CN109961433A - Product defects detection method, device and computer equipment - Google Patents
Product defects detection method, device and computer equipment Download PDFInfo
- Publication number
- CN109961433A CN109961433A CN201910247474.9A CN201910247474A CN109961433A CN 109961433 A CN109961433 A CN 109961433A CN 201910247474 A CN201910247474 A CN 201910247474A CN 109961433 A CN109961433 A CN 109961433A
- Authority
- CN
- China
- Prior art keywords
- picture
- defects detection
- target object
- detected
- default
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The application proposes a kind of product defects detection method, device and computer equipment, wherein method include: obtain detection request, and parsing carried out to detection request obtain include product surface picture to be detected；It whether include target object by default defects detection model inspection picture to be detected；If detecting, picture to be detected includes target object, and the corresponding target area of target object is marked, and extracts the target signature of target object, and carries out defect classification to target signature.As a result, by preset defects detection model it is real-time quick detect whether product includes defect, and classify to defect, improve the efficiency and precision of product defects detection.
Description
Technical field
This application involves automatic measurement technique field more particularly to a kind of product defects detection methods, device and computer
Equipment
Background technique
Traditional industry manufacturing industry production scene in, quality inspection is the key link in production procedure, for example, steel produce,
In the fields such as automobile manufacture, papermaking, battery manufacture, solar panels manufacture, printed wiring board, chip and liquid crystal display, to production
It is a kind of important means controlled product quality that the surface state of product, which carries out detection, judge product with the presence or absence of flaw and
Defect, and corresponding processing is done to product according to testing result.
In the related technology, it can be visually observed in production environment dependent on industry specialists by manual inspection mode
Photo provides judgement；The either artificial quality inspection mode of machine auxiliary, i.e., filtered by the quality inspection system with certain judgement
Fall and do not have defective photo, detection judgement is carried out by photo of the industry specialists to doubtful existing defects, however, aforesaid way efficiency
It is relatively low, and it is easy to appear erroneous judgement, cause testing result inaccurate.
Summary of the invention
The application is intended to solve at least some of the technical problems in related technologies.
Present applicant proposes a kind of product defects detection method, device and computer equipments, for solving in the prior art
Product defects detection mode efficiency is relatively low, and is easy to appear erroneous judgement, leads to the technical problem of testing result inaccuracy.
The application first aspect embodiment proposes a kind of product defects detection method, the described method comprises the following steps:
It obtains detection to request, and parsing is carried out to the detection request and obtains the picture to be detected comprising product surface；
It whether include target object by picture to be detected described in default defects detection model inspection；
If detecting, the picture to be detected includes target object, is carried out to the corresponding target area of the target object
Label, and the target signature of the target object is extracted, and defect classification is carried out to the target signature.
It is described by presetting defects detection model inspection institute as the mode in the cards of the first in the embodiment of the present application
State whether picture to be detected includes target object, comprising:
The picture feature in the picture to be detected in each region is calculated by the default defects detection model；
Judge whether the picture feature in described each region is default object features；
If judging any picture feature to preset object features, it will determine that the picture to be detected includes target object.
As second in the embodiment of the present application mode in the cards, described by presetting defects detection model inspection
Before whether the picture to be detected includes target object, further includes:
It obtains multiple to training sample picture；
Sample characteristics extraction is carried out to training sample picture to the multiple, by Faster RCNN algorithm to multiple institutes
It states sample characteristics and is trained the generation default defects detection model.
As the mode in the cards of the third in the embodiment of the present application, further includes:
The defect rank of the product surface is determined according to defect classification；
Different operation processings is determined according to the defect rank.
As the 4th kind of mode in the cards in the embodiment of the present application, further includes:
Obtain each testing result in preset time period；
Judge whether each testing result is accurate, and calculates the accuracy rate in the preset time period；
If the accuracy rate is less than preset standard, the default defects detection model is updated.
The product defects detection method of the embodiment of the present application is parsed by obtaining detection request, and to detection request
Obtain the picture to be detected comprising product surface；It whether include object by default defects detection model inspection picture to be detected
Body；If detecting, picture to be detected includes target object, and the corresponding target area of target object is marked, and is extracted
The target signature of target object, and defect classification is carried out to target signature.It is real-time quick by default defects detection model as a result,
Detect whether product includes defect, and classify to defect, improve the efficiency and precision of product defects detection.
The application second aspect embodiment proposes a kind of product defects detection device, comprising:
First obtains module, for obtaining detection request, and carries out parsing to detection request and obtains comprising product table
The picture to be detected in face；
Detection module, for whether including target object by picture to be detected described in default defects detection model inspection；
Categorization module, if for detecting that the picture to be detected includes target object, it is corresponding to the target object
Target area be marked, and extract the target signature of the target object, and defect point is carried out to the target signature
Class.
