CN109035248A - Defect detection method, apparatus, terminal device, server and storage medium - Google Patents
Defect detection method, apparatus, terminal device, server and storage medium Download PDFInfo
- Publication number
- CN109035248A CN109035248A CN201811031071.2A CN201811031071A CN109035248A CN 109035248 A CN109035248 A CN 109035248A CN 201811031071 A CN201811031071 A CN 201811031071A CN 109035248 A CN109035248 A CN 109035248A
- Authority
- CN
- China
- Prior art keywords
- image
- detected
- detection
- fault
- server
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
This application discloses fabric defect detection method, device, terminal device, server and storage mediums, this method includes receiving image to be detected, to be detected classified using the first detection model to described, wherein, it is to belong to fabric surface included in it to there are problems that fault image still falls within the normal picture that fault is not present in fabric included in it that first detection model, which is used to detect image to be detected,；If described image to be detected is classified as described problem image by first detection model, defect detection request is generated；Described image to be detected and defect detection request are sent to server, the server is equipped with the second detection model, wherein second detection model is used to detect the type of fault；Receive the prediction result for the fault type that the server is sent.This method, device, terminal device, server and storage medium can reduce the use cost of user.
Description
Technical field
This application involves Fabric Detection technical fields, in particular to fabric defect detection method, device, terminal device, clothes
Business device and storage medium.
Background technique
On the production line of the fabrics such as woven fabric, looped fabric, non-woven cloth, need whether to detect produced fabric
There are faults, for example, whether having spot, broken hole, fluffing etc. on fabric.
There are mainly two types of modes for current detection method: the first is pure manual detection mode, i.e., is stood by testing staff
Fabric defects is found in such a way that naked eyes detect before perching equipment and fault is marked or is recorded；Second auxiliary for machine
The detection mode helped is mainly found fabric defects by machine vision and computer program analysis and realizes classification.
In the case where the yield of fabric is very big, being detected by testing staff will have high labor costs the first detection mode,
Moreover, testing staff is easy fatigue after a period of time that works, to there is a possibility that erroneous detection occurs.Therefore, by detecting
Personnel are not high come the detection efficiency detected and accuracy in detection is not sufficiently stable.
Second of detection mode is generally by a whole set of program Solidification into terminal device, it is difficult to the development iteration of business, use
Family in use, if detection demand changes, needs replacing a whole set of program, higher cost.
Summary of the invention
In view of problem above, the embodiment of the present invention provides a kind of fabric defect detection method, device, terminal device, clothes
Business device and storage medium, can solve the technical issues of above-mentioned background technology part is mentioned.
In a first aspect, the fabric defect detection method of embodiment according to the invention, comprising: receive image to be detected, benefit
Classified with the first detection model to described image to be detected, wherein first detection model is to belong to for detection image
Fabric surface included in it has still falling within fabric included in it by fault image that there is no faults
Normal picture；If described image to be detected is classified as described problem image by first detection model, fault inspection is generated
Survey request；Described image to be detected and defect detection request are sent to server, the server is equipped with the second detection mould
Type, wherein second detection model is used to detect the type of fault；Receive the prediction for the fault type that the server is sent
As a result.
Second aspect, the fabric defect detection method of embodiment according to the invention, comprising: receive image to be detected and defect
Point detection request；It is requested according to the defect detection, described image to be detected is detected using the second detection model,
In, second detection model is used to detect the type of fault；Generate and export the prediction result of the fault type.
The third aspect, the fabric defects detection device of embodiment according to the invention, comprising: detection categorization module is used for
Image to be detected is received, is classified using the first detection model to described image to be detected, wherein first detection model
It is to belong to fabric surface included in it to there are problems that fault image is still fallen within included in it for detection image
The normal picture of fault is not present in fabric；Generation module is requested, if being used for first detection model for the mapping to be checked
As being classified as described problem image, then defect detection request is generated；Request sending module is used for described image to be detected and defect
Point detection request is sent to server, and the server is equipped with the second detection model, wherein second detection model is used for
Detect the type of fault；As a result receiving module, for receiving the prediction result for the fault type that the server is sent.
Fourth aspect, the fabric defects detection device of embodiment according to the invention, comprising: request receiving module is used for
Receive image to be detected and defect detection request；Fault categorization module utilizes the second inspection for requesting according to the defect detection
It surveys model to detect described image to be detected, wherein second detection model is used to detect the type of fault；As a result defeated
Module out, for generating and exporting the prediction result of the fault type.
5th aspect, the terminal device of embodiment according to the invention, comprising: processor, memory, communication interface and logical
Believe bus, the processor, the memory and the communication interface complete mutual communication by the communication bus；Institute
It states memory and is stored with executable instruction, wherein it is above-mentioned that the executable instruction executes the processor
Method described in first aspect.
