CN108154508B - Method, apparatus, storage medium and the terminal device of product defects detection positioning - Google Patents
Method, apparatus, storage medium and the terminal device of product defects detection positioning Download PDFInfo
- Publication number
- CN108154508B CN108154508B CN201810019380.1A CN201810019380A CN108154508B CN 108154508 B CN108154508 B CN 108154508B CN 201810019380 A CN201810019380 A CN 201810019380A CN 108154508 B CN108154508 B CN 108154508B
- Authority
- CN
- China
- Prior art keywords
- product
- training
- detection model
- picture
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
- G01N2021/8861—Determining coordinates of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Chemical & Material Sciences (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention proposes method, apparatus, storage medium and the terminal device of a kind of product defects detection positioning, wherein the described method includes: choosing the neural network structure of detection model to be trained according to the image data feature and product defects complexity of product to be measured;Obtain training data;Wherein, the training data includes the standard category and normal place of product defects in trained product picture and the trained product picture;According to the detection model for each neural network structure that training data training is chosen;The best detection model of fitting degree is chosen from the detection model after training carries out product defects detection to be supplied to detection service device;Wherein, the prediction classification and predicted position of product defects in the product picture that the detection model receives described in acquisition for being calculated according to the product picture received.Using the present invention, the accuracy of testing product defect can be improved.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of method, apparatus of product defects detection positioning, storage
Medium and terminal device.
Background technique
In traditional industry manufacturing industry production scene, quality inspection is the key link in production procedure.In steel production, automobile
In the fields such as manufacture, papermaking, battery manufacture, solar panels manufacture, a kind of important means controlled product quality is pair
The surface state of product is detected, and to judge that product whether there is flaw and defect, and does phase to product according to testing result
The processing answered.In traditional industry manufacturing industry production scene, quality inspection is the key link in production procedure.In steel production, vapour
In the fields such as vehicle manufacture, papermaking, battery manufacture, solar panels manufacture, a kind of important means controlled product quality is
The surface state of product is detected, to judge that product whether there is flaw and defect, and product is done according to testing result
Corresponding processing.In the production of traditional industry manufacturing industry, this quality inspection based on product surface state is mostly manual inspection or half
It automates optical instrument and assists quality inspection, not only inefficiency, but also be easy to appear erroneous judgement, in addition, the industry that this mode generates
Data are not easy to store, manage and mining again recycles.
Wherein, there are mainly two types of modes in the application of defects detection and positioning for existing quality inspection system.First method is
Pure artificial quality inspection mode, dependent on the experience of quality inspection personnel, quality inspection personnel visually observes the appearance photo and then empirically of product
Provide judgement;Second method is the semi-automatic quality inspection mode of machine auxiliary, mainly by the quality inspection system with certain judgement
It filters out and does not have defective photo, detection judgement is then carried out by photo of the quality inspection personnel to doubtful existing defects again.
But for the first quality inspection mode and second of quality inspection mode, the process of artificial quality inspection is all referred to, quality inspection is needed
Personnel carry out walkaround inspection in production scene, and manual record gets off to do subsequent processing again after finding defect.This method is not only
Low efficiency is easy erroneous judgement of failing to judge, and data are difficult to carry out secondary use excavation, and industrial production environment is often relatively more severe, right
Quality inspection personnel or the health and safety of producers will cause adverse effect.And for second of quality detection mode, due to it
Quality inspection examining system is developed in the quality inspection system based on traditional expert system or Feature Engineering, and detection decision rule is all
It is cured in machine based on experience, it is difficult to the development iteration of business, lead to the development quality inspection system with production technology
Detection accuracy is lower and lower, or even is reduced to complete unusable state.In addition, the detection decision rule of traditional quality inspection system is all
Solidified in advance by third-party vendor within hardware, when upgrading not only needs to carry out production line key technological transformation, but also price is high
It is expensive.Traditional quality inspection system safety, standardization, in terms of all there is obvious deficiency, be unfavorable for traditional industry
The optimization and upgrading of production line.
Summary of the invention
The embodiment of the present invention provides method, apparatus, storage medium and the terminal device of a kind of product defects detection positioning, with
Solve or alleviate the above technical problem in the prior art.
In a first aspect, the embodiment of the invention provides a kind of methods of product defects detection positioning, comprising:
According to the image data feature and product defects complexity of product to be measured, the nerve of detection model to be trained is chosen
Network structure;
Obtain training data;Wherein, the training data includes producing in training product picture and the trained product picture
The standard category and normal place of product defect;
According to the detection model for each neural network structure that training data training is chosen;
The best detection model of fitting degree is chosen from the detection model after training to be supplied to the progress of detection service device
Product defects detection;Wherein, the detection model is used to be calculated according to the product picture received to obtain the reception
To product picture in product defects prediction classification and predicted position.
With reference to first aspect, in the first embodiment of first aspect, the neural network structure include convolutional layer,
Pond layer and full articulamentum;And the neural network structure for choosing detection model to be trained, comprising:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond shape of pond layer
The mapping structure of formula and full articulamentum.