As the possible implementation of the first in the embodiment of the present application, the detection module is specifically used for:
The picture feature in the picture to be detected in each region is calculated by the default defects detection model；
Judge whether the picture feature in described each region is default object features；
If judging any picture feature to preset object features, it will determine that the picture to be detected includes target object.
As second in the embodiment of the present application possible implementation, further includes:
Second obtains module, multiple to training sample picture for obtaining；
Generation module passes through Faster RCNN for carrying out sample characteristics extraction to training sample picture to the multiple
Algorithm is trained multiple sample characteristics and generates the default defects detection model.
As the possible implementation of the third in the embodiment of the present application, further includes:
Determining module, for determining the defect rank of the product surface according to defect classification；
Processing module, for determining different operation processings according to the defect rank.
As the 4th kind of possible implementation in the embodiment of the present application, further includes:
Third obtains module, for obtaining each testing result in preset time period；
Judge computing module, for judging whether each testing result is accurate, and calculates in the preset time period
Interior accuracy rate；
Module is adjusted, if being less than preset standard for the accuracy rate, updates the default defects detection model.
The product defects detection device of the embodiment of the present application is parsed by obtaining detection request, and to detection request
Obtain the picture to be detected comprising product surface；It whether include object by default defects detection model inspection picture to be detected
Body；If detecting, picture to be detected includes target object, and the corresponding target area of target object is marked, and is extracted
The target signature of target object, and defect classification is carried out to target signature.It is real-time quick by default defects detection model as a result,
Detect whether product includes defect, and classify to defect, improve the efficiency and precision of product defects detection.
The application third aspect embodiment proposes a kind of computer equipment, including memory, processor and is stored in
On reservoir and the computer program that can run on a processor, when the processor executes described program, such as above-mentioned implementation is realized
Product defects detection method described in example.
The application fourth aspect embodiment proposes a kind of non-transitorycomputer readable storage medium, is stored thereon with meter
Calculation machine program realizes such as above-mentioned product defects detection method as described in the examples when the program is executed by processor.
The additional aspect of the application and advantage will be set forth in part in the description, and will partially become from the following description
It obtains obviously, or recognized by the practice of the application.
Detailed description of the invention
Fig. 1 is product defects detection method flow diagram provided by the embodiment of the present application one；
Fig. 2 is product defects detection method flow diagram provided by the embodiment of the present application two；
Fig. 3 is Faster RCNN algorithm structure schematic diagram provided by the embodiment of the present application；
Fig. 4 is product defects detection method flow diagram provided by the embodiment of the present application three；
Fig. 5 is product defects detection method flow diagram provided by the embodiment of the present application four；
Fig. 6 is product defects detection method flow diagram provided by the embodiment of the present application five；
Fig. 7 is product defects structure of the detecting device schematic diagram provided by the embodiment of the present application one；
Fig. 8 is product defects structure of the detecting device schematic diagram provided by the embodiment of the present application two；
Fig. 9 is product defects structure of the detecting device schematic diagram provided by the embodiment of the present application three；
Figure 10 is product defects structure of the detecting device schematic diagram provided by the embodiment of the present application four；
Figure 11 shows the block diagram for being suitable for the exemplary computer device for being used to realize the application embodiment.
Specific embodiment
Embodiments herein is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the application, and should not be understood as the limitation to the application.
Product defects automatic testing method, device and the calculating proposed according to the embodiment of the present application is described with reference to the accompanying drawings
Machine equipment.
Recognize for above-mentioned background technique, the prior art can pass through the people of manual inspection mode either machine auxiliary
Whether working medium procuratorial organ formula there is defect to judge product surface, relatively low there are manual operation efficiency and be easy to appear mistake
Sentence, leads to the technical problems such as testing result inaccuracy, the application proposes a kind of product defects detection method, by obtaining life in real time
Include the picture to be detected of product surface in producing line, picture to be detected is detected by default defects detection model, it can
Fast and accurately judge whether product includes defect, and determines position and the generic of defect.
Specifically, Fig. 1 is product defects detection method flow diagram provided by the embodiment of the present application one.
As shown in Figure 1, detection method includes the following steps for the product defects:
Step 101, it obtains detection to request, and parsing is carried out to detection request and obtains the mapping to be checked comprising product surface
Piece.
In practical applications, the product surface on production line is detected, can be set by installation associated picture acquisition
It is standby that the image comprising product surface on production line is acquired in real time, using the collected image comprising product surface as
Picture to be detected.
