Nothing Special   »   [go: up one dir, main page]

CN113538342B - Convolutional neural network-based aluminum aerosol can coating quality detection method - Google Patents

Convolutional neural network-based aluminum aerosol can coating quality detection method Download PDF

Info

Publication number
CN113538342B
CN113538342B CN202110711510.XA CN202110711510A CN113538342B CN 113538342 B CN113538342 B CN 113538342B CN 202110711510 A CN202110711510 A CN 202110711510A CN 113538342 B CN113538342 B CN 113538342B
Authority
CN
China
Prior art keywords
coating
fine tuning
aluminum aerosol
product
data
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
Application number
CN202110711510.XA
Other languages
Chinese (zh)
Other versions
CN113538342A (en
Inventor
张兴伟
陈满
张科
郭嘉楠
彭宇瑞
杨海林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shantou Oriental Technology Co ltd
Shantou University
Original Assignee
Shantou Oriental Technology Co ltd
Shantou University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shantou Oriental Technology Co ltd, Shantou University filed Critical Shantou Oriental Technology Co ltd
Priority to CN202110711510.XA priority Critical patent/CN113538342B/en
Publication of CN113538342A publication Critical patent/CN113538342A/en
Application granted granted Critical
Publication of CN113538342B publication Critical patent/CN113538342B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/241Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20081Training; Learning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20084Artificial neural networks [ANN]
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Quality & Reliability (AREA)
  • Probability & Statistics with Applications (AREA)
  • Image Analysis (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The invention discloses a convolutional neural network-based aluminum aerosol can coating quality detection method, which comprises the following steps: obtaining a photographing sample of an aluminum aerosol can product on a production line; classifying the photographed samples to finish marking the original image data; performing data enhancement and constructing a sample data set; establishing a grid model: the method comprises the steps of training a Fine tuning-GoogLeNet and a Fine tuning-ResNet18, inputting new image data, carrying out image classification prediction by using a trained defect discrimination model, and taking an image data label with highest prediction probability as a judgment basis. By adopting the invention, the quality detection accuracy of the inner coating and the outer coating of the aluminum aerosol can coating respectively reaches 99.38 percent and 96.4 percent, the efficiency is high, and compared with the manual visual detection, the speed is faster, the accuracy and the reliability are higher; compared with the traditional machine vision detection mode based on feature extraction, the method has stronger scene adaptation capability and better robustness, and can reduce the enterprise cost and improve the efficiency of product quality detection.