With reference to first aspect, it in second of embodiment of first aspect, is selected in the detection model from after training
Take the detection model that fitting degree is best, comprising:
For each detection model after training, by the detection model to training product figure in the training data
Piece is calculated, and the prediction result of training product picture in the training data is obtained;The prediction result includes that the training produces
The prediction classification and predicted position of the product defects of product picture;And according to training in the training data and the training data
The prediction result of product picture calculates the fitting degree of the detection model;And
According to the fitting degree of each detection model after training, fitting degree is chosen from the detection model after training
Best detection model.
With reference to first aspect, in the third embodiment of first aspect, the mode of the training include before signal to
It propagates and signal errors backpropagation.
With reference to first aspect, in the 4th kind of embodiment of first aspect, the iteration cut-off condition of the training is institute
What the standard category and normal place for stating the product defects of the product picture of training data were exported with currently trained detection model
Error amount between prediction result is less than error threshold.
Second aspect, the embodiment of the present invention provide a kind of device of product defects detection positioning, including
Structure is chosen module and is chosen for the image data feature and product defects complexity according to product to be measured wait instruct
The neural network structure of experienced detection model;
Data acquisition module, for obtaining training data;Wherein, the training data includes training product picture and described
The standard category and normal place of product defects in training product picture;
Detection model training module, the inspection of each neural network structure for being chosen according to training data training
Survey model;
Detection model chooses module, for chosen from the detection model after training the best detection model of fitting degree with
It is supplied to detection service device and carries out product defects detection;Wherein, the detection model be used for according to the product picture that receives into
Row calculates, the prediction classification and predicted position of product defects in the product picture received described in acquisition.
In conjunction with second aspect, in the first embodiment of second aspect, the neural network structure include convolutional layer,
Pond layer and full articulamentum;And the structure is chosen module and is specifically used for:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond shape of pond layer
The mapping structure of formula and full articulamentum.
In conjunction with second aspect, in second of embodiment of second aspect, the detection model chooses module and includes:
Prediction and the Fitting Calculation unit, for passing through the detection model pair for each detection model after training
Training product picture is calculated in the training data, obtains the prediction result of training product picture in the training data;
The prediction result includes the prediction classification and predicted position of the product defects of the training product picture;And according to the training
The prediction result of training product picture, calculates the fitting degree of the detection model in data and the training data;And
Model selection unit, for the fitting degree according to each detection model after training, from the detection after training
The best detection model of fitting degree is chosen in model.
In conjunction with second aspect, in the third embodiment of second aspect, the training mode include signal before
To propagation and signal errors backpropagation.
The function of described device can also execute corresponding software realization by hardware realization by hardware.It is described
Hardware or software include one or more modules corresponding with above-mentioned function.
It is described to deposit including processor and memory in the structure of product defects detection positioning in a possible design
Reservoir is used to store the side for supporting the device of product defects detection positioning to execute product defects detection positioning in above-mentioned first aspect
The program of method, the processor is configured to for executing the program stored in the memory.The traffic guidance figure generates
Device can also include communication interface, for product defects detection positioning device and other equipment or communication.
The third aspect, the embodiment of the present invention provide a kind of method of product defects detection positioning, comprising:
Receive the product picture of product to be measured;
The product picture is calculated according to currently stored detection model, product in the product picture is obtained and lacks
Sunken prediction classification and predicted position;Wherein, the detection model is from all selections generated according to training data training
The best detection model of fitting degree is chosen in the detection model of neural network structure, the neural network structure of the selection is root
It is chosen according to the image data feature and product defects complexity of the product to be measured, the training data includes training product figure
The standard category and normal place of product defects in piece and the trained product picture.
In conjunction with the third aspect, in the first embodiment of the third aspect, the neural network structure include convolutional layer,
Pond layer and full articulamentum;
The convolutional layer is used to be scanned convolution to product picture using the different convolution kernel of weight, therefrom extracts various
The feature of meaning, and export into characteristic pattern;
The pond layer, for carrying out dimensionality reduction operation to the characteristic pattern;
The full articulamentum, the feature for that will extract carry out mapped location.
Fourth aspect, the embodiment of the present invention provide a kind of device of product defects detection positioning, comprising:
Product picture receiving module, for receiving the product picture of product to be measured;
Predictor computation module obtains institute for calculating according to currently stored detection model the product picture
State the prediction classification and predicted position of product defects in product picture;Wherein, the detection model is instructed from according to training data
Practice and chooses the best detection model of fitting degree, the selection in the detection model of the neural network structure of all selections generated
Neural network structure be to be chosen according to the image data feature and product defects complexity of the product to be measured, the training
Data include the standard category and normal place of product defects in trained product picture and the trained product picture.
The function of stating device can also execute corresponding software realization by hardware realization by hardware.It is described hard
Part or software include one or more modules corresponding with above-mentioned function.
It is described to deposit including processor and memory in the structure of product defects detection positioning in a possible design
Reservoir is used to store the side for supporting the device of product defects detection positioning to execute product defects detection positioning in above-mentioned first aspect
The program of method, the processor is configured to for executing the program stored in the memory.The traffic guidance figure generates
Device can also include communication interface, for product defects detection positioning device and other equipment or communication.