It should be noted that above-mentioned associated picture acquisition equipment can be using high-precision in order to improve the accuracy of detection
Image Acquisition camera, by adjusting angle, light, filter, times mirror, focusing etc., for identical product acquisition surface, multiple are waited for
It detects picture, and processing is carried out to multiple pictures to be detected by default screening criteria such as clarity etc. to filter out target to be checked
Mapping piece carries out subsequent detection.
Therefore, the above-mentioned picture i.e. to be detected of the image comprising product surface that will be generated in real time on production line is converted into detection
Request, so that detection request is parsed by modes such as default analytical algorithms after receiving detection request, it is available
Picture to be detected comprising product surface.
It step 102, whether include target object by default defects detection model inspection picture to be detected.
Step 103, if detecting, picture to be detected includes target object, is carried out to the corresponding target area of target object
Label, and the target signature of target object is extracted, and defect classification is carried out to target signature.
It is detected it is understood that picture to be detected is input to default defects detection model, detects mapping to be checked
It whether include target object in piece, if including target object, i.e., the corresponding product of picture to be detected, which exists, to be lacked in picture to be detected
It falls into, then the corresponding target area position of target object in product is marked, that is, determine the position of defect in product, simultaneously
The target signature of target object is extracted, is classified according to target signature to defect, after being carried out according to defect type
Continuous processing.
It should be noted that the corresponding target object of different products identical may not may also have to, for example product is liquid
Crystal display screen, target object can be display screen upper projecting portion point, and product paper, target object can be paper surface for another example
Sundries etc..
If not including target object in picture to be detected, indicating the corresponding product of picture to be detected, there is no defects, then not
It is handled.
Wherein, default defects detection model is pre-generated, how to generate defects detection below with reference to Fig. 2 specific descriptions
Model.As shown in Figure 2, comprising:
Step 201, it obtains multiple to training sample picture.
Step 202, sample characteristics extraction is carried out to training sample picture to multiple, by Faster RCNN algorithm to more
A sample characteristics are trained the default defects detection model of generation.
Specifically, the default defects detection model of generation is trained to multiple sample characteristics by Faster RCNN algorithm,
Defects detection model can using depth convolutional neural networks structure (Deep Convolutional Neural Network,
Deep CNN), and the defects of object detection algorithms testing product in utilization computer vision position.
Wherein, Faster RCNN algorithm is used as object detection algorithms to detect defective locations, as shown in Figure 3
Faster RCNN algorithm and theory structure, using the original image on production line as the input of default defects detection model, defect
Classification and defective locations are as output.
Specifically, the basic network of Faster RCNN is using SE-ResNet, that is, in SE-ResNet facilities network
Start to train multiple sample characteristics and depth convolutional neural networks structure mainly by convolutional layer, pond layer and Quan Lian on network
Layer such as connects at the composition.Wherein, the convolution operation of convolutional layer is using the different convolution kernel of weight to original image or characteristic pattern
(feature map) is scanned convolution, therefrom extracts the feature of various meanings, and export into characteristic pattern.The pond of pond layer
Change operation and dimensionality reduction operation then is carried out to characteristic pattern, the main feature in keeping characteristics figure.
It should be noted that real-time collected picture to be detected is due to illumination, the external environments such as temperature in the production line
Influence it is possible that the case where deformation, fuzzy etc. are unfavorable for defects detection, but deep neural network model using its convolution,
Pondization operation can carry out more accurately judgement to picture to be detected is collected, to improve the efficiency and standard of defects detection
True rate.
Specifically, Faster RCNN algorithm utilizes the available characteristic pattern of convolution operation of disaggregated model, recycles candidate
Whether it includes specific object that Local Area Network (Region Proposal Network) calculates in a certain region of original image
Body: if recycling convolutional network multiple target Object Extraction to carry out target signature comprising target object, and target object is predicted
Classification and boundary box (bounding box), the i.e. target area of target object, if not including target object, without
Classification.
It combines network losses to do combined training after generating default defects detection model, optimizes default defects detection model
Parameter makes default defects detection model have higher detection efficiency and accuracy rate.
As a kind of mode in the cards, as shown in figure 4, being by default defects detection model inspection picture to be detected
No includes target object, comprising:
Step 301, the picture feature in picture to be detected in each region is calculated by default defects detection model.
Step 302, judge whether the picture feature in each region is default object features.
Step 303, if judging any picture feature to preset object features, it will determine that picture to be detected includes object
Body.