Description

Convolutional neural network-based aluminum aerosol can coating quality detection method
Technical Field
The invention relates to the technical field of product quality detection, in particular to a method for detecting the quality of an aluminum aerosol can coating based on a convolutional neural network.
Background
The aluminum aerosol can is favored by the packaging market by virtue of the characteristics of light weight, strong plasticity and the like, and is widely applied to the fields of industrial products, personal protection, cosmetic products and the like. The coating can be sprayed and printed on the inner surface and the outer surface of the tank body in the production process so as to shield the direct contact between the tank content and external substances and aluminum. Therefore, the spraying and printing processes directly affect the quality, qualification rate and production efficiency of the product. However, the coating quality of the aluminum aerosol can is necessary because the defects of coating bubbles, stains, concave-convex can bodies and the like are easily generated due to the influence of various uncertain factors in the spraying and printing processes, so that the product is not attractive, and the use is influenced (the concave-convex can body can influence the bearing capacity of the product, so that the service life is shortened, and even the personal safety of a user is influenced).
The traditional aluminum aerosol can coating quality detection method mainly depends on visual inspection of human eyes at present and has the defects of low detection efficiency, high labor cost, low accuracy, unreliable detection results and the like. The machine vision technology is used as a branch of rapid development in artificial intelligence, is applied to the field of automatic detection at present, but is mostly applied to detection of steel, aluminum pop cans and plane printed matters based on the traditional image feature recognition mode, and a machine vision detection system based on a convolutional neural network in the aspect of aluminum aerosol can coating quality is not available.
Disclosure of Invention
The technical problem to be solved by the embodiment of the invention is to provide a method for detecting the quality of an aluminum aerosol can coating based on a convolutional neural network. Can realize the simultaneous detection of the defects of bubbles, concave-convex, stains and the like of the inner and outer coatings of the aluminum aerosol can.
In order to solve the technical problems, the embodiment of the invention provides a method for detecting the quality of an aluminum aerosol can coating based on a convolutional neural network, which comprises the following steps:
s1: obtaining a photographing sample of an aluminum aerosol can product on a production line;
s2: dividing the shot sample into two major categories of an inner surface image and an outer surface image, and respectively classifying to finish marking of original image data;
s3: carrying out data enhancement and constructing an aluminum aerosol can inner and outer coating sample data set;
s4: adjusting the size of the image;
s5: based on a transfer learning mode, fine tuning is carried out on GoogLeNet and ResNet18 models to obtain two convolutional neural network models suitable for aluminum can cold spray coating image data sets: fine tuning-google net and Fine tuning-ResNet18;
s6: adjusting initial training parameters of a network, performing model training on the Fine tuning-GoogLeNet by using an inner coating data set of an aluminum aerosol can, and performing model training on the Fine tuning-ResNet18 by using an outer coating data set of the aluminum aerosol can to obtain a trained aluminum aerosol can defect judging model based on the Fine tuning-GoogLeNet and the Fine tuning-ResNet18;
s7: inputting new image data, carrying out image classification prediction by using a trained defect discrimination model, and taking an image data label with highest prediction probability as a judgment basis.
Wherein, the Fine tuning-google net is the image data of the inner coating sprayed by aluminum pot cold, and is adjusted based on GoogLeNet InceptionV1 network model, the adjusting method comprises:
s11: freezing other layers to make the input layer accept the input image size of 227 x 1;
s12: replacing the local response normalization layer in the original network with a batch normalization layer;
s13: a Dropout is arranged in front of the full connection layer;
s14: the number of output nodes of the full-connection layer is 3, and after the full-connection layer, the Softmax function is used for realizing the classified output of the images.
The method for normalizing the layers in batches comprises the following steps:
s121: for minimum batch data inputDefining normalized network responses as
S122: calculating the mean value of each minimum batch of data:
s123: calculating the variance of each minimum batch of data:
s124: and carrying out normalization processing on the minimum batch data by using the obtained mean value and variance to ensure that the mean value is 0 and the variance is 1:
s125: introducing trainable parametersScaling and translating the data: />
The Fine tuning-ResNet18 adjusts in a ResNet18 network model, and comprises the following steps:
1) Freezing other layers to make the input layer accept the input image with 224 x 1;
2) A Dropout is arranged in front of the full connection layer;
3) The FC layer output nodes are set to 5, and then the Softmax function is used to realize image classification output.
Wherein, the step S7 further includes: if the current product is the normal image label, recording the current product as a normal product and recording the current product as a detected product; otherwise, recording the product as the corresponding defect product of the label, and recording the product as the detected product.
The embodiment of the invention has the following beneficial effects: the invention builds an aluminum aerosol can defect judging model to realize the detection of product quality, and in the established data set, the accuracy of the inner and outer coating quality detection respectively reaches 99.38% and 96.4%; the detection speed of the inner coating single pot and the outer coating single pot is higher, and compared with the manual visual detection, the detection speed is higher, the accuracy and the reliability are higher; compared with the traditional machine vision detection mode based on feature extraction, the method has stronger scene adaptation capability and better robustness, and can reduce the enterprise cost and improve the efficiency of product quality detection.
Drawings
FIG. 1 is a schematic diagram of an aluminum aerosol can image dataset construction flow;
FIG. 2 is a schematic diagram of training flow of an aluminum aerosol can inner and outer coating defect discrimination model;
FIG. 3 is a flow chart of a method for detecting the quality of an inner coating and an outer coating of an aluminum aerosol can;
FIG. 4 is a schematic diagram of the network architecture of the Fine tuning-GoogleNet;
FIG. 5 is a schematic diagram of the structural design of the Fine tuning-ResNet18;
FIG. 6 is a graph showing the results of quality detection of an inner coating layer of an aluminum aerosol can with an inner coating bubble defect;
FIG. 7 is a graph showing the quality test results of the outer coating of an aluminum aerosol can with defects of the necking recess of the outer coating.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings, for the purpose of making the objects, technical solutions and advantages of the present invention more apparent.
The method for detecting the quality of the aluminum aerosol can coating based on the convolutional neural network, disclosed by the embodiment of the invention, is carried out by the following implementation steps.
1. The aluminum aerosol can products on the production line, including normal products and products with defects, are collected as shown in fig. 1.
2. And taking photos and sampling the inner surface and the outer surface of the acquired product by using a black-and-white camera. The obtained samples are divided into two major categories of inner surface images and outer surface images, and each category is divided into a plurality of minor categories according to normal products and products with different defects, so that the marking of the original image data is completed.
3. And carrying out data enhancement on the original image and constructing an aluminum aerosol can inner and outer coating sample data set. And the original data is expanded by using a spatial scale transformation (horizontal overturn, vertical overturn and horizontal vertical overturn) mode and a noise adding (Gaussian blur processing) mode, so that the fault tolerance and the robustness of a subsequent convolutional neural network model are enhanced. Based on the enhanced image data, an aluminum aerosol can inner coating sample data set and an aluminum aerosol can outer coating sample data set are constructed, the normal image and each type of coating image with defects are divided into a plurality of types by the data sets, and each type of image is marked with a data label.
The Gaussian blur processing mode is based on a two-dimensional Gaussian distribution principle:
for the original imageWherein->A Gaussian kernel constructed by using a two-dimensional Gaussian function and representing coordinate points of the original image, wherein +.>The sliding window convolution is performed thereon, and the process can be expressed asObtained->I.e. the image after gaussian blur processing.
4. The image is resized. The inner coating data image pixel size is adjusted to [227,227], and the outer coating data image pixel size is adjusted to [224,224].
5. And constructing a classified convolutional neural network. Based on a transfer learning mode, fine tuning is carried out on GoogLeNet and ResNet18 models to obtain two convolutional neural network models suitable for aluminum can cold spray coating image data sets: the Fine tuning-GoogLeNet and Fine tuning-ResNet18 are shown in FIGS. 2, 4, and 5.
The network structure of the Fine tuning-GoogLeNet aims at image data of an aluminum can cold spraying inner coating, and is adjusted on the basis of a GoogLeNet InceptionV network model, wherein the adjustment method comprises the following steps:
1) Freezing other layers to make the input layer accept the input image size of 227 x 1;
2) Replacing the local response normalization layer (Local Response Normalization, LRN) in the original network with a bulk normalization layer (Batch Normalization, BN);
the algorithm flow of the BN layer is as follows:
for input minimum batch data (mini-batch)Defining normalized network response as +.>
a) First, the mean value of each mini-batch is calculated:
b) The variance of each mini-batch is calculated:
c) And (3) carrying out normalization processing on the mini-batch by using the obtained mean value and variance to enable the mean value to be 0 and the variance to be 1:
d)
e) Introducing trainable parametersScaling and translating the data: />
3) Setting Dropout (temporary hidden neuron) in front of the full connection layer;
4) The number of output nodes of the full connection layer (Fully connected layer, FC) is 3, and after the full connection layer, the image classification output is realized by using a Softmax function.
The network can train and classify three types of images (normal, bubble, interior surface bulge) of the inner coating.
The Fine tuning-ResNet18 network structure is adjusted based on a ResNet18 network model aiming at an aluminum can cold spraying outer coating data set, and the method comprises the following steps:
1) Freezing other layers to make the input layer accept input image with size of
2) A Dropout is arranged in front of the full connection layer;
3) The FC layer output nodes are set to 5, and then the Softmax function is used to realize image classification output.
The network can train and classify five types of images (normal, bubble, tank dent, closing dent, stain) of the outer coating.
6. And (5) training a network model. And adjusting network initial training parameters, and respectively using the Fine tuning-GoogLeNet and the Fine tuning-ResNet18 to carry out model training on the inner coating and the outer coating of the aluminum aerosol can to obtain a trained aluminum aerosol can defect judging model based on the Fine tuning-GoogLeNet and the Fine tuning-ResNet 18.
7. And detecting and judging based on the trained defect judging model.
8. Inputting new image data, carrying out image classification prediction by using a trained defect discrimination model, and taking an image data label with highest prediction probability as a judgment basis. If the current product is the normal image label, recording the current product as a normal product and recording the current product as a detected product; otherwise, the defective product corresponding to the label is recorded, and the detected product is recorded at the same time, as shown in fig. 3.
The result of detecting the bubble defect with the inner coating on the inner coating quality of the aluminum aerosol can by applying the method of the invention is shown in figure 6, and the result of detecting the notch defect with the closing-in of the outer coating on the outer coating quality of the aluminum aerosol can is shown in figure 7.
The implementation of the invention has the following advantages:
1. the automatic detection method aiming at the characteristics of the spraying and printing process of the aluminum aerosol can is an effective supplement to the quality detection of the product of the existing high-speed aluminum aerosol can production line;
2. the method is different from the method for detecting the quality of the product based on the feature extraction of the visual inspection of human eyes and the traditional machine vision, uses a convolutional neural network and a transfer learning mode, and utilizes a Fine tuning-GoogLeNet and a Fine tuning-ResNet18 network to construct an aluminum aerosol can defect judging model so as to realize the detection of the quality of the product. In the established data set, the quality detection accuracy of the inner coating and the outer coating respectively reaches 99.38 percent and 96.4 percent; the detection speed of the inner coating and the outer coating is higher. Compared with manual visual inspection, the method has the advantages of higher speed, higher accuracy and higher reliability; compared with the traditional machine vision detection mode based on feature extraction, the method has stronger scene adaptation capability and better robustness. The enterprise cost can be reduced, and the efficiency of product quality detection can be improved.
The above disclosure is only a preferred embodiment of the present invention, and it is needless to say that the scope of the invention is not limited thereto, and therefore, the equivalent changes according to the claims of the present invention still fall within the scope of the present invention.

Claims (5)

1. The method for detecting the quality of the aluminum aerosol can coating based on the convolutional neural network is characterized by comprising the following steps of:
s1: obtaining a photographing sample of an aluminum aerosol can product on a production line;
s2: dividing the shot sample into two major categories of an inner surface image and an outer surface image, and respectively classifying to finish marking of original image data;
s3: carrying out data enhancement and constructing an aluminum aerosol can inner and outer coating sample data set;
s4: adjusting the size of the image;
s5: based on a transfer learning mode, fine tuning is carried out on GoogLeNet and ResNet18 models to obtain two convolutional neural network models suitable for aluminum can cold spray coating image data sets: fine tuning-google net and Fine tuning-ResNet18;
s6: adjusting initial training parameters of a network, performing model training on the Fine tuning-GoogLeNet by using an inner coating data set of an aluminum aerosol can, and performing model training on the Fine tuning-ResNet18 by using an outer coating data set of the aluminum aerosol can to obtain a trained aluminum aerosol can defect judging model based on the Fine tuning-GoogLeNet and the Fine tuning-ResNet18;
s7: inputting new image data, carrying out image classification prediction by using a trained defect discrimination model, and taking an image data label with highest prediction probability as a judgment basis.
2. The convolutional neural network-based aluminum aerosol can coating quality detection method of claim 1, wherein the Fine tuning-google net is based on GoogLeNet InceptionV network model for using aluminum can cold spray inner coating image data, the adjustment method comprises:
s11: freezing other layers to make the input layer accept the input image size of 227 x 1;
s12: replacing the local response normalization layer in the original network with a batch normalization layer;
s13: a Dropout is arranged in front of the full connection layer;
s14: the number of output nodes of the full-connection layer is 3, and after the full-connection layer, the Softmax function is used for realizing the classified output of the images.
3. The convolutional neural network-based aluminum aerosol can coating quality detection method of claim 2, wherein the batch normalization layer method comprises:
s121: for minimum batch data inputDefining normalized network responses as
S122: calculating the mean value of each minimum batch of data:
s123: calculating the variance of each minimum batch of data:
s124: and carrying out normalization processing on the minimum batch data by using the obtained mean value and variance to ensure that the mean value is 0 and the variance is 1:
s125: introducing trainable parametersScaling and translating the data: />
4. The convolutional neural network-based aluminum aerosol can coating quality detection method of claim 1, wherein the Fine tuning-ResNet18 is tuned in a ResNet18 network model, comprising the steps of:
1) Freezing other layers to make the input layer accept the input image with 224 x 1;
2) A Dropout is arranged in front of the full connection layer;
3) The FC layer output nodes are set to 5, and then the Softmax function is used to realize image classification output.
5. The method for detecting the coating quality of an aluminum aerosol can based on a convolutional neural network according to any one of claims 1 to 4, wherein the step S7 further comprises: if the current product is the normal image label, recording the current product as a normal product and recording the current product as a detected product; otherwise, recording the product as the corresponding defect product of the label, and recording the product as the detected product.
CN202110711510.XA 2021-06-25 2021-06-25 Convolutional neural network-based aluminum aerosol can coating quality detection method Active CN113538342B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110711510.XA CN113538342B (en) 2021-06-25 2021-06-25 Convolutional neural network-based aluminum aerosol can coating quality detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110711510.XA CN113538342B (en) 2021-06-25 2021-06-25 Convolutional neural network-based aluminum aerosol can coating quality detection method

Publications (2)

Publication Number Publication Date
CN113538342A CN113538342A (en) 2021-10-22
CN113538342B true CN113538342B (en) 2023-08-18

Family

ID=78125857

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110711510.XA Active CN113538342B (en) 2021-06-25 2021-06-25 Convolutional neural network-based aluminum aerosol can coating quality detection method

Country Status (1)

Country Link
CN (1) CN113538342B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115420756B (en) * 2022-11-07 2023-03-21 深圳市志橙半导体材料股份有限公司 Method and device for detecting surface defects of silicon carbide coating graphite plate
CN117288099B (en) * 2023-11-27 2024-01-30 江苏海德曼新材料股份有限公司 Machine vision-based aluminum-plastic composite board production quality detection and analysis method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102331424A (en) * 2010-07-12 2012-01-25 南通全技纺织涂层有限公司 Coating quality on-line detection method based on image recognition technology
CN110349126A (en) * 2019-06-20 2019-10-18 武汉科技大学 A kind of Surface Defects in Steel Plate detection method based on convolutional neural networks tape label
CN111709451A (en) * 2020-05-21 2020-09-25 五邑大学 Wine bottle surface defect detection method, electronic device and storage medium
CN112819748A (en) * 2020-12-16 2021-05-18 机科发展科技股份有限公司 Training method and device for strip steel surface defect recognition model

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3270095A1 (en) * 2016-07-13 2018-01-17 Sightline Innovation Inc. System and method for surface inspection
US10607119B2 (en) * 2017-09-06 2020-03-31 Kla-Tencor Corp. Unified neural network for defect detection and classification

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102331424A (en) * 2010-07-12 2012-01-25 南通全技纺织涂层有限公司 Coating quality on-line detection method based on image recognition technology
CN110349126A (en) * 2019-06-20 2019-10-18 武汉科技大学 A kind of Surface Defects in Steel Plate detection method based on convolutional neural networks tape label
CN111709451A (en) * 2020-05-21 2020-09-25 五邑大学 Wine bottle surface defect detection method, electronic device and storage medium
CN112819748A (en) * 2020-12-16 2021-05-18 机科发展科技股份有限公司 Training method and device for strip steel surface defect recognition model

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
大型储罐防腐蚀涂层厚度智能检测技术分析;张永吉;《全面腐蚀控制》;第34卷(第02期);48-49 *

Also Published As

Publication number Publication date
CN113538342A (en) 2021-10-22

Similar Documents

Publication Publication Date Title
CN111598861B (en) Improved Faster R-CNN model-based non-uniform texture small defect detection method
CN111553929B (en) Mobile phone screen defect segmentation method, device and equipment based on converged network
CN110210463B (en) Precise ROI-fast R-CNN-based radar target image detection method
CN109584248B (en) Infrared target instance segmentation method based on feature fusion and dense connection network
CN106650721B (en) A kind of industrial character identifying method based on convolutional neural networks
CN111223088B (en) Casting surface defect identification method based on deep convolutional neural network
CN106875373B (en) Mobile phone screen MURA defect detection method based on convolutional neural network pruning algorithm
CN111160249A (en) Multi-class target detection method of optical remote sensing image based on cross-scale feature fusion
CN113538342B (en) Convolutional neural network-based aluminum aerosol can coating quality detection method
CN113838034A (en) Candy packaging surface defect rapid detection method based on machine vision
CN113888461A (en) Method, system and equipment for detecting defects of hardware parts based on deep learning
CN113420776B (en) Multi-side joint detection article classification method based on model fusion
CN115775236A (en) Surface tiny defect visual detection method and system based on multi-scale feature fusion
CN112446871A (en) Tunnel crack identification method based on deep learning and OpenCV
CN116612106A (en) Method for detecting surface defects of optical element based on YOLOX algorithm
CN115829995A (en) Cloth flaw detection method and system based on pixel-level multi-scale feature fusion
CN113421223B (en) Industrial product surface defect detection method based on deep learning and Gaussian mixture
CN112508857A (en) Aluminum product surface defect detection method based on improved Cascade R-CNN
CN114972316A (en) Battery case end surface defect real-time detection method based on improved YOLOv5
CN109919150A (en) A kind of non-division recognition sequence method and system of 3D pressed characters
CN113989196B (en) Visual-sense-based method for detecting appearance defects of earphone silica gel gasket
CN110992301A (en) Gas contour identification method
CN114066810A (en) Method and device for detecting concave-convex point defects of packaging box
CN116523916B (en) Product surface defect detection method and device, electronic equipment and storage medium
CN112270404A (en) Detection structure and method for bulge defect of fastener product based on ResNet64 network

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