5th aspect, the embodiment of the invention provides a kind of computer readable storage mediums, fixed for product defects detection
Computer software instructions used in the device of position comprising for executing the side of product defects detection positioning in above-mentioned first aspect
Method is program involved in the device of product defects detection positioning in above-mentioned second aspect, or for executing the above-mentioned third aspect
The method of middle product defects detection positioning is program involved in the device of product defects detection positioning in above-mentioned fourth aspect.
Any one technical solution in above-mentioned technical proposal have the following advantages that or the utility model has the advantages that
The embodiment of the present invention waits instructing by the image data feature and product defects complexity according to product to be measured, pre-selection
Multiple neural network structures of experienced detection model, then using training data respectively to each neural network structure of selection
Detection model be trained, and choose the best detection model of fitting degree from the detection model after training, mind be contemplated
Influence of the organizational form through network structure to detection model training, and can be according to image data feature and product defects
Complexity chooses neural network structure, then the detection model that accordingly generates can be to the deformation on picture, fuzzy, illumination variation
Etc. robustness with higher, product product defects complexity is adapted to, effectively improves the accuracy of testing product defect.
Above-mentioned general introduction is merely to illustrate that the purpose of book, it is not intended to be limited in any way.Except foregoing description
Signal, except the aspect of property, embodiment and feature, by reference to attached drawing and the following detailed description, the present invention is further
Aspect, embodiment and feature will be and be readily apparent that.
Detailed description of the invention
In the accompanying drawings, unless specified otherwise herein, otherwise indicate the same or similar through the identical appended drawing reference of multiple attached drawings
Component or element.What these attached drawings were not necessarily to scale.It should be understood that these attached drawings depict only according to the present invention
Disclosed some embodiments, and should not serve to limit the scope of the present invention.
Fig. 1 is the flow diagram of the method for the product defects detection positioning that first embodiment of the invention provides;
Fig. 2 is the schematic illustration for the detection model that first embodiment of the invention provides;
Fig. 3 is the schematic illustration for the depth convolutional neural networks that first embodiment of the invention provides;
Fig. 4 is the process that the detection model of the method for the product defects detection positioning that first embodiment of the invention provides is chosen
Schematic diagram;
Fig. 5 is the structural schematic diagram of the device for the product defects detection positioning that second embodiment of the invention provides;
Fig. 6 is the flow diagram of the method for the product defects detection positioning that third embodiment of the invention provides;
Fig. 7 is the structural schematic diagram of the device for the product defects detection positioning that fourth embodiment of the invention provides;
Fig. 8 is the structural schematic diagram for the terminal device that third embodiment of the invention provides.
Specific embodiment
Hereinafter, certain exemplary embodiments are simply just described.As one skilled in the art will recognize that
Like that, without departing from the spirit or scope of the present invention, described embodiment can be modified by various different modes.
Therefore, attached drawing and description are considered essentially illustrative rather than restrictive.
Embodiment one
Referring to Fig. 1, being held the embodiment of the invention provides a kind of method of product defects detection positioning by training engine
Row, including step S10 is to step S40, it is specific as follows:
S10 chooses detection model to be trained according to the image data feature and product defects complexity of product to be measured
Neural network structure.
In embodiments of the present invention, when detecting to the surface state of product to be measured, the equipment such as video camera can be passed through
The product picture collection for acquiring product to be measured leads to the figure of picture due to the influence of picture pick-up device, photography technology and imaging environment
Sheet data feature has differences, for example, when the picture is high definition, the convolutional layer of neural network structure selected by detection model
Neuronal quantity can be reduced.And different types of product it includes the types of product defects be also different, for example,
In the case that product defects type is more, the neuronal quantity of the convolutional layer of neural network structure selected by detection model can be with
Increase, and the mapping structure of full articulamentum can also be adjusted further.In the specific implementation, neural network structure is pre-selection,
Can according to train come the detection effect of detection model further adjust.
As shown in Fig. 2, its schematic illustration for the detection model of the present embodiment.In the present embodiment, the detection mould
Type may include depth convolutional neural networks and defect location sorter network.
Wherein, the depth convolutional neural networks are used to extract the feature of product picture, and the feature is input to institute
It states in defect location sorter network.The depth convolutional neural networks can extract the characteristic information of product picture, such as may
The parts of images of existing defects extracts.
The defect location sorter network for judge the extracted feature of depth convolutional neural networks whether existing defects,
And obtain position and the classification of the defect.The defect location sorter network can use the objects such as RCNN, SSD, Mask RCNN
Physical examination survey and Image Segmentation Model, judge current feature whether existing defects, and if it exists, then judge the opposite position of current defect
It sets and generic.If there are many places defects in current feature, the relative position coordinates and classification of each defect are provided.