6th aspect, the server of embodiment according to the invention, comprising: processor, memory, communication interface and communication
Bus, the processor, the memory and the communication interface complete mutual communication by the communication bus；It is described
Memory is stored with executable instruction, wherein the executable instruction makes the processor execute above-mentioned the upon being performed
Method described in two aspects.
7th aspect, the storage medium of embodiment according to the invention are stored thereon with computer executable instructions,
In, the executable instruction makes computer execute method described in above-mentioned first aspect, above-mentioned second aspect upon being performed
The method.
It can be seen from the above that the scheme of the embodiment of the present invention is provided only for detection fabric in terminal device
Surface whether there is the first detection model of fault, and will be mounted in cloud service to the second detection model that fault is classified
On device, by the communication of terminal device and server, when detecting fabric surface, there are when fault, then carry out cloud detection and return
Pass result.The second detection model that server carries can carry out maintenance and upgrade according to the development of business by provider, work as user
It when having new detection demand, such as needs to increase fault type identification, then only need to be updated iteration in server side, without
The detection model of replacement terminal equipment reduces the use cost of user.
Detailed description of the invention
Fig. 1 is that present invention could apply to the schematic diagrames of exemplary system architecture therein；
Fig. 2 is the flow chart of the method for detection model training of one embodiment according to the invention；
Fig. 3 is the signaling process figure of the fabric defect detection method of one embodiment according to the invention；
Fig. 4 is the flow chart of the fabric defect detection method of one embodiment according to the invention；
Fig. 5 is the flow chart of the fabric defect detection method of another embodiment according to the invention；
Fig. 6 is the schematic diagram of the fabric defects detection device of one embodiment according to the invention；
Fig. 7 is the schematic diagram of the fabric defects detection device of another embodiment according to the invention；
Fig. 8 is the schematic diagram of the terminal device of one embodiment according to the invention；
Fig. 9 is the schematic diagram of the server of one embodiment according to the invention.
Specific embodiment
Theme described herein is discussed referring now to example embodiment.It should be understood that discussing these embodiments only
It is in order to enable those skilled in the art can better understand that being not to claim to realize theme described herein
Protection scope, applicability or the exemplary limitation illustrated in book.It can be in the protection scope for not departing from present disclosure
In the case of, the function and arrangement of the element discussed are changed.Each example can according to need, omit, substitute or
Add various processes or component.For example, described method can be executed according to described order in a different order, with
And each step can be added, omits or combine.In addition, feature described in relatively some examples is in other examples
It can be combined.
As used in this article, term " includes " and its modification indicate open term, are meant that " including but not limited to ".
Term "based" indicates " being based at least partially on ".Term " one embodiment " and " embodiment " expression " at least one implementation
Example ".Term " another embodiment " expression " at least one other embodiment ".Term " first ", " second " etc. may refer to not
Same or identical object.Here may include other definition, either specific or implicit.Unless bright in context
It really indicates, otherwise the definition of a term is consistent throughout the specification.
The scheme of the embodiment of the present invention is provided only for detection fabric surface with the presence or absence of the of fault in terminal device
One detection model, and the second detection model that fault is classified being mounted on cloud server, by terminal device and
The communication of server, when detecting fabric surface, there are when fault, then carry out cloud detection and return result.What server carried
Second detection model can carry out maintenance and upgrade, when user has new detection demand, example according to the development of business by provider
It such as needs to increase fault type identification, then only need to be updated iteration in server side, the detection without replacement terminal equipment
Model reduces the use cost of user.
Fig. 1 is shown can be using the exemplary system frame of the embodiment of fabric defect detection method or device of the invention
The schematic diagram of structure.As shown in Figure 1, the system architecture 100 may include terminal device 101,102,103, network 104 and server
105.Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104
It may include various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send data, message etc..Various client applications can be installed on terminal device 101,102,103, such as detected for calling
Detection application program of model etc..Terminal device 101,102,103 and server 105 can be such as computer or other are suitable
The electronic equipment with computing capability.
It should be understood that the quantity of terminal device, network and server in Fig. 1 is only schematical.According to practical need
It wants, can have any number of terminal device, network and server.
The scheme of the embodiment of the present invention includes model training stage and actually detected stage.
Fig. 2 shows the flow charts of the method for detection model training of one embodiment according to the invention.Fig. 2 institute
The method 200 shown corresponds to the first detection model and the second detection model training stage, is rolled up using training data training
Product neural network (CNN:Convolutional Neural Network) model (i.e. the first detection model) and SVM (Support
Vector Machine, support vector machines) classifier (i.e. the second detection model).CNN model be used for detect inputted include
The image of fabric belongs to normal picture or problem image, and SVM classifier is for being divided problem image by fault type
Class.Wherein normal picture refers to that the image of fault is not present in fabric surface included in it, and problem image, which refers to, wherein to be wrapped
There are the images of fault for the fabric surface contained.Method 200 shown in Fig. 2 can by computer or other suitably there is calculating energy
The electronic equipment of power is realized.
