CN111523540A - Metal surface defect detection method based on deep learning - Google Patents
Metal surface defect detection method based on deep learning Download PDFInfo
- Publication number
- CN111523540A CN111523540A CN202010306489.0A CN202010306489A CN111523540A CN 111523540 A CN111523540 A CN 111523540A CN 202010306489 A CN202010306489 A CN 202010306489A CN 111523540 A CN111523540 A CN 111523540A
- Authority
- CN
- China
- Prior art keywords
- deep learning
- metal surface
- feature
- data set
- inputting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 57
- 238000001514 detection method Methods 0.000 title claims abstract description 40
- 229910052751 metal Inorganic materials 0.000 title claims abstract description 37
- 239000002184 metal Substances 0.000 title claims abstract description 37
- 238000013135 deep learning Methods 0.000 title claims abstract description 33
- 238000000605 extraction Methods 0.000 claims abstract description 11
- 238000002372 labelling Methods 0.000 claims abstract description 3
- 238000000034 method Methods 0.000 claims description 16
- 238000011176 pooling Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 11
- 230000007613 environmental effect Effects 0.000 abstract description 5
- 230000008030 elimination Effects 0.000 abstract description 3
- 238000003379 elimination reaction Methods 0.000 abstract description 3
- 238000013136 deep learning model Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013434 data augmentation Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000005242 forging Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention provides a metal surface defect detection method based on deep learning, which comprises the following steps: step 1, determining the models of a camera and a light source according to the sizes of metal defect parts, building an image taking environment of the light source and the camera, and acquiring images; step 2, carrying out data set labeling and data set enhancement on the acquired image, inputting the image into a mixed feature pyramid network structure, and carrying out feature extraction; and 3, inputting the obtained feature map of each layer into the RPN to select a candidate region. The metal surface defect detection method based on deep learning provided by the invention realizes automatic detection and defect identification and positioning of the input image of the production line, automatic feature extraction, elimination of negative influence caused by manual feature extraction, overcoming the defect of larger influence of environmental factors in machine vision, and the mixed feature pyramid network structure has better performance for smaller targets and multi-scale problems.
Description
Technical Field
The invention relates to the technical field of metal surface defect detection, in particular to a metal surface defect detection method based on deep learning.
Background
At present, the surface quality detection of products has gained wide attention, so the detection technology has rapidly developed, and especially the detection algorithm based on deep learning has achieved many excellent results. The detection algorithm based on deep learning is higher than manual detection efficiency in the aspect of product surface quality control, the workload of inspectors can be reduced, the production cost of enterprises is reduced, the defect detection precision and positioning are more accurate, the production efficiency is improved, the automation of manufacturing industry is promoted, the intelligent development is realized, and the construction of intelligent factories is promoted. In addition, in the aspect of scientific research, deep learning belongs to a new field, theoretical results are combined with practice, so that the research results fall on the ground, better improvement is sought in practical application, deep development of related algorithms can be better promoted, research on the combination of the theoretical results of the deep learning and the practice application is developed, and the method has important research significance for the development of the deep learning.
Disclosure of Invention
The invention provides a metal surface defect detection method based on deep learning, and aims to solve the problem that machine vision is greatly influenced by environmental factors.
In order to achieve the above object, an embodiment of the present invention provides a method for detecting defects on a metal surface based on deep learning, including:
step 1, determining the models of a camera and a light source according to the sizes of metal defect parts, building an image taking environment of the light source and the camera, and acquiring images;
step 2, carrying out data set labeling and data set enhancement on the acquired image, inputting the image into a mixed feature pyramid network structure, and carrying out feature extraction;
step 3, inputting the obtained feature map of each layer into an RPN network, and selecting a candidate region;
step 4, combining the obtained feature map of each layer and the candidate region to enter ROIAlign, mapping the interesting region onto the feature map, dividing the interesting region into fixed sizes, and performing pooling operation;
and 5, inputting the output obtained by the ROI Align pooling into the convolutional layer to obtain the output of the convolutional layer, and inputting the output of the convolutional layer into the full-connection layer for classification and regression.
Wherein, the step 2 specifically comprises:
and the data set is marked by adopting a Labelimg tool for on-line processing, and the data set is enhanced by adopting rotation, defect image counterfeiting and scaling and random darkness.
Wherein, the step 2 further comprises:
and fusing the obtained feature graphs on the basis of the mixed feature pyramid network.
Wherein, the step 3 specifically comprises:
16x16 256-dimensional feature vectors are obtained by convolution with 3x3, an 18x16x16 feature map and a 36x16x16 feature map are obtained by convolution with 1x1 twice, and finally 16x16x9 results are obtained, wherein each result has four coordinates and two scores, and then the candidate frames are obtained after processing through the set size of Anchors.
Wherein, the step 4 specifically comprises:
the feature representation is integrated and output as a value.
Wherein, the step 5 specifically comprises:
and (5) forming a complete graph by the local features obtained by the convolutional layers through a weight matrix, and then classifying.
The scheme of the invention has the following beneficial effects:
the metal surface defect detection method based on deep learning in the embodiment of the invention realizes automatic detection and defect identification and positioning of the input image of the production line, automatic feature extraction, elimination of negative influence caused by manual feature extraction, overcoming the defect of large influence of environmental factors in machine vision, and the mixed feature pyramid network structure has better performance for smaller targets and multi-scale problems.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram of a deep learning model according to the present invention;
FIG. 3 is an architectural diagram of a hybrid signature pyramid network of the present invention;
FIG. 4 is a graph (1) of the average accuracy of the present invention;
FIG. 5 is a graph (2) showing the average accuracy of the expression effect of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following detailed description is made with reference to the accompanying drawings and specific embodiments.
The invention provides a metal surface defect detection method based on deep learning, aiming at the problem that the existing machine vision is greatly influenced by environmental factors.
As shown in fig. 1 to 5, an embodiment of the present invention provides a method for detecting defects on a metal surface based on deep learning, including: step 1, determining the models of a camera and a light source according to the size of a metal defect part, constructing an image taking environment of the light source and the camera, and acquiring an image; step 2, carrying out data set marking and data set enhancement on the acquired image, inputting the image into a mixed feature pyramid network structure, and carrying out feature extraction; step 3, inputting the obtained characteristic graph of each layer into an RPN network, and selecting candidate regions; step 4, combining the obtained feature map of each layer and the candidate region to enter ROIAlign, mapping the region of interest to the feature map, dividing the region of interest into fixed sizes, and performing pooling operation; and 5, inputting the output obtained by the ROIAlign pooling into the convolutional layer to obtain the output of the convolutional layer, and inputting the output of the convolutional layer into the full-connection layer for classification and regression.
The metal surface defect detection method based on deep learning provided by the embodiment of the invention takes a Foshan aluminum profile production line as a research, and finds that the existing production line mainly adopts manual detection, but the manual detection has low efficiency, the evaluation standard is difficult to accurately quantify, and the comprehensive automatic production is not realized, a large number of defect pictures are collected on site, and after data set analysis is carried out, the metal surface defect detection method based on deep learning selects a deep learning model as a starting point. In the detection of defects on a metal surface, compared with a smooth surface (such as an optical component), a photo of the metal surface may be prone to problems such as uneven illumination, strong reflection, background noise, stains and the like, and thus detection using a machine vision method is difficult. In addition, due to the limitation of production environment, the surface of the product is easy to generate pollutants, which is inevitable, and the influence of foreign matters such as dust and fiber is also included. In conclusion, for the detection of the metal surface defects, the deep learning model has better advantages in the metal surface defect detection.
The metal surface defect detection method based on deep learning in the embodiment of the invention is roughly divided into two parts, namely, a first stage, data set acquisition and picture preprocessing; and in the second stage, inputting the pictures into the mixed characteristic pyramid network for repeated training and optimization to obtain a network model with higher performance.
Wherein, the step 2 specifically comprises: and the data set is marked by adopting a Labelimg tool for on-line processing, and the data set is enhanced by adopting rotation, counterfeit defect images, zoom and random darkness.
Wherein, the step 2 further comprises: and fusing the obtained feature graphs on the basis of the mixed feature pyramid network.
In the method for detecting metal surface defects based on deep learning according to the embodiment of the invention, on the basis of the feature pyramid network, the feature graph obtained by fusion is cut to the same size in an interpolation or maximum pooling manner, then the feature graphs of the same size are added to obtain an average feature graph, and the average feature graph is cut to the size of the previous feature graph by up-down sampling or convolution.
Wherein, the step 3 specifically comprises: 16x16 256-dimensional feature vectors are obtained by convolution with 3x3, an 18x16x16 feature map and a 36x16x16 feature map are obtained by convolution with 1x1 twice, and finally 16x16x9 results are obtained, wherein each result has four coordinates and two scores, and then the candidate frames are obtained after processing through the set size of Anchors.
Wherein, the step 4 specifically comprises: the feature representation is integrated and output as a value.
Wherein, the step 5 specifically comprises: and (5) forming a complete graph by the local features obtained by the convolutional layers through a weight matrix, and then classifying.
The metal surface defect detection method based on deep learning described in the above embodiments of the present invention needs to train a deep learning model, continuously optimizes the deep learning model in the training process, for example, soft-NMS is used to replace NMS, analyzes the distribution of real frames in a training set by k-means clustering algorithm to obtain the anchors size of a data set, and replaces the final convolution layer with feasible convolution, so as to adapt to geometric deformation well, extract more accurate positions of features, and train and adjust parameter settings repeatedly until the requirements of robustness and accuracy of a production line are met.
The metal surface defect detection method based on deep learning of the embodiment of the invention comprises the following specific implementation steps: and acquiring pictures, wherein the total number of all types of defect pictures is 300, and the total number of the defect pictures is 3000, and in order to improve the correctness of a detection result, the number of each type of defect pictures reaches 3000 by adopting a data augmentation mode. And training the feature extraction part by combining the defect picture and the normal picture. The backbone networks used were pre-trained on ImageNet, with a learning rate of 0.001 after addition of the modified mixed feature pyramid network structure, and a batch size of 16, and for evaluation the detector was run on a single Titan X GPU with CUDA9 and CUDNN7, with a batch size of 1. In the training process, the deep learning model is optimized: the K-means clustering method replaces the fixed anchor size, approximately 1 percentage point is increased, the size is changed, and the proportion is unchanged. The ROI Align is used for replacing ROI Pooling, the ROI Align is a regional characteristic gathering mode, the problem of region mismatching caused by twice quantization in ROI Pooling operation is well solved, and experiments show that the ROI Pooling is replaced by the ROI Align in a detection task, so that the accuracy of a detection model can be improved. The reason that the last block of the original result-101 structure is changed into the deformable convolution is that in the implementation of the deformable convolution, a deviation needs to be learned based on the previous features, the previous features are strong enough to ensure that the deviation cannot be learnt disorderly, so that the last block is changed, the general flow of the metal surface defect detection method based on the deep learning is shown in fig. 1, the appearance effect of the metal surface defect detection method based on the deep learning is shown in fig. 4, the defects of scratch, missing bottom, orange peel and dirty points in fig. 4 are obviously improved, and the average accuracy reaches 86.31%.
Continuously optimizing the metal surface defect detection method based on deep learning through improvement, so that the overall average accuracy of the metal surface defect detection method based on deep learning is continuously improved, introducing a method for forging a defect image into a data set reinforcing part, analyzing the threshold distribution of the defect image, and eliminating the dependence of the method on the characteristic surrounding environment by adding small squares or random-shaped graphs; in the emotional part of the data set, a mode of mixing two images is tried, the final detection result is not friendly in consideration of excessive interference, so that the mode is abandoned, and meanwhile, in the training process, the soft-NMS is used for replacing the NMS. Fig. 5 shows the final performance of the metal surface defect detection method based on deep learning on a data set, the overall performance is good, the average accuracy of three defects, namely the mottled defect, the angular position bottom leakage defect and the pit formation in fig. 5 can reach 100%, the overall average accuracy is improved to 88.91%, and is improved by about 2%.
The metal surface defect detection method based on deep learning in the embodiment of the invention realizes automatic detection and defect identification and positioning of the input image of the production line, automatic feature extraction, elimination of negative influence caused by manual feature extraction, overcoming the defect of large influence of environmental factors in machine vision, and the mixed feature pyramid network structure has better performance for smaller targets and multi-scale problems.
The foregoing is a preferred embodiment of the present invention, and it should be noted that it is obvious to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and these modifications and improvements should be construed as the protection scope of the present invention.
Claims (6)
1. A metal surface defect detection method based on deep learning is characterized by comprising the following steps:
step 1, determining the models of a camera and a light source according to the sizes of metal defect parts, building an image taking environment of the light source and the camera, and acquiring images;
step 2, carrying out data set labeling and data set enhancement on the acquired image, inputting the image into a mixed feature pyramid network structure, and carrying out feature extraction;
step 3, inputting the obtained characteristic graph of each layer into an RPN network, and selecting candidate regions;
step 4, combining the obtained feature map of each layer and the candidate region to enter ROIAlign, mapping the region of interest to the feature map, dividing the region of interest into fixed sizes, and performing pooling operation;
and 5, inputting the output obtained by the ROIAlign pooling into the convolutional layer to obtain the output of the convolutional layer, and inputting the output of the convolutional layer into the full-connection layer for classification and regression.
2. The method for detecting defects on a metal surface based on deep learning of claim 1, wherein the step 2 specifically comprises:
and the data set is marked by adopting a Labelimg tool for on-line processing, and the data set is enhanced by adopting rotation, counterfeit defect images, zoom and random darkness.
3. The method for detecting the defects on the metal surface based on the deep learning as claimed in claim 2, wherein the step 2 further comprises:
and fusing the obtained feature graphs on the basis of the mixed feature pyramid network.
4. The method for detecting defects on a metal surface based on deep learning of claim 3, wherein the step 3 specifically comprises:
obtaining 16x16 256-dimensional feature vectors by using convolution of 3x3, obtaining an 18x16x16 feature map and a 36x16x16 feature map by using convolution of 1x1 twice, finally obtaining 16x16x9 results, wherein each result has four coordinates and two scores, and obtaining a candidate frame after processing through the set size of Anchors.
5. The method for detecting defects on a metal surface based on deep learning of claim 4, wherein the step 4 specifically comprises:
the feature representation is integrated and output as a value.
6. The method for detecting defects on a metal surface based on deep learning of claim 5, wherein the step 5 specifically comprises:
and (5) forming a complete graph by the local features obtained by the convolutional layers through a weight matrix, and then classifying.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010306489.0A CN111523540A (en) | 2020-04-17 | 2020-04-17 | Metal surface defect detection method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010306489.0A CN111523540A (en) | 2020-04-17 | 2020-04-17 | Metal surface defect detection method based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111523540A true CN111523540A (en) | 2020-08-11 |
Family
ID=71904421
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010306489.0A Pending CN111523540A (en) | 2020-04-17 | 2020-04-17 | Metal surface defect detection method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111523540A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184619A (en) * | 2020-08-25 | 2021-01-05 | 华北电力大学(保定) | Metal part surface defect detection method based on deep learning |
CN112508090A (en) * | 2020-12-04 | 2021-03-16 | 重庆大学 | External package defect detection method |
CN113012153A (en) * | 2021-04-30 | 2021-06-22 | 武汉纺织大学 | Aluminum profile flaw detection method |
CN113239786A (en) * | 2021-05-11 | 2021-08-10 | 重庆市地理信息和遥感应用中心 | Remote sensing image country villa identification method based on reinforcement learning and feature transformation |
CN113658108A (en) * | 2021-07-22 | 2021-11-16 | 西南财经大学 | Glass defect detection method based on deep learning |
WO2022160170A1 (en) * | 2021-01-28 | 2022-08-04 | 东莞职业技术学院 | Method and apparatus for detecting metal surface defects |
CN114863094A (en) * | 2022-05-31 | 2022-08-05 | 征图新视(江苏)科技股份有限公司 | Industrial image region-of-interest segmentation algorithm based on double-branch network |
WO2023168984A1 (en) * | 2022-03-09 | 2023-09-14 | 三门三友科技股份有限公司 | Area-array camera-based quality inspection method and system for cathode copper |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110443791A (en) * | 2019-08-02 | 2019-11-12 | 西安工程大学 | A kind of workpiece inspection method and its detection device based on deep learning network |
CN110782183A (en) * | 2019-11-07 | 2020-02-11 | 山东浪潮人工智能研究院有限公司 | Aluminum product defect detection method based on Faster R-CNN |
CN110853015A (en) * | 2019-11-12 | 2020-02-28 | 中国计量大学 | Aluminum profile defect detection method based on improved Faster-RCNN |
CN110929685A (en) * | 2019-12-10 | 2020-03-27 | 电子科技大学 | Pedestrian detection network structure based on mixed feature pyramid and mixed expansion convolution |
-
2020
- 2020-04-17 CN CN202010306489.0A patent/CN111523540A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110443791A (en) * | 2019-08-02 | 2019-11-12 | 西安工程大学 | A kind of workpiece inspection method and its detection device based on deep learning network |
CN110782183A (en) * | 2019-11-07 | 2020-02-11 | 山东浪潮人工智能研究院有限公司 | Aluminum product defect detection method based on Faster R-CNN |
CN110853015A (en) * | 2019-11-12 | 2020-02-28 | 中国计量大学 | Aluminum profile defect detection method based on improved Faster-RCNN |
CN110929685A (en) * | 2019-12-10 | 2020-03-27 | 电子科技大学 | Pedestrian detection network structure based on mixed feature pyramid and mixed expansion convolution |
Non-Patent Citations (1)
Title |
---|
双锴: "《计算机视觉》", 31 January 2020, 北京邮电大学出版社 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184619A (en) * | 2020-08-25 | 2021-01-05 | 华北电力大学(保定) | Metal part surface defect detection method based on deep learning |
CN112508090A (en) * | 2020-12-04 | 2021-03-16 | 重庆大学 | External package defect detection method |
WO2022160170A1 (en) * | 2021-01-28 | 2022-08-04 | 东莞职业技术学院 | Method and apparatus for detecting metal surface defects |
CN113012153A (en) * | 2021-04-30 | 2021-06-22 | 武汉纺织大学 | Aluminum profile flaw detection method |
CN113239786A (en) * | 2021-05-11 | 2021-08-10 | 重庆市地理信息和遥感应用中心 | Remote sensing image country villa identification method based on reinforcement learning and feature transformation |
CN113658108A (en) * | 2021-07-22 | 2021-11-16 | 西南财经大学 | Glass defect detection method based on deep learning |
WO2023168984A1 (en) * | 2022-03-09 | 2023-09-14 | 三门三友科技股份有限公司 | Area-array camera-based quality inspection method and system for cathode copper |
CN114863094A (en) * | 2022-05-31 | 2022-08-05 | 征图新视(江苏)科技股份有限公司 | Industrial image region-of-interest segmentation algorithm based on double-branch network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111523540A (en) | Metal surface defect detection method based on deep learning | |
CN106875381B (en) | Mobile phone shell defect detection method based on deep learning | |
WO2022236876A1 (en) | Cellophane defect recognition method, system and apparatus, and storage medium | |
Mathavan et al. | Use of a self-organizing map for crack detection in highly textured pavement images | |
CN108765402B (en) | Non-woven fabric defect detection and classification method | |
CN112613097A (en) | BIM rapid modeling method based on computer vision | |
CN113393426B (en) | Steel rolling plate surface defect detection method | |
CN110473201A (en) | A kind of automatic testing method and device of disc surface defect | |
CN113554631B (en) | Chip surface defect detection method based on improved network | |
CN112907519A (en) | Metal curved surface defect analysis system and method based on deep learning | |
CN113177924A (en) | Industrial production line product flaw detection method | |
US20210035305A1 (en) | Image-based classification system | |
CN104034637B (en) | Based on the online quality inspection device of diamond wire particle of machine vision | |
CN111696079B (en) | Surface defect detection method based on multitask learning | |
CN112489042B (en) | Method for detecting metal printing defects and surface damage based on super-resolution reconstruction | |
CN115439458A (en) | Industrial image defect target detection algorithm based on depth map attention | |
CN113962929A (en) | Photovoltaic cell assembly defect detection method and system and photovoltaic cell assembly production line | |
CN109544513A (en) | A kind of steel pipe end surface defect extraction knowledge method for distinguishing | |
CN113313107A (en) | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge | |
CN113627435A (en) | Method and system for detecting and identifying flaws of ceramic tiles | |
CN111753572A (en) | Complex background low-quality two-dimensional bar code detection method based on deep learning | |
CN111667465A (en) | Metal hand basin defect detection method based on far infrared image | |
CN114387262A (en) | Nut positioning detection method, device and system based on machine vision | |
CN111161228B (en) | Button surface defect detection method based on transfer learning | |
CN116797602A (en) | Surface defect identification method and device for industrial product detection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200811 |
|
RJ01 | Rejection of invention patent application after publication |