As shown in figure 3, it is the schematic illustration of depth convolutional neural networks provided in an embodiment of the present invention.By production line
On input of the original image as model, and the output of model is then to represent the label of defect classification, such as 0-n is respectively represented
N+1 class defect.Wherein, the structure of depth convolutional neural networks mainly by convolutional layer, pond layer, it is complete connect layer and connect etc. form.Its
In, convolution is scanned to original image or characteristic pattern (feature map) using weight different convolution kernel in convolutional layer, from
The middle feature for extracting various meanings, and export into characteristic pattern.Pond layer then carries out dimensionality reduction operation, keeping characteristics figure to characteristic pattern
In main feature.It, can be to photo on production line using this deep neural network model operated with convolution, pondization
The robustness with higher such as deformation, fuzzy, illumination variation, for classification task have it is higher can generalization.Full articulamentum
Then by the Feature Mapping of extraction to defect location sorter network.For different image data feature and product defects complexity,
It can be made up of design different number neuron, different pond forms and different full connection mapping structures corresponding
Deep neural network model, and be trained.
S20 obtains training data;Wherein, the training data includes training product picture and the trained product picture
The standard category and normal place of middle product defects.
In embodiments of the present invention, training data can be stored in tranining database, and training product picture can be pre-
First shooting, collecting gets well and detects identification to be stored in tranining database after the completion, in addition, the detection service device of operation detection model
Carrying out the testing result that detection calculating obtains to product picture also can be stored in tranining database.Carrying out convolutional neural networks
Training when, training data includes a large amount of image data with product defects and the image data pair without product defects
Convolutional neural networks are trained.
S30, according to the detection model for each neural network structure that training data training is chosen.That is, for choosing
Each neural network structure taken is all made of the training data and instructs to the corresponding detection model of the neural network structure
Practice, updates the corresponding detection model of the neural network structure.
Wherein, the iteration cut-off condition of the training of each model is the product defects of the product picture of the training data
Standard category and normal place and the prediction result of currently trained detection model output between error amount be less than error threshold
Value.
S40 chooses the best detection model of fitting degree, from the detection model after training to be supplied to detection service device
Carry out product defects detection;Wherein, the detection model is used to be calculated described in acquisition according to the product picture received
The prediction classification and predicted position of product defects in the product picture received.
In embodiments of the present invention, it can control multiple trained subelements to be respectively trained each detection model,
And when the lazy weight of training subelement, detection model can be trained by batch.
As shown in figure 4, the specific implementation process of this step S40 is as follows:
S41 produces each detection model after training by the detection model to training in the training data
Product picture is calculated, and the prediction result of training product picture in the training data is obtained;The prediction result includes the instruction
Practice the prediction classification and predicted position of the product defects of product picture;And according in the training data and the training data
The prediction result of training product picture, calculates the fitting degree of the detection model.
When training detection model, can continue to calculate each detection model using training data as examination criteria
The fitting degree of result is exported, it can be in a manner of linear fit, for example, least square method fitting or minimum through being returned to error
The modes such as return.
S42 chooses fitting according to the fitting degree of each detection model after training from the detection model after training
The best detection model of degree.
Embodiment two
Referring to Fig. 5, the embodiment of the present invention provides a kind of device of product defects detection positioning, comprising:
Structure chooses module 10, for the image data feature and product defects complexity according to product to be measured, choose to
The neural network structure of trained detection model;
Data acquisition module 20, for obtaining training data;Wherein, the training data includes training product picture and institute
State the standard category and normal place of product defects in trained product picture;
Detection model training module 30, for training each neural network structure chosen according to the training data
Detection model;
Detection model chooses module 40, for choosing the best detection model of fitting degree from the detection model after training
Product defects detection is carried out to be supplied to detection service device;Wherein, the detection model is used for according to the product picture received
It is calculated, the prediction classification and predicted position of product defects in the product picture received described in acquisition.
In conjunction with the embodiments two, in the first embodiment of embodiment two, the neural network structure include convolutional layer,
Pond layer and full articulamentum;And the structure is chosen module and is specifically used for:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond shape of pond layer
The mapping structure of formula and full articulamentum.
In conjunction with the embodiments two, in second of embodiment of embodiment two, the detection model chooses module and includes:
Prediction and the Fitting Calculation unit, for passing through the detection model pair for each detection model after training
Training product picture is calculated in the training data, obtains the prediction result of training product picture in the training data;
The prediction result includes the prediction classification and predicted position of the product defects of the training product picture;And according to the training
The prediction result of training product picture, calculates the fitting degree of the detection model in data and the training data;And
Model selection unit, for the fitting degree according to each detection model after training, from the detection after training
The best detection model of fitting degree is chosen in model.
In conjunction with the embodiments two, in the third embodiment of embodiment two, the training mode include signal before
To propagation and signal errors backpropagation.
The function of described device can also execute corresponding software realization by hardware realization by hardware.It is described
Hardware or software include one or more modules corresponding with above-mentioned function.
Embodiment three
Referring to Fig. 6, the embodiment of the present invention provides a kind of method of product defects detection positioning, it can be by detection service device
It executes, comprising:
S210 receives the product picture of product to be measured;
S220 calculates the product picture according to currently stored detection model, obtains in the product picture
The prediction classification and predicted position of product defects;Wherein, the detection model is all from being generated according to training data training
The best detection model of fitting degree, the neural network knot of the selection are chosen in the detection model of the neural network structure of selection
Structure is chosen according to the image data feature and product defects complexity of the product to be measured, and the training data includes training
The standard category and normal place of product defects in product picture and the trained product picture.
