CN111311544A - Floor defect detection method based on deep learning - Google Patents
Floor defect detection method based on deep learning Download PDFInfo
- Publication number
- CN111311544A CN111311544A CN202010058893.0A CN202010058893A CN111311544A CN 111311544 A CN111311544 A CN 111311544A CN 202010058893 A CN202010058893 A CN 202010058893A CN 111311544 A CN111311544 A CN 111311544A
- Authority
- CN
- China
- Prior art keywords
- image
- floor
- deep learning
- defect
- semantic segmentation
- 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.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 94
- 238000013135 deep learning Methods 0.000 title claims abstract description 60
- 238000001514 detection method Methods 0.000 title claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 45
- 230000011218 segmentation Effects 0.000 claims abstract description 44
- 238000013145 classification model Methods 0.000 claims abstract description 43
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000002372 labelling Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 17
- 238000012795 verification Methods 0.000 claims description 11
- 230000002950 deficient Effects 0.000 claims description 9
- 238000005520 cutting process Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 230000008014 freezing Effects 0.000 claims description 3
- 238000007710 freezing Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 206010063385 Intellectualisation Diseases 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000009408 flooring Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000011160 research 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
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
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)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a floor defect detection method based on deep learning, which comprises the following steps: 1) collecting a high-resolution image of a floor by using an industrial camera to establish a floor detection image library; 2) manually labeling the defect area of each defect image in the floor detection image library in the step 1) to form a floor detection image label library; 3) establishing a semantic segmentation model and an image classification model based on deep learning; 4) expanding the floor detection image library acquired in the step 1) by adopting a data enhancement technology; 5) carrying out deep learning training on the floor defect semantic segmentation model and the floor defect classification model on the image library expanded in the step 4); 6) and collecting floor surface images on line and carrying out defect detection on the floor surface based on the trained floor defect classification model. The invention achieves the precision of manual screening without manual interference, can reduce the production cost and promote the intellectualization of the floor industry.
Description
The technical field is as follows:
the invention relates to the technical field of floor defect detection, in particular to a floor defect detection method based on deep learning.
Background art:
the defect detection of the floor is a key link of quality control in the production process of the floor. The existing defect detection mode is mainly completed by manual naked eye screening, and the existing method has the following defects: the detection speed is slow, the detection result is greatly influenced by the experience of quality inspection workers, the detection precision is uncontrollable, the cost of the quality inspection workers is high, and the like. With the appearance and development of scientific technologies such as computer technology, artificial intelligence and the like, the object surface defect detection technology based on the machine vision technology comes to the fore, so that the object surface defect detection effect is improved to a great extent, the object surface defect detection rate is increased, and the influence on the accuracy of the defect detection result caused by factors such as scene conditions, subjective judgment and the like is avoided.
Detection techniques based on conventional image processing recognition techniques still require manual extraction of features, and the design of these features requires a great deal of a priori knowledge and experience. In the flooring industry, the surface of the product is mainly irregular and has a large variety, and the traditional image algorithm is not satisfactory in the situation. Deep learning is an emerging research field, and has the advantages of automatic learning of useful characteristics, strong anti-interference capability and high robustness. The deep learning technology is applied to floor defect detection, so that the existing pain point problem can be well solved.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
The invention content is as follows:
the invention aims to provide a floor defect detection method based on deep learning, so that the defects in the prior art are overcome.
In order to achieve the above object, the present invention provides a floor defect detection method based on deep learning, comprising the following steps:
1) collecting a high-resolution image of a floor by using an industrial camera to establish a floor detection image library;
2) manually labeling the defect area of each defect image in the floor detection image library in the step 1) to form a floor detection image label library;
3) establishing a semantic segmentation model and an image classification model based on deep learning;
4) expanding the floor detection image library acquired in the step 1) by adopting a data enhancement technology;
5) carrying out deep learning training on the floor defect semantic segmentation model and the floor defect classification model on the image library expanded in the step 4);
6) and acquiring an image of the floor surface on line, detecting the defect of the floor surface based on the trained floor defect classification model, and positioning the specific position of the floor defect by using the floor defect semantic segmentation model if the defect exists.
The floor detection image library in the step 1) is composed of high-resolution images collected by an industrial camera, the resolution is more than 1024 × 1024, the ratio of the number of normal image samples to the number of defective image samples in the image library is close to 1:1, the total number of samples in the image library is N, and N is more than 1000.