It is understood that default defects detection model detects picture to be detected, picture to be detected is calculated first
The picture feature of middle each region a, wherein picture to be detected can be divided into multiple regions, therefore, can be obtained by calculating
To multiple picture features of multiple regions, judge whether multiple picture features are default object features, if there is any one figure
Piece feature is default object features, it is determined that picture to be detected includes target object, that is to say, that any area of picture to be detected
Domain includes target object, it is determined that the corresponding product of picture to be detected includes defect.
Detecting defect, and after determining defect classification, there are many modes to the processing of faulty goods, as a kind of possibility
The mode of realization, as shown in Figure 5, further includes:
Step 401, classified according to defect and determine the defect rank of product surface.
Step 402, different operation processings is determined according to defect rank.
Specifically, after carrying out defect classification to target signature, classified according to defect and determine the defect rank of product surface,
That is, different defect ranks is determined according to different defect situations, for example, can be divided into defect according to defect size
Advanced, middle rank, rudimentary three kinds of grades determine different operation processings according to defect rank, and very big for defect is advanced scarce
Product is fallen into, can not be repaired by other means, discard processing can be taken, product corresponding for intermediate defect,
It does not influence that its price of reduction can be taken to be sold in the case where product use, product corresponding for rudimentary defect, if
It can be repaired by other means, normal product can be changed into and sold or used.
It should be noted that can be designed in conjunction with business scenario upon completion of the assays, can be made according to business demand
Meet the response of production environment scene requirement, for example, issuing alarm signal after detecting certain product surface existing defects, mentioning
Show staff；Data can also be stored, to carry out subsequent analysis；Manipulator motion can also be controlled, will be lacked
Product is fallen into take out from production line etc..
Further, it is also possible to will test the data such as result and respondent behavior as production log storage on line to Production database
In, in order to which defects detection model is trained optimization according to the data of storage, further increase the efficiency of product defects detection
And accuracy rate, and be conducive to the optimization and upgrading of industrial production line.
In order to further increase the efficiency and accuracy rate of detection, as a kind of mode in the cards, as shown in fig. 6, also
Include:
Step 501, each testing result in preset time period is obtained.
Step 502, judge whether each testing result is accurate, and calculate accuracy rate within a preset period of time.
Step 503, if accuracy rate is less than preset standard, default defects detection model is updated.
Specifically, after detecting a period of time, judge whether the accuracy rate of the testing result in a period reaches pre-
The standard being first arranged can update the parameter setting of default defects detection model if being less than preset standard, make default defects detection
Model more accurately detects faulty goods, improves the efficiency and accuracy rate of defects detection.
For example, for example accuracy rate whithin a period of time is calculated as 90 percent less than preset standard 9 percent
15, then the operation of update etc can be optimized, to model to improve detection accuracy.
The product defects detection method of the embodiment of the present application is parsed by obtaining detection request, and to detection request
Obtain the picture to be detected comprising product surface；It whether include object by default defects detection model inspection picture to be detected
Body；If detecting, picture to be detected includes target object, and the corresponding target area of target object is marked, and is extracted
The target signature of target object, and defect classification is carried out to target signature.It is real-time quick by default defects detection model as a result,
Detect whether product includes defect, and classify to defect, improve the efficiency and precision of product defects detection.
In order to realize above-described embodiment, the application also proposes a kind of product defects detection device.
As shown in fig. 7, the product defects detection device includes: the first acquisition module 601, detection module 602 and classification mould
Block 603.
First obtains module 601, for obtaining detection request, and carries out parsing to detection request and obtains comprising product surface
Picture to be detected.
Detection module 602, for whether including target object by default defects detection model inspection picture to be detected.
Categorization module 603, if for detecting that picture to be detected includes target object, target corresponding to target object
Region is marked, and extracts the target signature of target object, and carries out defect classification to target signature.
As a kind of mode in the cards, detection module 603 is specifically used for:
The picture feature in picture to be detected in each region is calculated by default defects detection model；
Judge whether the picture feature in each region is default object features；
If judging any picture feature to preset object features, it will determine that picture to be detected includes target object.
As the mode of alternatively possible realization, as shown in figure 8, the product defects detect dress on the basis of Fig. 7
It sets, further includes: second obtains module 604 and generation module 605.
Second obtains module 604, multiple to training sample picture for obtaining.
Generation module 605 passes through Faster RCNN for carrying out sample characteristics extraction to training sample picture to multiple
Algorithm is trained the default defects detection model of generation to multiple sample characteristics.
As another mode in the cards, as shown in Figure 9, further includes: determining module 606 and processing module 607.
Determining module 606 determines the defect rank of product surface for classifying according to defect.
Processing module 607, for determining different operation processings according to defect rank.
As another mode in the cards, as shown in Figure 10, further includes: third obtains module 608, judgement calculates mould
Block 609 and adjustment module 610.