Further, when detection model output prediction classification be existing defects, then can execute corresponding response action,
Such as: alarm operation can be executed, for reminding staff to check etc..Furthermore it is also possible to by real-time detection log into
Row storage.
And referring to Fig. 3, the neural network structure includes convolutional layer, pond layer and full articulamentum;
The convolutional layer is used to be scanned convolution to product picture using the different convolution kernel of weight, therefrom extracts various
The feature of meaning, and export into characteristic pattern;
The pond layer, for carrying out dimensionality reduction operation to the characteristic pattern;
The full articulamentum, the feature for that will extract carry out mapped location.
The technical effect that the embodiment of the present invention is implemented is consistent with implementation one, and details are not described herein.
Example IV
Referring to Fig. 7, the embodiment of the present invention provides a kind of device of product defects detection positioning, comprising:
Product picture receiving module 210, for receiving the product picture of product to be measured;
Predictor computation module 220 is obtained for being calculated according to currently stored detection model the product picture
The prediction classification and predicted position of product defects in the product picture;Wherein, the detection model is from according to training data
The best detection model of fitting degree, the choosing are chosen in the detection model of the neural network structure for all selections that training generates
The neural network structure taken is chosen according to the image data feature and product defects complexity of the product to be measured, the instruction
Practice the standard category and normal place that data include product defects in trained product picture and the trained product picture.
The function of described device can also execute corresponding software realization by hardware realization by hardware.It is described
Hardware or software include one or more modules corresponding with above-mentioned function.
It is described to deposit including processor and memory in the structure of product defects detection positioning in a possible design
Reservoir is used to store the side for supporting the device of product defects detection positioning to execute product defects detection positioning in above-mentioned first aspect
The program of method, the processor is configured to for executing the program stored in the memory.The traffic guidance figure generates
Device can also include communication interface, for product defects detection positioning device and other equipment or communication.
Embodiment five
The embodiment of the present invention also provides a kind of terminal device, as shown in figure 8, the equipment includes: memory 21 and processor
22, the computer program that can be run on processor 22 is stored in memory 21.Processor 22 executes the computer program
The method of product defects detection positioning in Shi Shixian above-described embodiment.The quantity of memory 21 and processor 22 can be one
Or it is multiple.
The equipment further include:
Communication interface 23, for the communication between processor 22 and peripheral equipment.
Memory 21 may include high speed RAM memory, it is also possible to further include nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage.
If memory 21, processor 22 and the independent realization of communication interface 23, memory 21, processor 22 and communication
Interface 23 can be connected with each other by bus and complete mutual communication.The bus can be industry standard architecture
(ISA, Industry Standard Architecture) bus, external equipment interconnection (PCI, Peripheral
Component) bus or extended industry-standard architecture (EISA, Extended Industry Standard
Component) bus etc..The bus can be divided into address bus, data/address bus, control bus etc..For convenient for expression, Fig. 8
In only indicated with a thick line, it is not intended that an only bus or a type of bus.
Optionally, in specific implementation, if memory 21, processor 22 and communication interface 23 are integrated in chip piece
On, then memory 21, processor 22 and communication interface 23 can complete mutual communication by internal interface.
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 included at least one embodiment or example of the invention.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the spy of different embodiments or examples described in this specification and different embodiments or examples
Sign is combined.
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." first " is defined as a result, the feature of " second " can be expressed or hidden
It include at least one this feature containing ground.In the description of the present invention, the meaning of " plurality " is two or more, unless otherwise
Clear specific restriction.
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 specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention 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 of the invention
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.
Computer-readable medium described in the embodiment of the present invention can be computer-readable signal media or computer can
Read storage medium either the two any combination.The more specific example of computer readable storage medium is at least (non-poor
Property list to the greatest extent) include the following: there is the electrical connection section (electronic device) of one or more wirings, 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 read-only memory (CDROM).In addition, computer readable storage medium even can be with
It is the paper or other suitable media that can print described program on it, because can be for example by paper or the progress of other media
Optical scanner is then edited, interpreted or is handled when necessary with other suitable methods and is described electronically to obtain
Program is then stored in computer storage.
In embodiments of the present invention, computer-readable signal media may include in a base band or as carrier wave a part
The data-signal of propagation, wherein carrying computer-readable program code.The data-signal of this propagation can use a variety of
Form, including but not limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media is also
It can be any computer-readable medium other than computer readable storage medium, which can send, pass
It broadcasts or transmits for instruction execution system, input method or device use or program in connection.Computer can
The program code for reading to include on medium can transmit with any suitable medium, including but not limited to: wirelessly, electric wire, optical cable, penetrate
Frequently (Radio Frequency, RF) etc. or above-mentioned any appropriate combination.
It should be appreciated that each section of the invention 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.It, and in another embodiment, can be under well known in the art for example, if realized with hardware
Any one of column technology or their combination are realized: having a logic gates for realizing logic function to data-signal
Discrete logic, with suitable combinational logic gate circuit specific integrated circuit, programmable gate array (PGA), scene
Programmable 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, each functional unit in each embodiment of the present invention can integrate in a processing module
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 readable storage medium storing program for executing.The storage medium can be read-only memory, disk or CD etc..
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in its various change or replacement,
These should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the guarantor of the claim
It protects subject to range.