The semantic segmentation model in the step 3) is composed of an Encoder Encoder and a Decoder Decoder, wherein the Encoder Encoder is a basic network and comprises a convolution layer, a batch normalization layer, a pooling layer and a linear activation layer and is used for automatically extracting depth features of an input image and outputting a group of feature maps; the Decoder comprises a convolution layer and an up-sampling layer and is used for screening a characteristic diagram and outputting a result image, wherein the resolution of the result image is the same as that of an input image.
The image classification model in the step 3) and the semantic segmentation model share parameters of an Encoder Encoder, and a convolution layer and a full connection layer are added on the basis of the Encoder Encoder to form the image classification model.
The data enhancement method in the step 4) comprises image horizontal turning, image vertical turning, image random brightness adjustment, random Gaussian noise disturbance and image random rotation, wherein the range of the image random rotation angle is between-10 degrees and 10 degrees.
When the defect detection is carried out on the floor surface in the step 6), cutting the floor surface image into a plurality of images with 1024 × 1024 resolution, inputting the images into the deep learning image classification model in the step 3), and if one slice is a defect image, judging that the whole image is a defect image; and (3) inputting the floor surface image, wherein the whole image is directly input when the deep learning semantic segmentation model is input in the step 3).
When the floor surface is detected in the step 6), the floor surface image is firstly classified by the floor defect classification model after deep learning training, if the floor surface image is not defective, the floor surface image is judged to be a normal product, and if the floor surface image is defective, the floor surface image is positioned at a specific defect position by the floor semantic segmentation model after deep learning training.
The step 5) of deep learning training of the floor defect semantic segmentation model and the floor defect classification model is carried out according to the following steps:
s1: randomly sampling the image library enhanced in the step 4) according to a ratio of 4:1 to divide the image library into a training set and a verification set;
s2: iterating the deep learning semantic segmentation model in the step 3) on a training set for 150 rounds at most, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final semantic segmentation model;
s3: and 3) reading part of parameters of an Encoder of a trained deep learning semantic segmentation model Encoder by a deep learning image classification model, freezing the part of parameters, iterating on a training set for 150 rounds at most, only finely adjusting a convolution layer and a full connection layer except the Encoder in the image classification model, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final classification model.
The step 4) of S1) cutting the enhanced image into a plurality of 1024 × 1024 resolution image inputs.
During the training, the deep learning semantic segmentation model takes DICELoss as an optimization target; during training, the deep learning classification model takes crossEntropyLoss as an optimization target.
And the floor defect classification model and the floor defect semantic segmentation model trained in the step 6 are pth files derived by solidifying the model trained in the step 5.
Compared with the prior art, the invention has the following beneficial effects:
the floor defect classification model and the floor defect semantic segmentation model are trained by adopting an image classification technology and a semantic segmentation technology based on deep learning, and a newly acquired image is detected in real time by taking the models as a basis, so that the high-precision defect classification and defect positioning can be carried out on the floor in real time; compared with the traditional image identification method, the method has the advantages that the key features in the floor image are automatically extracted, the complicated manual feature extraction link is omitted, and the detection speed and precision can be improved; the invention achieves the precision of manual screening without manual interference, can reduce the production cost and promote the intellectualization of the floor industry.
Description of the drawings:
FIG. 1 is a schematic diagram of a deep learning-based floor defect detection method of the present invention;
the specific implementation mode is as follows:
the following detailed description of specific embodiments of the invention is provided, but it should be understood that the scope of the invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
As shown in fig. 1, a method for detecting a floor defect based on deep learning includes the following steps:
1) collecting a high-resolution image of a floor by using an industrial camera to establish a floor detection image library;
2) manually labeling the defect area of each defect image in the floor detection image library in the step 1) to form a floor detection image label library;
3) establishing a semantic segmentation model and an image classification model based on deep learning;
4) expanding the floor detection image library acquired in the step 1) by adopting a data enhancement technology;
5) carrying out deep learning training on the floor defect semantic segmentation model and the floor defect classification model on the image library expanded in the step 4);
6) and acquiring an image of the floor surface on line, detecting the defect of the floor surface based on the trained floor defect classification model, and positioning the specific position of the floor defect by using the floor defect semantic segmentation model if the defect exists.
The floor detection image library in the step 1) is composed of high-resolution images collected by an industrial camera, the resolution is more than 1024 × 1024, the ratio of the number of normal image samples to the number of defective image samples in the image library is close to 1:1, the total number of samples in the image library is N, and N is more than 1000.
The semantic segmentation model in the step 3) is composed of an Encoder Encoder and a Decoder Decoder, wherein the Encoder Encoder is a basic network and comprises a convolution layer, a batch normalization layer, a pooling layer and a linear activation layer and is used for automatically extracting depth features of an input image and outputting a group of feature maps; the Decoder comprises a convolution layer and an up-sampling layer and is used for screening a characteristic diagram and outputting a result image, wherein the resolution of the result image is the same as that of an input image.
The image classification model in the step 3) and the semantic segmentation model share parameters of an Encoder Encoder, and a convolution layer and a full connection layer are added on the basis of the Encoder Encoder to form the image classification model.
The data enhancement method in the step 4) comprises image horizontal turning, image vertical turning, image random brightness adjustment, random Gaussian noise disturbance and image random rotation, wherein the range of the image random rotation angle is between-10 degrees and 10 degrees.
When the defect detection is carried out on the floor surface in the step 6), cutting the floor surface image into a plurality of images with 1024 × 1024 resolution, inputting the images into the deep learning image classification model in the step 3), and if one slice is a defect image, judging that the whole image is a defect image; and (3) inputting the floor surface image, wherein the whole image is directly input when the deep learning semantic segmentation model is input in the step 3).
When the floor surface is detected in the step 6), the floor surface image is firstly classified by the floor defect classification model after deep learning training, if the floor surface image is not defective, the floor surface image is judged to be a normal product, and if the floor surface image is defective, the floor surface image is positioned at a specific defect position by the floor semantic segmentation model after deep learning training.
The step 5) of deep learning training of the floor defect semantic segmentation model and the floor defect classification model is carried out according to the following steps:
s1: randomly sampling the image library enhanced in the step 4) according to a ratio of 4:1 to divide the image library into a training set and a verification set;
s2: iterating the deep learning semantic segmentation model in the step 3) on a training set for 150 rounds at most, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final semantic segmentation model;
s3: and 3) reading part of parameters of an Encoder of a trained deep learning semantic segmentation model Encoder by a deep learning image classification model, freezing the part of parameters, iterating on a training set for 150 rounds at most, only finely adjusting a convolution layer and a full connection layer except the Encoder in the image classification model, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final classification model.
The step 4) of S1) cutting the enhanced image into a plurality of 1024 × 1024 resolution image inputs.
During the training, the deep learning semantic segmentation model takes DICELoss as an optimization target; during training, the deep learning classification model takes crossEntropyLoss as an optimization target.
And the floor defect classification model and the floor defect semantic segmentation model trained in the step 6 are pth files derived by solidifying the model trained in the step 5.
In the training process, only data in the training set can perform data enhancement operation during iteration, the verification set does not perform data enhancement, the loss value of the model on the verification set can be calculated after each iteration is finished, and the model with the minimum loss value is stored as a Pythrch pth file. The initialization parameters used by the model during training are model parameters obtained by training on the ImageNet classification data set; during training, the parameter updating mode is Adam, the initial learning rate is 0.001, and the batch size is 8.
The floor defect classification model and the floor defect semantic segmentation model are trained by adopting an image classification technology and a semantic segmentation technology based on deep learning, and a newly acquired image is detected in real time by taking the models as a basis, so that the high-precision defect classification and defect positioning can be carried out on the floor in real time; compared with the traditional image identification method, the method has the advantages that the key features in the floor image are automatically extracted, the complicated manual feature extraction link is omitted, and the detection speed and precision can be improved; the invention achieves the precision of manual screening without manual interference, can reduce the production cost and promote the intellectualization of the floor industry.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (10)
1. A floor defect detection method based on deep learning is characterized in that: the following steps:
1) collecting a high-resolution image of a floor by using an industrial camera to establish a floor detection image library;
2) manually labeling the defect area of each defect image in the floor detection image library in the step 1) to form a floor detection image label library;
3) establishing a semantic segmentation model and an image classification model based on deep learning;
4) expanding the floor detection image library acquired in the step 1) by adopting a data enhancement technology;
5) carrying out deep learning training on the floor defect semantic segmentation model and the floor defect classification model on the image library expanded in the step 4);
6) and acquiring an image of the floor surface on line, detecting the defect of the floor surface based on the trained floor defect classification model, and positioning the specific position of the floor defect by using the floor defect semantic segmentation model if the defect exists.
2. The floor defect detection method based on deep learning as claimed in claim 1, wherein: the floor detection image library in the step 1) is composed of high-resolution images collected by an industrial camera, the resolution is more than 1024 × 1024, the ratio of the number of normal image samples to the number of defective image samples in the image library is close to 1:1, the total number of samples in the image library is N, and N is more than 1000.
3. The floor defect detection method based on deep learning as claimed in claim 1, wherein: the semantic segmentation model in the step 3) is composed of an Encoder Encoder and a Decoder Decoder, wherein the Encoder Encoder is a basic network and comprises a convolution layer, a batch normalization layer, a pooling layer and a linear activation layer and is used for automatically extracting depth features of an input image and outputting a group of feature maps; the Decoder comprises a convolution layer and an up-sampling layer and is used for screening a characteristic diagram and outputting a result image, wherein the resolution of the result image is the same as that of an input image.
4. The floor defect detection method based on deep learning as claimed in claim 1, wherein: the image classification model in the step 3) and the semantic segmentation model share parameters of an Encoder Encoder, and a convolution layer and a full connection layer are added on the basis of the Encoder Encoder to form the image classification model.
5. The floor defect detection method based on deep learning as claimed in claim 1, wherein: the data enhancement method in the step 4) comprises image horizontal turning, image vertical turning, image random brightness adjustment, random Gaussian noise disturbance and image random rotation, wherein the range of the image random rotation angle is between-10 degrees and 10 degrees.
6. The floor defect detection method based on deep learning as claimed in claim 1, wherein: when the defect detection is carried out on the floor surface in the step 6), cutting the floor surface image into a plurality of images with 1024 × 1024 resolution, inputting the images into the deep learning image classification model in the step 3), and if one slice is a defect image, judging that the whole image is a defect image; and (3) inputting the floor surface image, wherein the whole image is directly input when the deep learning semantic segmentation model is input in the step 3).
7. The floor defect detection method based on deep learning as claimed in claim 1, wherein: when the floor surface is detected in the step 6), the floor surface image is firstly classified by the floor defect classification model after deep learning training, if the floor surface image is not defective, the floor surface image is judged to be a normal product, and if the floor surface image is defective, the floor surface image is positioned at a specific defect position by the floor semantic segmentation model after deep learning training.
8. The floor defect detection method based on deep learning as claimed in claim 1, wherein: the step 5) of deep learning training of the floor defect semantic segmentation model and the floor defect classification model is carried out according to the following steps:
s1: randomly sampling the image library enhanced in the step 4) according to a ratio of 4:1 to divide the image library into a training set and a verification set;
s2: iterating the deep learning semantic segmentation model in the step 3) on a training set for 150 rounds at most, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final semantic segmentation model;
s3: and 3) reading part of parameters of an Encoder of a trained deep learning semantic segmentation model Encoder by a deep learning image classification model, freezing the part of parameters, iterating on a training set for 150 rounds at most, only finely adjusting a convolution layer and a full connection layer except the Encoder in the image classification model, stopping training if the loss value is converged in the training process, and selecting the model which best appears on a verification set as a final classification model.
9. The floor defect detection method based on deep learning as claimed in claim 7, wherein: the step 4) of S1) cutting the enhanced image into a plurality of 1024 × 1024 resolution image inputs.
10. The floor defect detection method based on deep learning as claimed in claim 7, wherein: during the training, the deep learning semantic segmentation model takes DICELoss as an optimization target; during training, the deep learning classification model takes crossEntropyLoss as an optimization target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010058893.0A CN111311544B (en) | 2020-01-19 | 2020-01-19 | Floor defect detection method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010058893.0A CN111311544B (en) | 2020-01-19 | 2020-01-19 | Floor defect detection method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111311544A true CN111311544A (en) | 2020-06-19 |
| CN111311544B CN111311544B (en) | 2024-04-26 |
Family
ID=71158236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010058893.0A Active CN111311544B (en) | 2020-01-19 | 2020-01-19 | Floor defect detection method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111311544B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111986199A (en) * | 2020-09-11 | 2020-11-24 | 征图新视(江苏)科技股份有限公司 | Unsupervised deep learning-based wood floor surface flaw detection method |
| CN112132784A (en) * | 2020-08-22 | 2020-12-25 | 安徽大学 | Method for classifying and segmenting industrial magnetic tile defect image based on small sample deep convolution neural network |
| CN112365443A (en) * | 2020-10-16 | 2021-02-12 | 珠海市奥德维科技有限公司 | Hexahedron defect detection method and medium based on deep learning |
| CN112990392A (en) * | 2021-05-20 | 2021-06-18 | 四川大学 | New material floor defect target detection system based on improved YOLOv5 algorithm |
| CN113283541A (en) * | 2021-06-15 | 2021-08-20 | 无锡锤头鲨智能科技有限公司 | Automatic floor sorting method |
| CN113344888A (en) * | 2021-06-17 | 2021-09-03 | 四川启睿克科技有限公司 | Surface defect detection method and device based on combined model |
| CN113627315A (en) * | 2021-08-06 | 2021-11-09 | 上海华测导航技术股份有限公司 | Method for preprocessing data from edge-end camera to algorithmic reasoning process |
| CN113781430A (en) * | 2021-09-09 | 2021-12-10 | 北京云屿科技有限公司 | Glove surface defect detection method and system based on deep learning |
| CN113837209A (en) * | 2020-06-23 | 2021-12-24 | 乐达创意科技股份有限公司 | Method and system for improved machine learning using data for training |
| CN115240031A (en) * | 2022-07-21 | 2022-10-25 | 无锡锤头鲨智能科技有限公司 | Method and system for generating plate surface defects based on generation countermeasure network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175988A (en) * | 2019-04-25 | 2019-08-27 | 南京邮电大学 | Cloth defect inspection method based on deep learning |
| CN110458814A (en) * | 2019-07-29 | 2019-11-15 | 山东艾雷维特智能科技有限公司 | A kind of detection method of wood single-plate defect, apparatus and system |
| CN110473173A (en) * | 2019-07-24 | 2019-11-19 | 熵智科技(深圳)有限公司 | A kind of defect inspection method based on deep learning semantic segmentation |
| CN110554052A (en) * | 2019-10-12 | 2019-12-10 | 南宁市威锐康商贸有限公司 | artificial board surface defect detection method and system |
| WO2019233166A1 (en) * | 2018-06-04 | 2019-12-12 | 杭州海康威视数字技术股份有限公司 | Surface defect detection method and apparatus, and electronic device |
-
2020
- 2020-01-19 CN CN202010058893.0A patent/CN111311544B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019233166A1 (en) * | 2018-06-04 | 2019-12-12 | 杭州海康威视数字技术股份有限公司 | Surface defect detection method and apparatus, and electronic device |
| CN110175988A (en) * | 2019-04-25 | 2019-08-27 | 南京邮电大学 | Cloth defect inspection method based on deep learning |
| CN110473173A (en) * | 2019-07-24 | 2019-11-19 | 熵智科技(深圳)有限公司 | A kind of defect inspection method based on deep learning semantic segmentation |
| CN110458814A (en) * | 2019-07-29 | 2019-11-15 | 山东艾雷维特智能科技有限公司 | A kind of detection method of wood single-plate defect, apparatus and system |
| CN110554052A (en) * | 2019-10-12 | 2019-12-10 | 南宁市威锐康商贸有限公司 | artificial board surface defect detection method and system |
Non-Patent Citations (1)
| Title |
|---|
| 王理顺，钟勇，李振东，贺宜龙: "基于深度学习的织物缺陷在线检测算法", 计算机应用, vol. 39, no. 7, 10 July 2019 (2019-07-10), pages 2125 - 2128 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113837209A (en) * | 2020-06-23 | 2021-12-24 | 乐达创意科技股份有限公司 | Method and system for improved machine learning using data for training |
| CN112132784A (en) * | 2020-08-22 | 2020-12-25 | 安徽大学 | Method for classifying and segmenting industrial magnetic tile defect image based on small sample deep convolution neural network |
| CN111986199A (en) * | 2020-09-11 | 2020-11-24 | 征图新视(江苏)科技股份有限公司 | Unsupervised deep learning-based wood floor surface flaw detection method |
| CN111986199B (en) * | 2020-09-11 | 2024-04-16 | 征图新视(江苏)科技股份有限公司 | Method for detecting surface flaws of wood floor based on unsupervised deep learning |
| CN112365443A (en) * | 2020-10-16 | 2021-02-12 | 珠海市奥德维科技有限公司 | Hexahedron defect detection method and medium based on deep learning |
| CN112990392A (en) * | 2021-05-20 | 2021-06-18 | 四川大学 | New material floor defect target detection system based on improved YOLOv5 algorithm |
| CN113283541A (en) * | 2021-06-15 | 2021-08-20 | 无锡锤头鲨智能科技有限公司 | Automatic floor sorting method |
| CN113283541B (en) * | 2021-06-15 | 2022-07-22 | 无锡锤头鲨智能科技有限公司 | Automatic floor sorting method |
| WO2022262409A1 (en) * | 2021-06-15 | 2022-12-22 | 无锡锤头鲨智能科技有限公司 | Automatic floor sorting method |
| CN113344888A (en) * | 2021-06-17 | 2021-09-03 | 四川启睿克科技有限公司 | Surface defect detection method and device based on combined model |
| CN113627315A (en) * | 2021-08-06 | 2021-11-09 | 上海华测导航技术股份有限公司 | Method for preprocessing data from edge-end camera to algorithmic reasoning process |
| CN113627315B (en) * | 2021-08-06 | 2024-04-30 | 上海华测导航技术股份有限公司 | Method for preprocessing process data from edge camera to algorithm reasoning |
| CN113781430A (en) * | 2021-09-09 | 2021-12-10 | 北京云屿科技有限公司 | Glove surface defect detection method and system based on deep learning |
| CN113781430B (en) * | 2021-09-09 | 2023-08-25 | 北京云屿科技有限公司 | Glove surface defect detection method and system based on deep learning |
| CN115240031A (en) * | 2022-07-21 | 2022-10-25 | 无锡锤头鲨智能科技有限公司 | Method and system for generating plate surface defects based on generation countermeasure network |
| CN115240031B (en) * | 2022-07-21 | 2023-10-27 | 无锡锤头鲨智能科技有限公司 | Board surface defect generation method and system based on generation countermeasure network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111311544B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111311544B (en) | Floor defect detection method based on deep learning | |
| CN108961217B (en) | Surface defect detection method based on regular training | |
| CN111833306B (en) | Defect detection method and model training method for defect detection | |
| CN113870235B (en) | Circular stamping part defect detection method based on quantum firework arc edge extraction | |
| CN111414971B (en) | Convolutional neural network-based finished tea variety and grade identification method | |
| CN110927171A (en) | Bearing roller chamfer surface defect detection method based on machine vision | |
| CN112037219A (en) | Metal surface defect detection method based on two-stage convolution neural network | |
| CN114612472B (en) | SegNet improvement-based leather defect segmentation network algorithm | |
| CN111582093A (en) | Automatic small target detection method in high-resolution image based on computer vision and deep learning | |
| CN107622276B (en) | Deep learning training method based on combination of robot simulation and physical sampling | |
| CN110889837A (en) | Cloth flaw detection method with flaw classification function | |
| CN113706490B (en) | Wafer defect detection method | |
| CN109859199B (en) | Method for detecting quality of freshwater seedless pearls through SD-OCT image | |
| CN114255212A (en) | FPC surface defect detection method and system based on CNN | |
| CN112907561A (en) | Notebook appearance flaw detection method based on deep learning | |
| CN115797314B (en) | Method, system, equipment and storage medium for detecting surface defects of parts | |
| CN112884746A (en) | Character defect intelligent detection algorithm based on edge shape matching | |
| CN109685030A (en) | A kind of mug rim of a cup defects detection classification method based on convolutional neural networks | |
| CN113673618A (en) | Tobacco insect target detection method fused with attention model | |
| CN117723564B (en) | Packaging bag printing quality detection method and system based on image transmission | |
| CN115018790A (en) | Workpiece surface defect detection method based on anomaly detection | |
| CN107194432B (en) | Refrigerator door body identification method and system based on deep convolutional neural network | |
| CN111192261A (en) | Method for identifying lens defect types | |
| CN112598646B (en) | Capacitance defect detection method and device, electronic equipment and storage medium | |
| CN115471684A (en) | Injection molding workpiece template matching method, electronic equipment and storage medium |
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 |