As shown in Fig. 2, receiving the image of multiple original shootings in box 202.Wherein, the image of multiple original shooting
Including multiple normal pictures and multiple problem images.
In box 204, image labeling (Image Annotation) processing is executed to the image of multiple original shooting, with
Obtain first sample image set SP1.Wherein, each of first sample image set SP1 sample image is to multiple original
One of image of the image of shooting executes what image labeling was handled.Image labeling processing is known technology, herein
Omit descriptions thereof.
In box 206, gray processing processing is executed to first sample image set SP1, it will be in first sample image set SP1
Each sample image is converted to gray level image.
In box 208, some or all sample images are chosen from the first sample image set SP1 that gray processing is handled and are made
For drawing of seeds picture.It such as, but not limited to, include more problem images in selected drawing of seeds picture, because of usual situation
Under, first sample image set SP1 includes less problem.
In box 210, one or many angularly rotations, mirror image are executed to each drawing of seeds picture and/or other are suitable
Operation, with from obtaining one or more images derived from each drawing of seeds picture.Wherein, the first sample of gray processing processing
Sample image in image set SP1 and the second sample graph image set is together to form from the image obtained derived from each drawing of seeds picture
SP2。
Under normal conditions, problem sample image is fewer than normal sample image, such as problem sample image and normal sample
The ratio of image may be 1: 10~1: 20, thus problem sample image and normal sample figure in first sample image set SP1
The quantity of picture be it is unbalanced, and unbalanced sample will lead to when carrying out neural metwork training last training result occur it is different
Normal deviation.Therefore, in 208 selected seed image of box, normal sample image of the sample image than selection the problem of selection
It is more, so as to after the angularly operation of rotation, mirror image etc. of process box 210 in obtained the second sample graph image set SP2,
The quantity of problem sample image and normal sample image is balance, abnormal variation occurs to avoid training result.In addition, passing through
The operation of box 206 and 208 can increase the quantity of training sample (for example, can be by 2500 sample images after treatment
Obtain the sample image more than 50000 or even 100000), and with the increase of training samples, the mind that finally training obtains
There is higher accuracy in detection through network model and classifier.
Box 206-210 constitutes the image preprocessing process (Image Preprocessing) of method 200.
In box 212, the property parameters of each gray level image in the second sample graph image set SP2 are obtained, wherein the attribute
Parameter includes but is not limited to the length of image, width etc..
In box 214, from each rule chosen in the second sample graph image set SP2 in its property parameters the first rule set of satisfaction
Multiple images then, as training the third sample graph image set SP3 of CNN model.Wherein, first rule set is for defining
The condition that sample image suitable for training CNN model needs to meet.For example, the first rule set is defined suitable for CNN model
Length limitation, width limitation that sample image needs to meet etc..Third sample graph image set SP3 includes multiple normal pictures and multiple
Problem image.
In box 216, each rule that its property parameters meets Second Rule concentration are chosen from the second sample graph image set SP2
Multiple images then, as training the 4th sample graph image set SP4 of SVM classifier.Wherein, the Second Rule collection is for fixed
Justice is suitable for the condition that the sample image of training SVM classifier needs to meet.For example, the definition of Second Rule collection is suitable for training
Length limitation, width limitation that the sample image of SVM classifier needs to meet etc..4th sample graph image set SP4 includes multiple more
The problem of kind fault is classified image.Fault classification is such as, but not limited to spot, yarn defect, float, printing and dyeing fault, side defect, fold, latitude
Tiltedly, broken hole, hook silk, sanding unevenness, blur, fluffing, scratch, roll line, stop Mark.Wherein, spot includes greasy dirt, rust spot, color
Point, spot, mildew, auxiliary agent spot；Yarn defect includes dead cotton, slubbing, flyings, thick young yarn, soiled yarn, the dry unevenness of item；Float include broken yarn,
Knot, dropped stitch, rotten needle, try to stop people from fighting each other it is elastic, disconnected try to stop people from fighting each other, spacing is unstable, cloth cover plays snake, filling is shown up, color fibre, needle path, mistake yarn, yarn
Trace, try to stop people from fighting each other it is show-through, try to stop people from fighting each other and show up；Printing and dyeing fault include bite, stamp displacement, stamp staining, stamp cross that bottom, stamp is bad, contaminates
Flower two tone colour, loses colour, difference；When defect includes pin hole, double needle hole, crimping, rotten side, narrow envelope, wealthy envelope；Fold includes intermediate folder
Trace, cloth cover corrugation, folding line；Skew of weft includes twill, arch.