Third obtains module 608, for obtaining each testing result in preset time period.
Judge computing module 609, for judging whether each testing result is accurate, and calculates standard within a preset period of time
True rate.
Module 610 is adjusted, if being less than preset standard for accuracy rate, updates default defects detection model.
It should be noted that the aforementioned device that the embodiment is also applied for the explanation of embodiment of the method, herein not
It repeats again.
The product defects detection device of the embodiment of the present application is parsed by obtaining detection request, and to detection request
Obtain the picture to be detected comprising product surface；It whether include object by default defects detection model inspection picture to be detected
Body；If detecting, picture to be detected includes target object, and the corresponding target area of target object is marked, and is extracted
The target signature of target object, and defect classification is carried out to target signature.It is real-time quick by default defects detection model as a result,
Detect whether product includes defect, and classify to defect, improve the efficiency and precision of product defects detection.
In order to realize above-described embodiment, the application also proposes computer equipment, including memory, processor and is stored in
On reservoir and the computer program that can run on a processor, when the processor executes described program, such as above-mentioned implementation is realized
Product defects detection method described in example.
In order to realize above-described embodiment, the application also proposes a kind of non-transitorycomputer readable storage medium, deposits thereon
Computer program is contained, such as above-mentioned product defects detection method as described in the examples is realized when which is executed by processor.
Figure 11 shows the block diagram for being suitable for the exemplary computer device for being used to realize the application embodiment.Figure 11 is shown
Computer equipment 12 be only an example, should not function to the embodiment of the present application and use scope bring any restrictions.
As shown in Figure 10, computer equipment 12 is showed in the form of universal computing device.The component of computer equipment 12 can
To include but is not limited to: one or more processor or processing unit 16, system storage 28 connect different system components
The bus 18 of (including system storage 28 and processing unit 16).
Bus 18 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (Industry Standard
Architecture；Hereinafter referred to as: ISA) bus, microchannel architecture (Micro Channel Architecture；Below
Referred to as: MAC) bus, enhanced isa bus, Video Electronics Standards Association (Video Electronics Standards
Association；Hereinafter referred to as: VESA) local bus and peripheral component interconnection (Peripheral Component
Interconnection；Hereinafter referred to as: PCI) bus.
Computer equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that computer equipment 12 accesses, including volatile and non-volatile media, moveable and immovable medium.
Memory 28 may include the computer system readable media of form of volatile memory, such as random access memory
Device (Random Access Memory；Hereinafter referred to as: RAM) 30 and/or cache memory 32.Computer equipment 12 can be with
It further comprise other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example,
Storage system 34 can be used for reading and writing immovable, non-volatile magnetic media, and (Figure 11 do not show, commonly referred to as " hard drive
Device ").Although being not shown in Fig. 5, the disk for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided and driven
Dynamic device, and to removable anonvolatile optical disk (such as: compact disc read-only memory (Compact Disc Read Only
Memory；Hereinafter referred to as: CD-ROM), digital multi CD-ROM (Digital Video Disc Read Only
Memory；Hereinafter referred to as: DVD-ROM) or other optical mediums) read-write CD drive.In these cases, each driving
Device can be connected by one or more data media interfaces with bus 18.Memory 28 may include that at least one program produces
Product, the program product have one group of (for example, at least one) program module, and it is each that these program modules are configured to perform the application
The function of embodiment.
Program/utility 40 with one group of (at least one) program module 42 can store in such as memory 28
In, such program module 42 include but is not limited to operating system, one or more application program, other program modules and
It may include the realization of network environment in program data, each of these examples or certain combination.Program module 42 is usual
Execute the function and/or method in embodiments described herein.
Computer equipment 12 can also be with one or more external equipments 14 (such as keyboard, sensing equipment, display 24
Deng) communication, can also be enabled a user to one or more equipment interact with the computer equipment 12 communicate, and/or with make
The computer equipment 12 any equipment (such as network interface card, the modulatedemodulate that can be communicated with one or more of the other calculating equipment
Adjust device etc.) communication.This communication can be carried out by input/output (I/O) interface 22.Also, computer equipment 12 may be used also
To pass through network adapter 20 and one or more network (such as local area network (Local Area Network；Hereinafter referred to as:
LAN), wide area network (Wide Area Network；Hereinafter referred to as: WAN) and/or public network, for example, internet) communication.Such as figure
Shown, network adapter 20 is communicated by bus 18 with other modules of computer equipment 12.It should be understood that although not showing in figure
Out, other hardware and/or software module can be used in conjunction with computer equipment 12, including but not limited to: microcode, device drives
Device, redundant processing unit, external disk drive array, RAID system, tape drive and data backup storage system etc..
