection model server and is lacked
Fall into testing result.
5. method according to claim 1-3, which is characterized in that the defects detection result, comprising: defect class
Other and/or defective patterns position；
It is described that the corresponding display screen peripheral circuit of the display screen peripheral circuit image is determined according to the defects detection result
Quality, comprising:
According to production phase information and the defects detection as a result, determining the corresponding display of the display screen peripheral circuit image
Shield the quality of peripheral circuit.
6. method according to claim 1-3, which is characterized in that described true according to the defects detection result
After the quality for determining the corresponding display screen peripheral circuit of the display screen peripheral circuit image, further includes:
If it is determined that the display screen peripheral circuit is damage circuit, then following one or more operations are executed:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
The display screen peripheral circuit image and the defects detection result are input in the defects detection model so as to excellent
Change the defects detection model.
7. a kind of display screen peripheral circuit detection device characterized by comprising
Receiving module, the quality testing request sent for receiving the console being deployed on display screen peripheral circuit production line,
Outside display screen comprising the image capture device acquisition on the display screen peripheral circuit production line in the quality testing request
Enclose circuit image；
Preprocessing module obtains size and defects detection model for zooming in or out the display screen peripheral circuit image
The consistent testing image of input size requirements, wherein the defects detection model is with historic defects display screen peripheral circuit
Image carries out what full convolutional network FCN algorithm training obtained；
Processing module obtains defects detection result for the testing image to be input in defects detection model；
Determining module, for being determined outside the corresponding display screen of the display screen peripheral circuit image according to the defects detection result
Enclose the quality of circuit.
8. device according to claim 7, which is characterized in that
The processing module is also used to obtain in the described display screen peripheral circuit image is input in defects detection model
Before defects detection result, with the actual pixels classification of historic defects display screen peripheral circuit image to the defects detection model
The FCN algorithm training is carried out, so that the defects detection model exports the historic defects display screen peripheral circuit image
Prediction pixel classification, penalty values between the actual pixels classification are lower than default loss threshold value.
9. device according to claim 7, which is characterized in that
The preprocessing module, be also used to it is described the display screen peripheral circuit image is zoomed in or out before to described aobvious
Display screen peripheral circuit image carries out image preprocessing, wherein described image pretreatment includes one or more in following processing:
Cutting edge, shearing, rotation.
10. according to the described in any item devices of claim 7-9, which is characterized in that
The processing module is specifically used for determining the detection model server of carrying process resource according to load balancing；It will
The testing image, which is input to operate in the defects detection model on the detection model server, obtains defects detection
As a result.
11. according to the described in any item devices of claim 7-9, which is characterized in that the defects detection result, comprising: defect
Classification and/or defective patterns position；
The determining module is specifically used for according to production phase information and the defects detection as a result, determining the display screen
The quality of the corresponding display screen peripheral circuit of peripheral circuit image.
12. according to the described in any item devices of claim 7-9, which is characterized in that the processing module is also used at described
According to the defects detection result determine the corresponding display screen peripheral circuit of the display screen peripheral circuit image quality it
Afterwards, however, it is determined that the display screen peripheral circuit is damage circuit, then executes following one or more operations:
Warning message is sent to production manager by controller；
It is stored the defects detection result as log into Production database by controller；
Production control instruction is sent to eliminate defect to the console by controller；
The display screen peripheral circuit image and the defects detection result are input in the defects detection model so as to excellent
Change the defects detection model.
13. a kind of electronic equipment, including processor, memory and it is stored on the memory and can runs on a processor
Computer program, which is characterized in that the processor is realized as described in the claims any one of 1-6 when executing described program
Method.