In box 218, use the image of third sample graph image set SP3 as training data, training obtains CNN model.
In box 220, use the image of the 4th sample graph image set SP4 as training data, training obtains SVM classifier.
Trained CNN model is equipped on terminal device, and trained SVM classifier is equipped on server.
Fig. 3 shows the signaling process figure of cloth inspection method provided in an embodiment of the present invention, and this method 300 includes:
Step S302, terminal device receive image to be detected T.
Optionally, image to be detected T can be acquired by CCD industrial camera, then be sent out by wired or wireless mode
It send to terminal device.Wherein above-mentioned CCD (Charge Coupled Device, photosensitive coupling component) is to be used in digital camera
Record the semiconductor subassembly of light variation.When it is implemented, CCD industrial camera can by the network switch or other modes with
Terminal device connection.
Step S304, terminal device execute pretreatment to image T, such as, but not limited to, image T are converted to grayscale image
As etc..
Step S306, terminal device classify to pretreated image T using with trained CNN model.
Step S308, if image T is classified as normal picture by CNN model, terminal device determines that image T is included
Fault is not present in fabric surface, and then process terminates.
Step S310, if image T is classified as problem image by CNN model, it is fixed that terminal device executes fault to image T
Position and image dividing processing.Wherein, fault positioning and image dividing processing split defect regions from image, are conducive to
The subsequent quick and accurate detection to fault type.Fault positioning and image dividing processing are known technologies, herein omission pair
Its description.
Step S312, terminal device execute image enhancement and normalizing to the image T through fault positioning and image dividing processing
Change processing, to enhance information useful in image and be converted to satisfactory canonical form.At image enhancement and normalization
Reason is known technology, omits descriptions thereof herein.
Step S314, terminal device generate defect detection request.
Step S316, terminal device will request to send by the image T and defect detection of image enhancement and normalized
To server.
Step S318, server request to carry out detection point to image T using with trained SVM classifier according to defect detection
Class obtains the prediction result of the fault type for the fabric surface that image T is included.
Prediction result is sent to terminal device by step S320, server.
Step S322, terminal device receives the prediction result that server is sent, and executes corresponding response action, then flows
Journey terminates.
Optionally, terminal device executes corresponding response action, and such as, but not limited to, Storage Estimation result passes through display
Equipment shows prediction result, executes operation etc. of alarming.
Process, which can be seen that scheme provided in an embodiment of the present invention, as shown in Figure 2 to detect fabric in local terminal
With the presence or absence of fault, when detecting, then using the type of cloud server detection fault, to pass through this distribution side there are when fault
Method, can beyond the clouds server carry out fault types of database maintenance and update, when user has new detection demand, such as need
Increase fault type identification, then only need to be updated iteration in server side, without the detection model of replacement terminal equipment,
Reduce the use cost of user.
Other modifications
It will be understood by those skilled in the art that in an embodiment of the present invention, detection point of the image to be detected in terminal device
In class process, by a series of processing, what is finally obtained can be the characteristic of image to be detected, when it is implemented, eventually
Image to be detected is sent to server and can be by end equipment is sent to server for the characteristic of image to be detected.
It will be understood by those skilled in the art that although in the above embodiments, method 300 includes executing pre- place to image T
The step S304 of reason, however, the present invention is not limited thereto.In other embodiments of the invention, such as, but not limited to, exist
It has been suitable in the case where being classified using model under the original state of image T, method 300 can not also include to image T
Execute pretreated step S304.
It will be understood by those skilled in the art that although in the above embodiments, method 300 includes carrying out fault to image T
The step S310 of positioning and image dividing processing, however, the present invention is not limited thereto.In other embodiments of the invention,
Method 300 can not also include the steps that executing image T fault positioning and image dividing processing S310.
It will be understood by those skilled in the art that although in the above embodiments, method 300 includes carrying out image to image T
The step S312 of enhancing and normalized, however, the present invention is not limited thereto.In other embodiments of the invention, side
Method 300 can not also include the steps that executing image enhancement and normalized S312 to image T.
It will be understood by those skilled in the art that although in the above embodiments, method 200 includes holding to the received image of institute
The box 204 of row image labeling processing, however, the present invention is not limited thereto.In other embodiments of the invention, such as but
It is not limited to, in the case where the 202 received image of institute of box has executed image labeling processing, method 200 can not also include
To received image execute the box 204 of image labeling processing.
It will be understood by those skilled in the art that although in the above embodiments, method 200 includes box 206-208 to spread out
Raw more sample images, however, the present invention is not limited thereto.In other embodiments of the invention, such as but do not limit to
In in the case where the quantity of existing sample image is enough, method 200 can not also include box 206-208.