Processing unit 16 by the program that is stored in system storage 28 of operation, thereby executing various function application and
Data processing, such as realize the plot jump method based on speech recognition referred in previous embodiment.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is contained at least one embodiment or example of the application.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present application, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing custom logic function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the application includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, Lai Zhihang function, this should be by the application
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system, including the system of processor or other can be held from instruction
The instruction fetch of row system, device or equipment and the system executed instruction) it uses, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium ", which can be, any may include, stores, communicates, propagates or pass
Defeated program is for instruction execution system, device or equipment or the dress used in conjunction with these instruction execution systems, device or equipment
It sets.The more specific example (non-exhaustive list) of computer-readable medium include the following: there is the electricity of one or more wirings
Interconnecting piece (electronic device), portable computer diskette box (magnetic device), random access memory (RAM), read-only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can print described program on it or other are suitable
Medium, because can then be edited, be interpreted or when necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each section of the application can be realized with hardware, software, firmware or their combination.Above-mentioned
In embodiment, software that multiple steps or method can be executed in memory and by suitable instruction execution system with storage
Or firmware is realized.Such as, if realized with hardware in another embodiment, following skill well known in the art can be used
Any one of art or their combination are realized: have for data-signal is realized the logic gates of logic function from
Logic circuit is dissipated, the specific integrated circuit with suitable combinational logic gate circuit, programmable gate array (PGA), scene can compile
Journey gate array (FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
It, can also be in addition, can integrate in a processing module in each functional unit in each embodiment of the application
It is that each unit physically exists alone, can also be integrated in two or more units in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized and when sold or used as an independent product in the form of software function module, also can store in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..Although having been shown and retouching above
Embodiments herein is stated, it is to be understood that above-described embodiment is exemplary, and should not be understood as the limit to the application
System, those skilled in the art can be changed above-described embodiment, modify, replace and become within the scope of application
Type.
Claims (12)
1. a kind of product defects detection method, which comprises the following steps:
It obtains detection to request, and parsing is carried out to the detection request and obtains the picture to be detected comprising product surface；
It whether include target object by picture to be detected described in default defects detection model inspection；
If detecting, the picture to be detected includes target object, is marked to the corresponding target area of the target object
Note, and the target signature of the target object is extracted, and defect classification is carried out to the target signature.
2. the method as described in claim 1, which is characterized in that described by be detected described in default defects detection model inspection
Whether picture includes target object, comprising:
The picture feature in the picture to be detected in each region is calculated by the default defects detection model；
Judge whether the picture feature in described each region is default object features；
If judging any picture feature to preset object features, it will determine that the picture to be detected includes target object.
3. the method as described in claim 1, which is characterized in that described by be checked described in default defects detection model inspection
Before whether mapping piece includes target object, further includes:
It obtains multiple to training sample picture；
Sample characteristics extraction is carried out to training sample picture to the multiple, by Faster RCNN algorithm to multiple samples
Eigen, which is trained, generates the default defects detection model.
4. the method as described in claim 1, which is characterized in that further include:
The defect rank of the product surface is determined according to defect classification；
Different operation processings is determined according to the defect rank.
5. the method as described in claim 1, which is characterized in that further include:
Obtain each testing result in preset time period；
Judge whether each testing result is accurate, and calculates the accuracy rate in the preset time period；
If the accuracy rate is less than preset standard, the default defects detection model is updated.
6. a kind of product defects detection device characterized by comprising
First obtains module, for obtaining detection request, and carries out parsing to the detection request and obtains comprising product surface
Picture to be detected；
Detection module, for whether including target object by picture to be detected described in default defects detection model inspection；
Categorization module, if for detecting that the picture to be detected includes target object, mesh corresponding to the target object
Mark region is marked, and extracts the target signature of the target object, and carries out defect classification to the target signature.
7. device as claimed in claim 6, which is characterized in that the detection module is specifically used for:
The picture feature in the picture to be detected in each region is calculated by the default defects detection model；
Judge whether the picture feature in described each region is default object features；
If judging any picture feature to preset object features, it will determine that the picture to be detected includes target object.
8. device as claimed in claim 6, which is characterized in that further include:
Second obtains module, multiple to training sample picture for obtaining；
Generation module passes through Faster RCNN algorithm for carrying out sample characteristics extraction to training sample picture to the multiple
Multiple sample characteristics are trained and generate the default defects detection model.
9. device as claimed in claim 8, which is characterized in that further include:
Determining module, for determining the defect rank of the product surface according to defect classification；
Processing module, for determining different operation processings according to the defect rank.