Claims (8)
1. a kind of method of product defects detection positioning characterized by comprising
According to the image data feature and product defects complexity of product to be measured, the neural network of detection model to be trained is chosen
Structure;
Obtain training data;Wherein, the training data includes that product in product picture and the trained product picture is trained to lack
Sunken standard category and normal place;
According to the detection model for each neural network structure that training data training is chosen;
The best detection model of fitting degree is chosen from the detection model after training carries out product to be supplied to detection service device
Defects detection;Wherein, the detection model is used to be calculated to receive described in acquisition according to the product picture received
The prediction classification and predicted position of product defects in product picture;
Wherein, the iteration cut-off condition of the training be the training data product picture product defects standard category and
Error amount between normal place and the prediction result of currently trained detection model output is less than error threshold;
Wherein, the neural network structure includes convolutional layer, pond layer and full articulamentum;And described choose detection to be trained
The neural network structure of model, comprising:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond form of pond layer and
The mapping structure of full articulamentum;
The best detection model of fitting degree is chosen in the detection model from after training, comprising:
For each detection model after training, by the detection model in the training data training product picture into
Row calculates, and obtains the prediction result of training product picture in the training data;The prediction result includes the training product figure
The prediction classification and predicted position of the product defects of piece;And according to training product in the training data and the training data
The prediction result of picture calculates the fitting degree of the detection model;And
According to the fitting degree of each detection model after training, it is best that fitting degree is chosen from the detection model after training
Detection model.
2. the method for product defects detection positioning as described in claim 1, which is characterized in that the mode of the training includes letter
Number propagated forward and signal errors backpropagation.
3. a kind of method of product defects detection positioning characterized by comprising
Receive the product picture of product to be measured;
The product picture is calculated according to currently stored detection model, obtains product defects in the product picture
Predict classification and predicted position;Wherein, the detection model is the nerve from all selections generated according to training data training
The best detection model of fitting degree is chosen in the detection model of network structure, the neural network structure of the selection is according to institute
State product to be measured image data feature and product defects complexity choose, the training data include training product picture and
The standard category and normal place of product defects in the trained product picture;
Wherein, the iteration cut-off condition of the training be the training data product picture product defects standard category and
Error amount between normal place and the prediction result of currently trained detection model output is less than error threshold;
Wherein, the neural network structure includes convolutional layer, pond layer and full articulamentum;And the neural network knot of the selection
Structure is chosen according to the image data feature and product defects complexity of the product to be measured, comprising:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond form of pond layer and
The mapping structure of full articulamentum;
Fitting journey is chosen in the detection model of the neural network structure from all selections generated according to training data training
Spending best detection model includes:
For each detection model after training, by the detection model in the training data training product picture into
Row calculates, and obtains the prediction result of training product picture in the training data;The prediction result includes the training product figure
The prediction classification and predicted position of the product defects of piece;And according to training product in the training data and the training data
The prediction result of picture calculates the fitting degree of the detection model;And
According to the fitting degree of each detection model after training, it is best that fitting degree is chosen from the detection model after training
Detection model.
4. the method for product defects detection positioning as claimed in claim 3, which is characterized in that
The convolutional layer is used to be scanned convolution to product picture using the different convolution kernel of weight, therefrom extracts various meanings
Feature, and export into characteristic pattern;
The pond layer, for carrying out dimensionality reduction operation to the characteristic pattern;
The full articulamentum, the feature for that will extract carry out mapped location.
5. a kind of device of product defects detection positioning, which is characterized in that including
Structure is chosen module and is chosen to be trained for the image data feature and product defects complexity according to product to be measured
The neural network structure of detection model;
Data acquisition module, for obtaining training data;Wherein, the training data includes training product picture and the training
The standard category and normal place of product defects in product picture;
Detection model training module, the detection mould of each neural network structure for being chosen according to training data training
Type;
Detection model chooses module, for choosing the best detection model of fitting degree from the detection model after training to provide
Product defects detection is carried out to detection service device;Wherein, the detection model according to the product picture received based on carrying out
It calculates, the prediction classification and predicted position of product defects in the product picture received described in acquisition;
Wherein, the iteration cut-off condition of the training be the training data product picture product defects standard category and
Error amount between normal place and the prediction result of currently trained detection model output is less than error threshold;
Wherein, the neural network structure includes convolutional layer, pond layer and full articulamentum;And to choose module specific for the structure
For:
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond form of pond layer and
The mapping structure of full articulamentum;
The detection model chooses module
Prediction and the Fitting Calculation unit, for for each detection model after training, passing through the detection model to described
Training product picture is calculated in training data, obtains the prediction result of training product picture in the training data;It is described
Prediction result includes the prediction classification and predicted position of the product defects of the training product picture;And according to the training data
With the prediction result of training product picture in the training data, the fitting degree of the detection model is calculated;And
Model selection unit, for the fitting degree according to each detection model after training, from the detection model after training
It is middle to choose the best detection model of fitting degree.