It will be understood by those skilled in the art that although in the above embodiments, method 200 includes box 208-210 with flat
The quantity of weighing apparatus problem sample image and normal sample image and more sample images are obtained, however, the present invention does not limit to
In this.In other embodiments of the invention, such as, but not limited to, the problem sample graph in the 202 received image of institute of box
The quantity of picture and normal sample image be balance and quantity it is enough in the case where, method 200 can not also the side of including
Frame 208-210.
Although method 200 includes box 206 with by sample it will be understood by those skilled in the art that in the above embodiments
Image is converted into gray level image, however, the present invention is not limited thereto.In other embodiments of the invention, such as but not office
It is limited to, in the case where the 202 received image of institute of box has been gray level image, method 200 can not also include box 206.
It will be understood by those skilled in the art that although in the above embodiments, method 200 includes box 212-216 to select
The sample image for being suitable for training CNN model and classifier is taken, however, the present invention is not limited thereto.In other of the invention
In embodiment, method 200 can not also include box 212-216.
It will be understood by those skilled in the art that although in the above embodiments, the first detection model is CNN model, however,
The present invention is not limited thereto.In other embodiments of the invention, the first detection model is also possible to BP neural network model,
Or other kinds of neural network model.
It will be understood by those skilled in the art that although in the above embodiments, the second detection model is SVM classifier, so
And the present invention is not limited thereto.In other implementations of the invention, the second detection model can also be Bayes classifier,
Nearest Neighbor Classifier, linear classifier or other kinds of classifier.
Fig. 4 shows the flow chart of the fabric defect detection method of one embodiment according to the invention.Side shown in Fig. 4
Method 400 is applied to terminal device.
As shown in figure 4, method 400 may include, in box 402, image to be detected is received, the first detection model pair is utilized
Described image to be detected is classified, wherein first detection model is to belong to knit included in it for detection image
Object surface has that fault image still falls within the normal picture that fault is not present in fabric included in it；
Method 400 can also include: in box 404, if first detection model classifies described image to be detected
For described problem image, then defect detection request is generated.
Method 400 can also include: that described image to be detected and defect detection request are sent to service in box 406
Device, the server is equipped with the second detection model, wherein second detection model is used to detect the type of fault.
Method 400 can also include: to receive the prediction result for the fault type that the server is sent in box 408.
In one aspect, if described image to be detected is classified as described problem image, institute by first detection model
State method further include: fault positioning and image dividing processing are carried out to described image to be detected；Correspondingly, by the mapping to be checked
Picture and defect detection request be sent to server, comprising: will by fault positioning and image dividing processing image to be detected and
Defect detection request is sent to server.
On the other hand, first detection model is convolutional neural networks model and second detection model
Including Bayes classifier, Nearest Neighbor Classifier, linear classifier, SVM classifier one of which.
Fig. 5 shows the flow chart of the fabric defect detection method of another embodiment according to the invention.It is shown in fig. 5
Method 500 is applied to server.
As shown in figure 5, method 500 may include, in box 502, image to be detected and defect detection request are received.
Method 500 can also include: to be requested according to the defect detection, in box 504 using the second detection model to institute
It states image to be detected to be detected, wherein second detection model is used to detect the type of fault.
Method 500 can also include: to generate and export the prediction result of the fault type in box 506.
Fig. 6 shows the schematic diagram of the fabric defects detection device of one embodiment according to the invention, dress shown in fig. 6
Setting 600 can use the mode of software, hardware or software and hardware combining to realize.Device 600 is applied to terminal device.
As shown in fig. 6, device 600 may include detection categorization module 602, request generation module 604, request sending module
606 and result receiving module 608.Detection categorization module 602 is for receiving image to be detected, using the first detection model to described
Image to be detected is classified, wherein first detection model is to belong to fabric table included in it for detection image
Face has that fault image still falls within the normal picture that fault is not present in fabric included in it.Request generation module
If described image to be detected is classified as described problem image for first detection model by 604, defect detection is generated
Request.Request sending module 606 is used to described image to be detected and defect detection request being sent to server, the server
Equipped with the second detection model, wherein second detection model is used to detect the type of fault.As a result receiving module 608 is used
In the prediction result for receiving the fault type that the server is sent.
In one aspect, if described image to be detected is classified as described problem image, institute by first detection model
Stating device further includes locating segmentation module, for carrying out fault positioning and image dividing processing to described image to be detected；Accordingly
, the request sending module by image to be detected and defect detection of fault positioning and image dividing processing for that will request
It is sent to server.
On the other hand, first detection model is convolutional neural networks model and second detection model
Including Bayes classifier, Nearest Neighbor Classifier, linear classifier, SVM classifier one of which.