14. a kind of storage medium, which is characterized in that instruction is stored in the storage medium, when run on a computer,
So that computer executes as the method according to claim 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810709408.4A CN109087281A (en) | 2018-07-02 | 2018-07-02 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810709408.4A CN109087281A (en) | 2018-07-02 | 2018-07-02 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109087281A true CN109087281A (en) | 2018-12-25 |
Family
ID=64836933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810709408.4A Pending CN109087281A (en) | 2018-07-02 | 2018-07-02 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109087281A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886964A (en) * | 2019-03-29 | 2019-06-14 | 北京百度网讯科技有限公司 | Circuit board defect detection method, device and equipment |
| CN111738338A (en) * | 2020-06-23 | 2020-10-02 | 征图新视(江苏)科技股份有限公司 | Defect detection method applied to motor coil based on cascaded expansion FCN network |
| CN113127272A (en) * | 2019-12-31 | 2021-07-16 | 华为技术有限公司 | Screen detection method and electronic equipment for screen detection |
| CN113409290A (en) * | 2021-06-29 | 2021-09-17 | 北京兆维电子(集团)有限责任公司 | Method and device for detecting appearance defects of liquid crystal display and storage medium |
| CN116416884A (en) * | 2023-06-12 | 2023-07-11 | 深圳市彤兴电子有限公司 | Testing device and testing method for display module |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118044A (en) * | 2015-06-16 | 2015-12-02 | 华南理工大学 | Method for automatically detecting defects of wheel-shaped cast product |
| CN106934800A (en) * | 2017-02-28 | 2017-07-07 | 北京科技大学 | A kind of metal plate and belt detection method of surface flaw and device based on YOLO9000 networks |
| CN107016405A (en) * | 2017-02-24 | 2017-08-04 | 中国科学院合肥物质科学研究院 | A kind of insect image classification method based on classification prediction convolutional neural networks |
| CN107392896A (en) * | 2017-07-14 | 2017-11-24 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of Wood Defects Testing method and system based on deep learning |
| CN107564002A (en) * | 2017-09-14 | 2018-01-09 | 广东工业大学 | Plastic tube detection method of surface flaw, system and computer-readable recording medium |
| CN107561738A (en) * | 2017-08-30 | 2018-01-09 | 湖南理工学院 | TFT LCD surface defect quick determination methods based on FCN |
| CN107607554A (en) * | 2017-09-26 | 2018-01-19 | 天津工业大学 | A kind of Defect Detection and sorting technique of the zinc-plated stamping parts based on full convolutional neural networks |
| CN107966447A (en) * | 2017-11-14 | 2018-04-27 | 浙江大学 | A kind of Surface Flaw Detection method based on convolutional neural networks |
| CN108038846A (en) * | 2017-12-04 | 2018-05-15 | 国网山东省电力公司电力科学研究院 | Transmission line equipment image defect detection method and system based on multilayer convolutional neural networks |
| CN108230317A (en) * | 2018-01-09 | 2018-06-29 | 北京百度网讯科技有限公司 | Steel plate defect detection sorting technique, device, equipment and computer-readable medium |
| CN108230318A (en) * | 2018-01-09 | 2018-06-29 | 北京百度网讯科技有限公司 | Ladle defects detection sorting technique, device, equipment and computer-readable medium |
-
2018
- 2018-07-02 CN CN201810709408.4A patent/CN109087281A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118044A (en) * | 2015-06-16 | 2015-12-02 | 华南理工大学 | Method for automatically detecting defects of wheel-shaped cast product |
| CN107016405A (en) * | 2017-02-24 | 2017-08-04 | 中国科学院合肥物质科学研究院 | A kind of insect image classification method based on classification prediction convolutional neural networks |
| CN106934800A (en) * | 2017-02-28 | 2017-07-07 | 北京科技大学 | A kind of metal plate and belt detection method of surface flaw and device based on YOLO9000 networks |
| CN107392896A (en) * | 2017-07-14 | 2017-11-24 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of Wood Defects Testing method and system based on deep learning |
| CN107561738A (en) * | 2017-08-30 | 2018-01-09 | 湖南理工学院 | TFT LCD surface defect quick determination methods based on FCN |
| CN107564002A (en) * | 2017-09-14 | 2018-01-09 | 广东工业大学 | Plastic tube detection method of surface flaw, system and computer-readable recording medium |
| CN107607554A (en) * | 2017-09-26 | 2018-01-19 | 天津工业大学 | A kind of Defect Detection and sorting technique of the zinc-plated stamping parts based on full convolutional neural networks |
| CN107966447A (en) * | 2017-11-14 | 2018-04-27 | 浙江大学 | A kind of Surface Flaw Detection method based on convolutional neural networks |
| CN108038846A (en) * | 2017-12-04 | 2018-05-15 | 国网山东省电力公司电力科学研究院 | Transmission line equipment image defect detection method and system based on multilayer convolutional neural networks |
| CN108230317A (en) * | 2018-01-09 | 2018-06-29 | 北京百度网讯科技有限公司 | Steel plate defect detection sorting technique, device, equipment and computer-readable medium |
| CN108230318A (en) * | 2018-01-09 | 2018-06-29 | 北京百度网讯科技有限公司 | Ladle defects detection sorting technique, device, equipment and computer-readable medium |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886964A (en) * | 2019-03-29 | 2019-06-14 | 北京百度网讯科技有限公司 | Circuit board defect detection method, device and equipment |
| CN113127272A (en) * | 2019-12-31 | 2021-07-16 | 华为技术有限公司 | Screen detection method and electronic equipment for screen detection |
| CN113127272B (en) * | 2019-12-31 | 2024-05-14 | 华为技术有限公司 | Screen detection method and screen detection electronic equipment |
| CN111738338A (en) * | 2020-06-23 | 2020-10-02 | 征图新视(江苏)科技股份有限公司 | Defect detection method applied to motor coil based on cascaded expansion FCN network |
| CN113409290A (en) * | 2021-06-29 | 2021-09-17 | 北京兆维电子(集团)有限责任公司 | Method and device for detecting appearance defects of liquid crystal display and storage medium |
| CN113409290B (en) * | 2021-06-29 | 2023-12-15 | 北京兆维电子(集团)有限责任公司 | Method and device for detecting appearance defects of liquid crystal display, and storage medium |
| CN116416884A (en) * | 2023-06-12 | 2023-07-11 | 深圳市彤兴电子有限公司 | Testing device and testing method for display module |
| CN116416884B (en) * | 2023-06-12 | 2023-08-18 | 深圳市彤兴电子有限公司 | Testing device and testing method for display module |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109085174A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| US11380232B2 (en) | Display screen quality detection method, apparatus, electronic device and storage medium | |
| KR102229594B1 (en) | Display screen quality detection method, device, electronic device and storage medium | |
| CN109087281A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| CN108921840A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| EP3937128A2 (en) | Image defect detection method and apparatus, electronic device, storage medium and product | |
| EP3520045B1 (en) | Image-based vehicle loss assessment method, apparatus, and system, and electronic device | |
| CN108564104A (en) | Product defects detection method, device, system, server and storage medium | |
| CN108921839A (en) | Continuous casting billet quality detection method, device, electronic equipment and storage medium | |
| CN109064446A (en) | Display screen quality determining method, device, electronic equipment and storage medium | |
| CN108961239A (en) | Continuous casting billet quality detection method, device, electronic equipment and storage medium | |
| CN115131283A (en) | Defect detection and model training method, device, equipment and medium for target object | |
| CN109724993A (en) | Detection method, device and the storage medium of the degree of image recognition apparatus | |
| CN112613569A (en) | Image recognition method, and training method and device of image classification model | |
| CN109087282A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
| CN112001336A (en) | Pedestrian boundary crossing alarm method, device, equipment and system | |
| US11120541B2 (en) | Determination device and determining method thereof | |
| CN112949490A (en) | Device action detection method and device, electronic device and readable storage medium | |
| CN111859370A (en) | Method, apparatus, electronic device and computer-readable storage medium for identifying service | |
| CN116563969B (en) | Equipment inspection method, device, equipment and storage medium | |
| CN117541846A (en) | Production safety identification method based on big data and cloud management system | |
| CN114332051A (en) | Image-inference-based power transmission and distribution line equipment asset general survey method | |
| CN114529990A (en) | Photographing behavior recognition method and device, electronic equipment and storage medium |
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 |