6. a kind of device of product defects detection positioning characterized by comprising
Product picture receiving module, for receiving the product picture of product to be measured;
Predictor computation module obtains the production for calculating according to currently stored detection model the product picture
The prediction classification and predicted position of product defects in product picture;Wherein, the detection model is given birth to from according to training data training
At all selections neural network structure detection model in choose the best detection model of fitting degree, the mind of the selection
It through network structure is chosen according to the image data feature and product defects complexity of the product to be measured, the training data
Standard category and normal place including product defects in training product picture and the trained product picture;
Wherein, the iteration cut-off condition of the training be the training data product picture product defects standard category and
Error amount between normal place and the prediction result of currently trained detection model output is less than error threshold
Wherein, the neural network structure includes convolutional layer, pond layer and full articulamentum;And the neural network knot of the selection
Structure is to include: according to what the image data feature and product defects complexity of the product to be measured were chosen
Choose the neuronal quantity of the convolutional layer of the neural network structure of detection model to be trained, the pond form of pond layer and
The mapping structure of full articulamentum;
Fitting journey is chosen in the detection model of the neural network structure from all selections generated according to training data training
Spending best detection model includes:
For each detection model after training, by the detection model in the training data training product picture into
Row calculates, and obtains the prediction result of training product picture in the training data;The prediction result includes the training product figure
The prediction classification and predicted position of the product defects of piece;And according to training product in the training data and the training data
The prediction result of picture calculates the fitting degree of the detection model;And
According to the fitting degree of each detection model after training, it is best that fitting degree is chosen from the detection model after training
Detection model.
7. a kind of terminal device for realizing product defects detection positioning, which is characterized in that the terminal device includes:
One or more processors;
Storage device, for storing one or more programs;
When one or more of programs are executed by one or more of processors, so that one or more of processors
The method for realizing the product defects detection positioning as described in any in claim 1-4.
8. a kind of computer readable storage medium, is stored with computer program, which is characterized in that the program is executed by processor
The method of product defects detection positioning of the Shi Shixian as described in any in claim 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810019380.1A CN108154508B (en) | 2018-01-09 | 2018-01-09 | Method, apparatus, storage medium and the terminal device of product defects detection positioning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810019380.1A CN108154508B (en) | 2018-01-09 | 2018-01-09 | Method, apparatus, storage medium and the terminal device of product defects detection positioning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108154508A CN108154508A (en) | 2018-06-12 |
CN108154508B true CN108154508B (en) | 2019-05-24 |
Family
ID=62461007
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810019380.1A Active CN108154508B (en) | 2018-01-09 | 2018-01-09 | Method, apparatus, storage medium and the terminal device of product defects detection positioning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108154508B (en) |
Families Citing this family (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108876781A (en) * | 2018-06-26 | 2018-11-23 | 广东工业大学 | Surface defect recognition method based on SSD algorithm |
CN108921840A (en) * | 2018-07-02 | 2018-11-30 | 北京百度网讯科技有限公司 | Display screen peripheral circuit detection method, device, electronic equipment and storage medium |
CN108921839A (en) * | 2018-07-02 | 2018-11-30 | 北京百度网讯科技有限公司 | Continuous casting billet quality detection method, device, electronic equipment and storage medium |
CN109087280A (en) * | 2018-07-02 | 2018-12-25 | 北京百度网讯科技有限公司 | Continuous casting billet quality detection method, device, control equipment, server and medium |
CN108961238A (en) | 2018-07-02 | 2018-12-07 | 北京百度网讯科技有限公司 | Display screen quality determining method, device, electronic equipment and storage medium |
CN110717881A (en) * | 2018-07-12 | 2020-01-21 | 长鑫存储技术有限公司 | Wafer defect identification method and device, storage medium and terminal equipment |
CN109187549A (en) * | 2018-07-26 | 2019-01-11 | 上海工程技术大学 | A kind of detection method for backing layer edging defect detection station after rearview mirror |
CN108982522B (en) * | 2018-08-09 | 2022-04-26 | 北京百度网讯科技有限公司 | Method and apparatus for detecting pipe defects |
CN109086827A (en) * | 2018-08-10 | 2018-12-25 | 北京百度网讯科技有限公司 | Method and apparatus for detecting monocrystaline silicon solar cell defect |
CN108960214A (en) * | 2018-08-17 | 2018-12-07 | 中控智慧科技股份有限公司 | Fingerprint enhancement binarization method, device, equipment, system and storage medium |
CN109239075B (en) | 2018-08-27 | 2021-11-30 | 北京百度网讯科技有限公司 | Battery detection method and device |
CN109544522A (en) * | 2018-11-12 | 2019-03-29 | 北京科技大学 | A kind of Surface Defects in Steel Plate detection method and system |