Fig. 7 shows the schematic diagram of the fabric defects detection device of another embodiment according to the invention, shown in Fig. 7
Device 700 can use the mode of software, hardware or software and hardware combining to realize.Device 700 is applied to server.
As shown in fig. 7, device 700 may include request receiving module 702, fault categorization module 704 and result output mould
Block 706.Request receiving module 702 is for receiving image to be detected and defect detection request.Fault categorization module 704 is used for basis
The defect detection request, detects described image to be detected using the second detection model, wherein the second detection mould
Type is used to detect the type of fault.As a result output module 706 is used to generate and export the prediction result of the fault type.
Fig. 8 shows the schematic diagram of the terminal device of one embodiment according to the invention.As shown in figure 8, terminal device
800 may include processor 802, memory 804, communication interface 806 and communication bus 808, processor 802,804 and of memory
Communication interface 806 completes mutual communication by communication bus 808.
Memory 804 is stored with executable instruction, wherein the executable instruction makes processor 802 upon being performed
Execute method 200, method shown in Fig. 3 300 or method shown in Fig. 4 400 shown in Fig. 2.
Fig. 9 shows the schematic diagram of the server of one embodiment according to the invention.As shown in figure 9, server 900 can
To include processor 902, memory 904, communication interface 906 and communication bus 908, processor 902, memory 904 and communication
Interface 906 completes mutual communication by communication bus 908.
Memory 904 is stored with executable instruction, wherein the executable instruction makes processor 902 upon being performed
Execute method 200 or method shown in fig. 5 500 shown in Fig. 2.
The communication bus that above-mentioned terminal device and server are mentioned can be Peripheral Component Interconnect standard (Peripheral
Pomponent Interconnect, abbreviation PCI) bus or expanding the industrial standard structure (Extended Industry
Standard Architecture, abbreviation EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control
Bus processed etc..Only to be indicated with a thick line in figure convenient for indicating, it is not intended that an only bus or a type of total
Line.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, abbreviation RAM), also may include
Nonvolatile memory (non-volatile memory), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
Abbreviation CPU), network processing unit (Network Processor, abbreviation NP) etc.；It can also be digital signal processor
(Digital Signal Processing, abbreviation DSP), specific integrated circuit (Application Specific
Integrated Circuit, abbreviation ASIC), field programmable gate array (Field-Programmable Gate Array,
Abbreviation FPGA), graphics processor (Graphics Processing Unit, abbreviation GPU) or other programmable logic device,
Discrete gate or transistor logic, discrete hardware components.
The present invention also provides a kind of storage mediums, are stored thereon with computer executable instructions, wherein the executable finger
Order makes upon being performed shown in computer execution method shown in Fig. 2, method shown in Fig. 3, method shown in Fig. 4 or Fig. 5
Method.
The specific embodiment illustrated above in conjunction with attached drawing describes exemplary embodiment, it is not intended that may be implemented
Or fall into all embodiments of the protection scope of claims." exemplary " meaning of the term used in entire this specification
Taste " be used as example, example or illustration ", be not meant to than other embodiments " preferably " or " there is advantage ".For offer pair
The purpose of the understanding of described technology, specific embodiment include detail.However, it is possible in these no details
In the case of implement these technologies.In some instances, public in order to avoid the concept to described embodiment causes indigestion
The construction and device known is shown in block diagram form.
The foregoing description of present disclosure is provided so that any those of ordinary skill in this field can be realized or make
Use present disclosure.To those skilled in the art, the various modifications carried out to present disclosure are apparent
, also, can also answer generic principles defined herein in the case where not departing from the protection scope of present disclosure
For other modifications.Therefore, present disclosure is not limited to examples described herein and design, but disclosed herein with meeting
Principle and novel features widest scope it is consistent.
Claims (11)
1. fabric defect detection method, comprising:
Image to be detected is received, is classified using the first detection model to described image to be detected, wherein first detection
Model is to belong to fabric surface included in it and there are problems that fault image is still fallen within wherein to be wrapped for detection image
The normal picture of fault is not present in the fabric contained；
If described image to be detected is classified as described problem image by first detection model, generates defect detection and ask
It asks；
Described image to be detected and defect detection request are sent to server, the server equipped with the second detection model,
Wherein, second detection model is used to detect the type of fault；
Receive the prediction result for the fault type that the server is sent.
2. according to the method described in claim 1, wherein, if described image to be detected is classified as by first detection model
Described problem image, the method also includes:
Fault positioning and image dividing processing are carried out to described image to be detected；
Correspondingly, described image to be detected and defect detection request are sent to server, comprising:
Server will be sent to by image to be detected and defect detection request of fault positioning and image dividing processing.
3. according to the method described in claim 1, wherein,
First detection model is convolutional neural networks model, and
Second detection model include Bayes classifier, Nearest Neighbor Classifier, linear classifier, SVM classifier wherein one
Kind.
4. fabric defect detection method, comprising:
Receive image to be detected and defect detection request；
It is requested according to the defect detection, described image to be detected is detected using the second detection model, wherein described the
Two detection models are used to detect the type of fault；
Generate and export the prediction result of the fault type.
5. fabric defects detection device, comprising:
Detection categorization module classifies to described image to be detected using the first detection model for receiving image to be detected,
Wherein, first detection model is to belong to fabric surface included in it to there are problems that fault image for detection image
Still fall within the normal picture that fault is not present in fabric included in it；
Generation module is requested, if described image to be detected is classified as described problem image for first detection model,
Then generate defect detection request；
Request sending module, for described image to be detected and defect detection request to be sent to server, the server is taken
It is loaded with the second detection model, wherein second detection model is used to detect the type of fault；
As a result receiving module, for receiving the prediction result for the fault type that the server is sent.
6. device according to claim 5, wherein if described image to be detected is classified as by first detection model
Described problem image, described device further include:
Locating segmentation module, for carrying out fault positioning and image dividing processing to described image to be detected；
Correspondingly, the request sending module by fault for that will position image to be detected and fault with image dividing processing
Detection request is sent to server.
7. device according to claim 5, wherein
First detection model is convolutional neural networks model, and
Second detection model include Bayes classifier, Nearest Neighbor Classifier, linear classifier, SVM classifier wherein one
Kind.
8. fabric defects detection device, comprising:
Request receiving module, for receiving image to be detected and defect detection request；
Fault categorization module, for according to the defect detection request, using the second detection model to described image to be detected into
Row detection, wherein second detection model is used to detect the type of fault；
As a result output module, for generating and exporting the prediction result of the fault type.
9. terminal device, comprising:
Processor, memory, communication interface and communication bus, the processor, the memory and the communication interface pass through
The communication bus completes mutual communication；
The memory is stored with executable instruction, wherein the executable instruction holds the processor
Method described in row claim 1-3 any one.
10. server, comprising:
Processor, memory, communication interface and communication bus, the processor, the memory and the communication interface pass through
The communication bus completes mutual communication；
The memory is stored with executable instruction, wherein the executable instruction holds the processor
Row method as claimed in claim 4.
11. storage medium is stored thereon with computer executable instructions, wherein the executable instruction makes upon being performed
Computer perform claim requires method described in 1-3 any one, method as claimed in claim 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811031071.2A CN109035248A (en) | 2018-09-05 | 2018-09-05 | Defect detection method, apparatus, terminal device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811031071.2A CN109035248A (en) | 2018-09-05 | 2018-09-05 | Defect detection method, apparatus, terminal device, server and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109035248A true CN109035248A (en) | 2018-12-18 |
Family
ID=64623484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811031071.2A Pending CN109035248A (en) | 2018-09-05 | 2018-09-05 | Defect detection method, apparatus, terminal device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109035248A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580645A (en) * | 2018-12-20 | 2019-04-05 | 深圳灵图慧视科技有限公司 | Defects identification equipment |
| CN109801261A (en) * | 2018-12-20 | 2019-05-24 | 深圳灵图慧视科技有限公司 | The surface-defect detection method and equipment of long products in traveling |
| CN109881356A (en) * | 2018-12-25 | 2019-06-14 | 西安工程大学 | Hosiery machine knitting needle on-line measuring device and method based on SVM image classification |
| US20210004945A1 (en) * | 2019-07-05 | 2021-01-07 | International Business Machines Corporation | Image defect identification |
| CN113638104A (en) * | 2021-10-14 | 2021-11-12 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
| US11295439B2 (en) | 2019-10-16 | 2022-04-05 | International Business Machines Corporation | Image recovery |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207186A (en) * | 2013-04-08 | 2013-07-17 | 北京经纬纺机新技术有限公司 | Identification method for defect detection of automatic cloth inspecting machine and system thereof |
| CN106845556A (en) * | 2017-02-09 | 2017-06-13 | 东华大学 | A kind of fabric defect detection method based on convolutional neural networks |
| CN106996935A (en) * | 2017-02-27 | 2017-08-01 | 华中科技大学 | A kind of multi-level fuzzy judgment Fabric Defects Inspection detection method and system |
| WO2018008860A1 (en) * | 2016-07-06 | 2018-01-11 | 김일열 | Defect detection device |
| CN108320278A (en) * | 2018-01-09 | 2018-07-24 | 北京百度网讯科技有限公司 | Product defects detect localization method, device, equipment and computer-readable medium |
-
2018
- 2018-09-05 CN CN201811031071.2A patent/CN109035248A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207186A (en) * | 2013-04-08 | 2013-07-17 | 北京经纬纺机新技术有限公司 | Identification method for defect detection of automatic cloth inspecting machine and system thereof |
| WO2018008860A1 (en) * | 2016-07-06 | 2018-01-11 | 김일열 | Defect detection device |
| CN106845556A (en) * | 2017-02-09 | 2017-06-13 | 东华大学 | A kind of fabric defect detection method based on convolutional neural networks |
| CN106996935A (en) * | 2017-02-27 | 2017-08-01 | 华中科技大学 | A kind of multi-level fuzzy judgment Fabric Defects Inspection detection method and system |
| CN108320278A (en) * | 2018-01-09 | 2018-07-24 | 北京百度网讯科技有限公司 | Product defects detect localization method, device, equipment and computer-readable medium |
Non-Patent Citations (2)
| Title |
|---|
| 王永灿: "工业流水线织物疵点检测及分类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 邹超: "布匹疵点在线检测***研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580645A (en) * | 2018-12-20 | 2019-04-05 | 深圳灵图慧视科技有限公司 | Defects identification equipment |
| CN109801261A (en) * | 2018-12-20 | 2019-05-24 | 深圳灵图慧视科技有限公司 | The surface-defect detection method and equipment of long products in traveling |
| CN109881356A (en) * | 2018-12-25 | 2019-06-14 | 西安工程大学 | Hosiery machine knitting needle on-line measuring device and method based on SVM image classification |
| US20210004945A1 (en) * | 2019-07-05 | 2021-01-07 | International Business Machines Corporation | Image defect identification |
| WO2021005426A1 (en) * | 2019-07-05 | 2021-01-14 | International Business Machines Corporation | Image defect identification |
| US11030738B2 (en) * | 2019-07-05 | 2021-06-08 | International Business Machines Corporation | Image defect identification |
| GB2600587A (en) * | 2019-07-05 | 2022-05-04 | Ibm | Image defect identification |
| GB2600587B (en) * | 2019-07-05 | 2023-11-15 | Ibm | Image defect identification |
| US11295439B2 (en) | 2019-10-16 | 2022-04-05 | International Business Machines Corporation | Image recovery |
| CN113638104A (en) * | 2021-10-14 | 2021-11-12 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
| CN113638104B (en) * | 2021-10-14 | 2022-02-08 | 海门市恒昌织带有限公司 | Intelligent yarn cleaning control method and system for bobbin winder |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109035248A (en) | Defect detection method, apparatus, terminal device, server and storage medium | |
| US20220414856A1 (en) | A fabric defect detection method based on multi-modal deep learning | |
| CN109187579A (en) | Fabric defect detection method and device, computer equipment and computer-readable medium | |
| CN108564104A (en) | Product defects detection method, device, system, server and storage medium | |
| KR20170138559A (en) | Washing machine control method and washing machine | |
| CN109242846A (en) | Method, apparatus and equipment for fabric surface defects detection | |
| CN109215022A (en) | Cloth inspection method, device, terminal device, server, storage medium and system | |
| CN106296689B (en) | Flaw detection method, system and device | |
| CN110378887A (en) | Screen defect inspection method, apparatus and system, computer equipment and medium | |
| Shahrabadi et al. | Defect detection in the textile industry using image-based machine learning methods: a brief review | |
| CN117152484B (en) | Small target cloth flaw detection method based on improved YOLOv5s | |
| CN109348216A (en) | A kind of combination treatment method of bad point detection peace field calibration | |
| CN113822842A (en) | Industrial defect detection method based on multi-task learning | |
| CN108924547A (en) | TV motherboard test method, device, system and test host computer | |
| CN109325940A (en) | Textile detecting method and device, computer equipment and computer-readable medium | |
| CN109300117A (en) | Nerve network system, electronic equipment and machine readable media | |
| CN109325941A (en) | Surface properties detection method and device, computer equipment and computer-readable medium | |
| CN109146875A (en) | Method, apparatus and equipment for fabric surface defects detection | |
| CN109684875A (en) | Cloth detects recording method, device, equipment and storage medium | |
| CN103984967A (en) | Automatic detection system and automatic detection method applied to commodity label detection | |
| CN109166119A (en) | Fabric defect detection method, device, equipment and machine readable media | |
| CN109472771A (en) | Detection method, device and the detection device of maize male ears | |
| Wu et al. | Objective rating method for fabric pilling based on LSNet network | |
| CN212846839U (en) | Fabric information matching system | |
| CN113504975A (en) | Multi-task cooperative scheduling method for online semantic segmentation machine vision detection |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181218 |