10. device as claimed in claim 6, which is characterized in that further include:
Third obtains module, for obtaining each testing result in preset time period；
Judge computing module, for judging whether each testing result is accurate, and calculates in the preset time period
Accuracy rate；
Module is adjusted, if being less than preset standard for the accuracy rate, updates the default defects detection model.
11. a kind of computer equipment, which is characterized in that including memory, processor and store on a memory and can handle
The computer program run on device when the processor executes described program, realizes such as production as claimed in any one of claims 1 to 5
Product defect inspection method.
12. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, which is characterized in that the program
Such as product defects detection method as claimed in any one of claims 1 to 5 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910247474.9A CN109961433A (en) | 2019-03-29 | 2019-03-29 | Product defects detection method, device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910247474.9A CN109961433A (en) | 2019-03-29 | 2019-03-29 | Product defects detection method, device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109961433A true CN109961433A (en) | 2019-07-02 |
Family
ID=67025323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910247474.9A Pending CN109961433A (en) | 2019-03-29 | 2019-03-29 | Product defects detection method, device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961433A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599484A (en) * | 2019-09-19 | 2019-12-20 | 京东方科技集团股份有限公司 | Defect detection method, device and storage medium |
| CN111179261A (en) * | 2019-12-31 | 2020-05-19 | 深圳市太赫兹科技创新研究院 | Defect detection method, system, terminal device and storage medium |
| CN111311099A (en) * | 2020-02-18 | 2020-06-19 | 中车贵阳车辆有限公司 | Railway wagon quality inspection information system |
| CN111402220A (en) * | 2020-03-11 | 2020-07-10 | 北京百度网讯科技有限公司 | Method and device for acquiring information |
| CN111696095A (en) * | 2020-06-12 | 2020-09-22 | 北京百度网讯科技有限公司 | Method and device for detecting surface defects of object |
| CN111812118A (en) * | 2020-06-24 | 2020-10-23 | 阿丘机器人科技(苏州)有限公司 | PCB detection method, device, equipment and computer readable storage medium |
| CN111929328A (en) * | 2020-09-09 | 2020-11-13 | 深兰人工智能芯片研究院（江苏）有限公司 | Zipper defect detection method and device |
| CN111999308A (en) * | 2020-08-27 | 2020-11-27 | 银河水滴科技（北京）有限公司 | Defect detection control method, control device and control system |
| CN112950560A (en) * | 2021-02-20 | 2021-06-11 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Electronic component defect detection method, device and system |
| CN113409308A (en) * | 2021-07-16 | 2021-09-17 | 李华洲 | Circuit board fault defect detection method, device, detection equipment and system |
| CN114119612A (en) * | 2022-01-25 | 2022-03-01 | 武汉精立电子技术有限公司 | Picture-based appearance detection method and system |
| CN115592226A (en) * | 2021-08-03 | 2023-01-13 | 苏州楚翰真空科技有限公司（Cn） | Production and quality inspection integrated method and system for vacuum cups |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156809A (en) * | 2015-04-24 | 2016-11-23 | 阿里巴巴集团控股有限公司 | For updating the method and device of disaggregated model |
| CN108257121A (en) * | 2018-01-09 | 2018-07-06 | 北京百度网讯科技有限公司 | The newer method, apparatus of product defects detection model, storage medium and terminal device |
| CN108564104A (en) * | 2018-01-09 | 2018-09-21 | 北京百度网讯科技有限公司 | Product defects detection method, device, system, server and storage medium |
| CN109344716A (en) * | 2018-08-31 | 2019-02-15 | 深圳前海达闼云端智能科技有限公司 | Training method, detection method, device, medium and equipment of living body detection model |
-
2019
- 2019-03-29 CN CN201910247474.9A patent/CN109961433A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156809A (en) * | 2015-04-24 | 2016-11-23 | 阿里巴巴集团控股有限公司 | For updating the method and device of disaggregated model |
| CN108257121A (en) * | 2018-01-09 | 2018-07-06 | 北京百度网讯科技有限公司 | The newer method, apparatus of product defects detection model, storage medium and terminal device |
| CN108564104A (en) * | 2018-01-09 | 2018-09-21 | 北京百度网讯科技有限公司 | Product defects detection method, device, system, server and storage medium |
| CN109344716A (en) * | 2018-08-31 | 2019-02-15 | 深圳前海达闼云端智能科技有限公司 | Training method, detection method, device, medium and equipment of living body detection model |
Non-Patent Citations (2)
| Title |
|---|
| 徐科 等: "《金属表面质量在线检测技术》", 31 October 2016 * |
| 赖传雅: "《赖传雅文集》", 28 February 2016, 南宁：广西科学技 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599484B (en) * | 2019-09-19 | 2023-01-10 | 京东方科技集团股份有限公司 | Defect detection method, device and storage medium |
| CN110599484A (en) * | 2019-09-19 | 2019-12-20 | 京东方科技集团股份有限公司 | Defect detection method, device and storage medium |
| CN111179261A (en) * | 2019-12-31 | 2020-05-19 | 深圳市太赫兹科技创新研究院 | Defect detection method, system, terminal device and storage medium |
| CN111311099A (en) * | 2020-02-18 | 2020-06-19 | 中车贵阳车辆有限公司 | Railway wagon quality inspection information system |
| CN111402220A (en) * | 2020-03-11 | 2020-07-10 | 北京百度网讯科技有限公司 | Method and device for acquiring information |
| CN111696095A (en) * | 2020-06-12 | 2020-09-22 | 北京百度网讯科技有限公司 | Method and device for detecting surface defects of object |
| CN111696095B (en) * | 2020-06-12 | 2023-08-08 | 北京百度网讯科技有限公司 | Method and device for detecting surface defects of object |
| CN111812118A (en) * | 2020-06-24 | 2020-10-23 | 阿丘机器人科技(苏州)有限公司 | PCB detection method, device, equipment and computer readable storage medium |
| CN111999308A (en) * | 2020-08-27 | 2020-11-27 | 银河水滴科技（北京）有限公司 | Defect detection control method, control device and control system |
| CN111929328A (en) * | 2020-09-09 | 2020-11-13 | 深兰人工智能芯片研究院（江苏）有限公司 | Zipper defect detection method and device |
| CN112950560A (en) * | 2021-02-20 | 2021-06-11 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Electronic component defect detection method, device and system |
| CN113409308A (en) * | 2021-07-16 | 2021-09-17 | 李华洲 | Circuit board fault defect detection method, device, detection equipment and system |
| CN115592226A (en) * | 2021-08-03 | 2023-01-13 | 苏州楚翰真空科技有限公司（Cn） | Production and quality inspection integrated method and system for vacuum cups |
| CN115592226B (en) * | 2021-08-03 | 2023-07-21 | 苏州楚翰真空科技有限公司 | Thermos cup production quality inspection integrated method and system |
| CN114119612A (en) * | 2022-01-25 | 2022-03-01 | 武汉精立电子技术有限公司 | Picture-based appearance detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109961433A (en) | Product defects detection method, device and computer equipment | |
| CN109949305A (en) | Method for detecting surface defects of products, device and computer equipment | |
| CN109871895A (en) | The defect inspection method and device of circuit board | |
| CN100414558C (en) | Automatic fingerprint distinguishing system and method based on template learning | |
| Cai et al. | A pointer meter recognition method based on virtual sample generation technology | |
| CN110120036A (en) | A kind of multiple dimensioned tire X-ray defect detection method | |
| Yu et al. | Detecting gear surface defects using background‐weakening method and convolutional neural network | |
| CN109978868A (en) | Toy appearance quality determining method and its relevant device | |
| CN109919135A (en) | Behavioral value method, apparatus based on deep learning | |
| CN116340796B (en) | Time sequence data analysis method, device, equipment and storage medium | |
| CN110413513A (en) | Publication quality control system, method and the device of software | |
| CN109977826A (en) | The classification recognition methods of object and device | |
| CN110084137A (en) | Data processing method, device and computer equipment based on Driving Scene | |
| Le et al. | Novel framework for optical film defect detection and classification | |
| CN113762182A (en) | Milling cutter wear state prediction method and device based on depth network self-adaptation | |
| CN108847284A (en) | The human-body biological age calculates apparatus and system | |
| CN110335242A (en) | A kind of tire X-ray defect detection method based on multi-model fusion | |
| CN116664846B (en) | Method and system for realizing 3D printing bridge deck construction quality monitoring based on semantic segmentation | |
| CN112232288A (en) | Satellite map target identification method and system based on deep learning | |
| Wan et al. | Robust and accurate sub-pixel extraction method of laser stripes in complex circumstances | |
| Fang et al. | Illumination-robust milling surface roughness machine vision inspection based on MAML++ network | |
| Zhang et al. | A YOLOv3‐Based Industrial Instrument Classification and Reading Recognition Method | |
| Zhang et al. | A dual-structure attention-based multi-level feature fusion network for automatic surface defect detection | |
| CN114092415A (en) | Large-size light guide plate defect visual detection method | |
| CN107403192A (en) | A kind of fast target detection method and system based on multi-categorizer |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190702 |