CN109598712A (en) * | 2018-11-30 | 2019-04-09 | 北京百度网讯科技有限公司 | Quality determining method, device, server and the storage medium of plastic foam cutlery box |
CN109584235A (en) * | 2018-11-30 | 2019-04-05 | 北京百度网讯科技有限公司 | Quality determining method, device, server and the storage medium of plastic foam cutlery box |
CN109919925A (en) * | 2019-03-04 | 2019-06-21 | 联觉(深圳)科技有限公司 | Printed circuit board intelligent detecting method, system, electronic device and storage medium |
CN110120044A (en) * | 2019-05-20 | 2019-08-13 | 北京百度网讯科技有限公司 | Article defects detection method and device, edge device, server |
CN110675359A (en) * | 2019-06-29 | 2020-01-10 | 创新奇智(南京)科技有限公司 | Defect sample generation method and system for steel coil surface and electronic equipment |
CN110349145B (en) * | 2019-07-09 | 2022-08-16 | 京东方科技集团股份有限公司 | Defect detection method, defect detection device, electronic equipment and storage medium |
CN111369517B (en) * | 2020-02-28 | 2023-08-01 | 创新奇智(合肥)科技有限公司 | Automatic quality inspection method and device for solar panel, electronic equipment and storage medium |
CN111709455A (en) * | 2020-05-22 | 2020-09-25 | 五邑大学 | Evaluation method of wine bottle surface defect model, electronic device and storage medium |
CN111693534B (en) | 2020-06-12 | 2023-09-15 | 北京百度网讯科技有限公司 | Surface defect detection method, model training method, device, equipment and medium |
CN111929327A (en) * | 2020-09-09 | 2020-11-13 | 深兰人工智能芯片研究院(江苏)有限公司 | Cloth defect detection method and device |
CN112669271A (en) * | 2020-12-22 | 2021-04-16 | 深圳酷派技术有限公司 | Object surface defect detection method, related device and computer storage medium |
CN112907533A (en) * | 2021-02-10 | 2021-06-04 | 武汉精测电子集团股份有限公司 | Detection model training method, device, equipment and readable storage medium |
CN114119554A (en) * | 2021-11-29 | 2022-03-01 | 哈尔滨工业大学 | Surface microdefect detection method and device based on convolutional neural network |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106355579A (en) * | 2016-09-14 | 2017-01-25 | 南京理工大学 | Defect detecting method of cigarette carton surface wrinkles |
CN106485324A (en) * | 2016-10-09 | 2017-03-08 | 成都快眼科技有限公司 | A kind of convolutional neural networks optimization method |
CN106845556A (en) * | 2017-02-09 | 2017-06-13 | 东华大学 | A kind of fabric defect detection method based on convolutional neural networks |
CN107123114A (en) * | 2017-04-21 | 2017-09-01 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of cloth defect inspection method and device based on machine learning |
CN107133943B (en) * | 2017-04-26 | 2018-07-06 | 贵州电网有限责任公司输电运行检修分公司 | A kind of visible detection method of stockbridge damper defects detection |
CN107316295A (en) * | 2017-07-02 | 2017-11-03 | 苏州大学 | A kind of fabric defects detection method based on deep neural network |
CN107316300B (en) * | 2017-07-17 | 2020-12-15 | 杭州盈格信息技术有限公司 | Tire X-ray defect detection method based on deep convolutional neural network |
CN107451997A (en) * | 2017-07-31 | 2017-12-08 | 南昌航空大学 | A kind of automatic identifying method of the welding line ultrasonic TOFD D scanning defect types based on deep learning |
-
2018
- 2018-01-09 CN CN201810019380.1A patent/CN108154508B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN108154508A (en) | 2018-06-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108154508B (en) | Method, apparatus, storage medium and the terminal device of product defects detection positioning | |
CN108257121B (en) | Method, apparatus, storage medium and the terminal device that product defects detection model updates | |
CN108230318B (en) | Ladle defects detection classification method, device, equipment and computer-readable medium | |
Li et al. | Automatic pixel‐level multiple damage detection of concrete structure using fully convolutional network | |
CN111739075B (en) | Deep network lung texture recognition method combining multi-scale attention | |
CN108320278A (en) | Product defects detect localization method, device, equipment and computer-readable medium | |
CN108230317A (en) | Steel plate defect detection sorting technique, device, equipment and computer-readable medium | |
CN109241967B (en) | Thyroid ultrasound image automatic identification system based on deep neural network, computer equipment and storage medium | |
CN101103373B (en) | Method for improved image segmentation | |
CN106296636B (en) | Print the detection method and detection means of image | |
CN105574550A (en) | Vehicle identification method and device | |
CN103743452B (en) | Contact-free level measures system | |
CN104913797A (en) | Pointer type instrument number reading and recognition method and system | |
CN109886928A (en) | A kind of target cell labeling method, device, storage medium and terminal device | |
CN108711148A (en) | A kind of wheel tyre defect intelligent detecting method based on deep learning | |
CN110264444A (en) | Damage detecting method and device based on weak segmentation | |
CN104363815A (en) | Image processing device, image processing method, and image processing program | |
CN111950812B (en) | Method and device for automatically identifying and predicting rainfall | |
CN110321933A (en) | A kind of fault recognition method and device based on deep learning | |
CN112819821B (en) | Cell nucleus image detection method | |
CN111598098A (en) | Water gauge water line detection and effectiveness identification method based on full convolution neural network | |
CN112614133A (en) | Three-dimensional pulmonary nodule detection model training method and device without anchor point frame | |
CN113951834A (en) | Alzheimer disease classification prediction method based on visual Transformer algorithm | |
CN111563439A (en) | Aquatic organism disease detection method, device and equipment | |
CN111931721B (en) | Method and device for detecting color and number of annual inspection label and electronic equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |