CN111415329A - Workpiece surface defect detection method based on deep learning - Google Patents
Workpiece surface defect detection method based on deep learning Download PDFInfo
- Publication number
- CN111415329A CN111415329A CN202010105860.7A CN202010105860A CN111415329A CN 111415329 A CN111415329 A CN 111415329A CN 202010105860 A CN202010105860 A CN 202010105860A CN 111415329 A CN111415329 A CN 111415329A
- Authority
- CN
- China
- Prior art keywords
- workpiece
- convolution
- prediction
- feature
- size
- 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
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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a workpiece surface defect detection method based on deep learning, which specifically comprises the following steps: collecting workpiece images under different backgrounds and illumination conditions; preprocessing the acquired workpiece image; constructing a deep convolutional neural network model to obtain feature maps of 6 different layers; performing multi-scale feature fusion prediction by adopting a feature pyramid feature map, obtaining and generating 4 anchor box prediction target boundary boxes by using a K-means clustering algorithm, and predicting categories by using a cross entropy loss function; removing redundant prediction bounding boxes through a non-maximum suppression algorithm; and outputting the position information and the category of the workpiece surface defect. The invention solves the problems of low detection efficiency and poor precision of manual detection and physical detection methods, overcomes the problem of poor adaptability of the traditional machine vision defect detection, improves the detection efficiency and accuracy of the surface defects of workpieces, reduces the labor cost, can be quickly adapted to the surface defect detection of novel products, shortens the development period and improves the flexibility.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a workpiece surface defect detection method based on deep learning.
Background
The workpiece is a component of a precision instrument, and the surface quality and the shape structure of the workpiece have great influence on the performance of the instrument. In the process of processing, transporting and assembling workpieces, factors such as change and abrasion of tool strokes, material characteristics of the workpieces, friction and collision among the workpieces and the like can cause defects such as scratches, convex powder, concave-convex collision, cracks and the like on the surfaces of the workpieces, so that the attractiveness of the workpieces is influenced, the service performance of the workpieces is also influenced, and even serious potential safety hazards are brought to instruments. Therefore, it is urgently needed to develop a rapid and accurate method for detecting surface defects of workpieces, which has important practical application value for popularization of product quality detection technology.
At present, the workpiece quality detection methods mainly include a manual detection method, an instrument-based physical detection method and a machine vision detection method. The manual detection method is that whether the surface of the workpiece has defects or not is observed manually and visually, the type of the defects is judged by using empirical knowledge, the manual detection method is strong in subjective consciousness, time-consuming and labor-consuming, and the detection precision is difficult to guarantee. The physical detection method based on the instrument comprises ultrasonic detection, magnetic particle detection, eddy current detection, X-ray detection and the like, and can meet the surface defect detection of common workpieces, but the instrument and equipment have the defects of large volume, high price, complex maintenance, low detection efficiency and poor accuracy. Because the machine vision detection method has the advantages of no damage, accuracy, rapidness, reliability and the like, the machine vision technology is used for carrying out nondestructive detection on the surface defects of the workpiece, the subjective difference and the visual fatigue of people can be eliminated, the labor cost is reduced, the detection efficiency and the detection precision are improved, and the detection grading error is reduced. Compared with the traditional machine vision method, the deep learning method can directly learn the features from the bottom data and has higher complex structure expression capability, so that the automatic learning process is used for completely replacing the manual design of the features, the method can be suitable for different products, the development period is shortened, the flexibility is improved, and the method is quickly suitable for the surface defect detection of novel products.
In summary, the prior art has the following disadvantages:
(1) the manual detection method has strong subjective consciousness, is time-consuming and labor-consuming, and has difficult guarantee of detection precision.
(2) The physical detection method based on the instrument has the advantages of large instrument and equipment volume, high price, complex maintenance, low detection efficiency and poor accuracy.
(3) The traditional machine vision detection method needs manual feature extraction, and the product adaptability is poor.
Disclosure of Invention
In view of the above, in order to solve the above problems in the prior art, the invention provides a workpiece surface defect detection method based on deep learning, which improves the detection efficiency and accuracy of workpiece surface defects, reduces labor cost, can quickly adapt to the surface defect detection of a novel product, shortens the development period, and improves flexibility.
The invention solves the problems through the following technical means:
a workpiece surface defect detection method based on deep learning comprises the following steps:
collecting workpiece images under different backgrounds and illumination conditions;
preprocessing the acquired workpiece image;
constructing a deep convolutional neural network model to obtain feature maps of 6 different layers;
performing multi-scale feature fusion prediction by adopting a feature pyramid feature map, obtaining and generating 4 anchor box prediction target boundary boxes by using a K-means clustering algorithm, and predicting categories by using a cross entropy loss function;
removing redundant prediction bounding boxes through a non-maximum suppression algorithm;
and outputting the position information and the category of the workpiece surface defect.
Further, acquiring the workpiece images under different backgrounds and illumination conditions specifically comprises:
the CCD camera is used for shooting workpiece images under different backgrounds and illumination conditions, and the workpiece images comprise complex environments of a single background, complex backgrounds, darkness, brightness, soft light, shielding and overlapping.
Further, the preprocessing of the acquired workpiece image specifically includes:
firstly, performing data enhancement on an obtained workpiece image, wherein the data enhancement comprises optical transformation and geometric transformation; the optical transformation comprises random brightness adjustment, random contrast adjustment and random channel adjustment, and the geometric transformation comprises rotation, stretching, translation, horizontal turning, vertical turning, random picture cutting and random zooming;
then, generating corresponding marking information for the expanded image data set by using a marking tool; the marking information comprises position information and categories of the surface defects of the workpiece in the sample, wherein the categories are normal, scratch, convex powder, concave-convex and crack;
finally, dividing the data set into a training set and a testing set; wherein, the training set is 70% of the total sample amount, and the testing set is 30%.
Further, constructing a deep convolutional neural network model to obtain feature maps of 6 different layers specifically includes:
in the deep convolutional neural network module, each convolutional layer is operated by a Re L u function and normalized Batchnormalization;
(1) the basic network structure comprises a VGG16 basic feature extraction module, a first 13 convolution layers of VGG16 are firstly passed through and modified, dense Conv + stride Conv is used for replacing dense Conv + maxporoling in a VGG16 network structure, wherein the convolution kernel size of dense Conv is 3 × 3, the step size is 1, the convolution kernel size of stride Conv is 3 × 3, the step size is 2, then two convolutions, Conv6 and Conv7 are used for replacing full connection layers, FC6 and FC7, of VGG16 for further extracting features, wherein Conv6 is a void convolution, the convolution kernel size is 3 × 3, the void number is 6, the step size is 1, the convolution kernel size of Conv7 is 1 × 1, and the step size is 1;
(2) and (3) depth convolutional layers, namely adding 4 depth convolutional layers on the basic network structure to further extract higher semantic information, wherein the depth convolutional layers are Conv8, Conv9, Conv10 and Conv11 respectively, and each depth convolutional layer consists of 2 convolutions, namely a convolution kernel with the size of 1 × 1 and the step size of 1 and a convolution kernel with the size of 3 × 3 and the step size of 2.
Further, performing multi-scale feature fusion prediction by using a feature pyramid feature map, obtaining 4 anchor box prediction target bounding boxes by using a K-means clustering algorithm, and predicting categories by using a cross entropy loss function, wherein the method specifically comprises the following steps:
obtaining 6 feature maps which are 63 × 63, 32 × 32, 16 × 16, 8 × 8, 4 × 4 and 1 × 1 by utilizing a deep convolutional neural network, performing multi-scale feature fusion prediction on the feature maps of 6 different layers by adopting a feature pyramid method, clustering by using a K-means algorithm to obtain 4 anchor box prediction target boundary frames, and predicting the category by using a cross entropy loss function;
the FPN overall architecture comprises 4 parts, namely a top-down network, a bottom-up network, transverse connection and convolution fusion;
top-down networking: is a deep convolutional neural network;
performing convolution with the size of 1 × 1 to reduce the number of channels on a characteristic diagram 1 × 1 obtained by Conv11 in the depth convolution layer to obtain F6, and then sequentially performing 2 times of nearest neighbor upsampling operation;
performing 1 × 1 convolution operation on bottom-layer feature maps obtained by Conv9, Conv8, Conv7, Conv5 and Conv4 in the deep convolutional neural network respectively to ensure that the number of channels is fixed to 256, and adding the up-sampled high semantic features element by element to obtain F5, F4, F3, F2 and F1;
convolution fusion F5, F4, F3, F2 and F1 are fused by convolution with the size of 3 × 3.
Further, removing redundant prediction bounding boxes through a non-maximum suppression algorithm specifically comprises:
removing repeated prediction bounding boxes by using a modified algorithm Soft-NMS of a non-maximum suppression algorithm;
the concrete expression of the Soft-NMS function is as follows:
in the formula, siFor each bounding box score, M is the bounding box with the highest current score, biIs a certain bounding box remaining, NtTo set the threshold value 0.5.
Compared with the prior art, the invention has the beneficial effects that at least:
(1) the dense conv + stride conv structure is used for replacing dense conv + maxporoling in VGG16 for down-sampling, more characteristic information is reserved, the speed is higher, and the detection efficiency is effectively improved.
(2) The method has the advantages that the deep convolutional neural network extraction features are built, 4 deep convolutional layers are added on the basis of a network structure with VGG16 as a basic network structure, multi-scale feature fusion prediction is carried out by adopting FPN, higher semantic information is extracted, the detection accuracy is improved, the generalization capability of a model is enhanced, and the flexibility is improved.
(3) By using the Soft-NMS algorithm to remove repeated prediction bounding boxes, the problem of missed detection caused by overlapping of a plurality of workpieces can be solved, and the recall rate and the robustness of detection are improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is an overall flow chart of the method for detecting surface defects of a workpiece based on deep learning according to the present invention;
FIG. 2 is a detailed step diagram of the method for detecting surface defects of a workpiece based on deep learning according to the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It should be noted that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and all other embodiments obtained by those skilled in the art without any inventive work based on the embodiments of the present invention belong to the protection scope of the present invention.
Examples
As shown in fig. 1, the invention discloses a workpiece surface defect detection method based on deep learning, which specifically comprises the following steps: collecting workpiece images under different backgrounds and illumination conditions; preprocessing the acquired workpiece image; constructing a deep convolutional neural network model to obtain feature maps of 6 different layers; performing multi-scale feature fusion prediction by adopting a feature pyramid feature map, obtaining and generating 4 anchor box prediction target boundary boxes by using a K-means clustering algorithm, and predicting categories by using a cross entropy loss function; removing redundant prediction bounding boxes through a non-maximum suppression algorithm; and outputting the position information and the category of the workpiece surface defect.
As shown in fig. 2, the method for detecting surface defects of a workpiece based on deep learning of the present invention specifically includes the following steps:
step 1: workpiece image acquisition
The CCD camera is used for shooting the workpiece images under different backgrounds and illumination conditions, and the workpiece images comprise complex environments such as a single background, a complex background, darkness, brightness, soft light, shielding, overlapping and the like.
Step 2: image pre-processing
First, data enhancement is performed on the workpiece image obtained in step 1. Data enhancement mainly includes optical transformation and geometric transformation. The optical transformation comprises random brightness adjustment, random contrast adjustment and random channel adjustment, and the geometric transformation comprises rotation, stretching, translation, horizontal turning, vertical turning, random picture cutting and random zooming.
Then, the corresponding labeling information is generated for the expanded image data set by using a labeling tool. The marking information comprises position information and categories of the workpiece surface defects in the sample, wherein the categories are normal, scratch, convex powder, concave-convex and crack.
Finally, the data set is divided into a training set and a testing set. Wherein, the training set is 70% of the total sample amount, and the testing set is 30%.
And step 3: constructing deep convolutional neural network to extract features
And in the deep convolutional neural network module, each convolutional layer is operated by an Re L u function and normalized Batchnormalization.
(1) The basic network structure adopts VGG16 as a basic feature extraction module, firstly, the first 13 convolution layers of VGG16 are passed through and modified, dense Conv + stride Conv is used to replace dense Conv + maxporoling in a VGG16 network structure, wherein the convolution kernel size of dense Conv is 3 × 3, the step size is 1, the convolution kernel size of stride Conv is 3 × 3, the step size is 2, then, two convolutions, Conv6 and Conv7 are used to replace full connection layers, FC6 and FC7 of VGG16 to further extract features, wherein Conv6 is a hole convolution, the convolution kernel size is 3 × 3, the number of holes is 6, the step size is 1, the convolution kernel size of Conv7 is 1 × 1, and the step size is 1.
(2) And adding 4 depth convolutional layers on the basic network structure, and further extracting higher semantic information, wherein the depth convolutional layers are Conv8, Conv9, Conv10 and Conv11 respectively, and each depth convolutional layer consists of 2 convolutions, namely a convolution kernel with the size of 1 × 1 and the step size of 1 and a convolution kernel with the size of 3 × 3 and the step size of 2.
And 4, step 4: multi-scale feature fusion prediction
The method comprises the steps of obtaining 6 Feature maps which are 63 × 63, 32 × 32, 16 × 16, 8 × 8, 4 × 4 and 1 × 1 through a deep convolutional neural Network, conducting multi-scale Feature fusion prediction on the Feature maps of 6 different layers through a Feature Pyramid (FPN) method, clustering through a K-means algorithm to obtain 4 anchor box prediction target boundary boxes, and predicting categories through a cross entropy loss function.
The FPN overall architecture comprises 4 parts of top-down network, bottom-up network, transverse connection and convolution fusion.
Top-down networking: for deep convolutional neural networks
And (3) performing convolution with the size of 1 × 1 on a characteristic diagram 1 × 1 obtained by Conv11 in the depth convolutional layer to reduce the number of channels to obtain F6, and then sequentially performing 2 times of nearest neighbor upsampling operation.
And (3) performing 1 × 1 convolution operation on bottom-layer feature maps obtained by Conv9, Conv8, Conv7, Conv5 and Conv4 in the deep convolutional neural network respectively to ensure that the number of channels is fixed to 256, and adding the up-sampled high semantic features element by element to obtain F5, F4, F3, F2 and F1.
Convolution fusion F5, F4, F3, F2 and F1 are fused by convolution with the size of 3 × 3.
And 5: non-maxima suppression algorithm deduplication
A modified algorithm Soft-NMS, which is a Non-Maximum Suppression algorithm (NMS), is used to remove duplicate prediction bounding boxes. The concrete expression of the Soft-NMS function is as follows:
in the formula, siFor each bounding box score, M is the bounding box with the highest current score, biIs a certain bounding box remaining, NtTo set the threshold value 0.5.
Step 6: the result of the detection
And outputting the position information and the category of the workpiece surface defect.
According to the invention, a dense conv + strand conv structure is used for replacing dense conv + maxporoling in VGG16 for downsampling, more characteristic information is reserved, the speed is higher, and the detection efficiency is effectively improved.
According to the invention, the deep convolutional neural network extraction features are constructed, 4 deep convolutional layers are added on a network structure taking VGG16 as a basic network structure, and multi-scale feature fusion prediction is carried out by adopting FPN (field programmable gate array), so that higher semantic information is extracted, the detection accuracy is improved, the generalization capability of a model is enhanced, and the flexibility is improved.
The method uses the Soft-NMS algorithm to remove repeated prediction bounding boxes, can solve the problem of missed detection caused by overlapping of a plurality of workpieces, and improves the recall rate and the robustness of detection.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (6)
1. A workpiece surface defect detection method based on deep learning is characterized by comprising the following steps:
collecting workpiece images under different backgrounds and illumination conditions;
preprocessing the acquired workpiece image;
constructing a deep convolutional neural network model to obtain feature maps of 6 different layers;
performing multi-scale feature fusion prediction by adopting a feature pyramid feature map, obtaining and generating 4 anchor box prediction target boundary boxes by using a K-means clustering algorithm, and predicting categories by using a cross entropy loss function;
removing redundant prediction bounding boxes through a non-maximum suppression algorithm;
and outputting the position information and the category of the workpiece surface defect.
2. The method for detecting the surface defects of the workpiece based on the deep learning as claimed in claim 1, wherein the step of collecting the workpiece images under different backgrounds and illumination conditions is specifically as follows:
the CCD camera is used for shooting workpiece images under different backgrounds and illumination conditions, and the workpiece images comprise complex environments of a single background, complex backgrounds, darkness, brightness, soft light, shielding and overlapping.
3. The workpiece surface defect detection method based on deep learning of claim 1, wherein the preprocessing of the acquired workpiece image is specifically:
firstly, performing data enhancement on an obtained workpiece image, wherein the data enhancement comprises optical transformation and geometric transformation; the optical transformation comprises random brightness adjustment, random contrast adjustment and random channel adjustment, and the geometric transformation comprises rotation, stretching, translation, horizontal turning, vertical turning, random picture cutting and random zooming;
then, generating corresponding marking information for the expanded image data set by using a marking tool; the marking information comprises position information and categories of the surface defects of the workpiece in the sample, wherein the categories are normal, scratch, convex powder, concave-convex and crack;
finally, dividing the data set into a training set and a testing set; wherein, the training set is 70% of the total sample amount, and the testing set is 30%.
4. The workpiece surface defect detection method based on deep learning of claim 1, wherein the step of constructing the deep convolutional neural network model to obtain the feature maps of 6 different layers is specifically as follows:
in the deep convolutional neural network module, each convolutional layer is operated by a Re L u function and normalized Batchnormalization;
(1) the basic network structure comprises a VGG16 basic feature extraction module, a first 13 convolution layers of VGG16 are firstly passed through and modified, dense Conv + stride Conv is used for replacing dense Conv + maxporoling in a VGG16 network structure, wherein the convolution kernel size of dense Conv is 3 × 3, the step size is 1, the convolution kernel size of stride Conv is 3 × 3, the step size is 2, then two convolutions, Conv6 and Conv7 are used for replacing full connection layers, FC6 and FC7, of VGG16 for further extracting features, wherein Conv6 is a void convolution, the convolution kernel size is 3 × 3, the void number is 6, the step size is 1, the convolution kernel size of Conv7 is 1 × 1, and the step size is 1;
(2) and (3) depth convolutional layers, namely adding 4 depth convolutional layers on the basic network structure to further extract higher semantic information, wherein the depth convolutional layers are Conv8, Conv9, Conv10 and Conv11 respectively, and each depth convolutional layer consists of 2 convolutions, namely a convolution kernel with the size of 1 × 1 and the step size of 1 and a convolution kernel with the size of 3 × 3 and the step size of 2.
5. The workpiece surface defect detection method based on deep learning of claim 1, wherein a feature pyramid feature map is adopted for multi-scale feature fusion prediction, a K-means clustering algorithm is used to obtain 4 anchor box prediction target bounding boxes, and the prediction categories by using a cross entropy loss function are specifically as follows:
obtaining 6 feature maps which are 63 × 63, 32 × 32, 16 × 16, 8 × 8, 4 × 4 and 1 × 1 by utilizing a deep convolutional neural network, performing multi-scale feature fusion prediction on the feature maps of 6 different layers by adopting a feature pyramid method, clustering by using a K-means algorithm to obtain 4 anchor box prediction target boundary frames, and predicting the category by using a cross entropy loss function;
the FPN overall architecture comprises 4 parts, namely a top-down network, a bottom-up network, transverse connection and convolution fusion;
top-down networking: is a deep convolutional neural network;
performing convolution with the size of 1 × 1 to reduce the number of channels on a characteristic diagram 1 × 1 obtained by Conv11 in the depth convolution layer to obtain F6, and then sequentially performing 2 times of nearest neighbor upsampling operation;
performing 1 × 1 convolution operation on bottom-layer feature maps obtained by Conv9, Conv8, Conv7, Conv5 and Conv4 in the deep convolutional neural network respectively to ensure that the number of channels is fixed to 256, and adding the up-sampled high semantic features element by element to obtain F5, F4, F3, F2 and F1;
convolution fusion F5, F4, F3, F2 and F1 are fused by convolution with the size of 3 × 3.
6. The method for detecting the surface defects of the workpiece based on the deep learning as claimed in claim 1, wherein the removing of the redundant prediction bounding box by the non-maximum suppression algorithm is specifically as follows:
removing repeated prediction bounding boxes by using a modified algorithm Soft-NMS of a non-maximum suppression algorithm;
the concrete expression of the Soft-NMS function is as follows:
in the formula, siFor each bounding box score, M is the bounding box with the highest current score, biIs a certain bounding box remaining, NtTo set the threshold value 0.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010105860.7A CN111415329B (en) | 2020-02-20 | 2020-02-20 | Workpiece surface defect detection method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010105860.7A CN111415329B (en) | 2020-02-20 | 2020-02-20 | Workpiece surface defect detection method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111415329A true CN111415329A (en) | 2020-07-14 |
| CN111415329B CN111415329B (en) | 2023-08-01 |
Family
ID=71492719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010105860.7A Active CN111415329B (en) | 2020-02-20 | 2020-02-20 | Workpiece surface defect detection method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111415329B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329861A (en) * | 2020-11-06 | 2021-02-05 | 北京工业大学 | Layered feature fusion method for multi-target detection of mobile robot |
| CN112419237A (en) * | 2020-11-03 | 2021-02-26 | 中国计量大学 | Automobile clutch master cylinder groove surface defect detection method based on deep learning |
| CN112686843A (en) * | 2020-12-21 | 2021-04-20 | 福建新大陆软件工程有限公司 | Wood board defect detection method and system based on neural network |
| CN112734760A (en) * | 2021-03-31 | 2021-04-30 | 惠州高视科技有限公司 | Semiconductor bump defect detection method, electronic device, and storage medium |
| CN112950584A (en) * | 2021-03-01 | 2021-06-11 | 哈尔滨工程大学 | Coating surface defect identification method based on deep learning |
| CN113096098A (en) * | 2021-04-14 | 2021-07-09 | 大连理工大学 | Casting appearance defect detection method based on deep learning |
| CN113111719A (en) * | 2021-03-16 | 2021-07-13 | 苏州海宸威视智能科技有限公司 | Vehicle detection method under unmanned aerial vehicle visual angle based on reconstructed brain-like convolution |
| CN114076794A (en) * | 2020-08-19 | 2022-02-22 | 宝山钢铁股份有限公司 | Automatic detection device and detection method for near-surface defects of small strip steel square billet |
| CN114581806A (en) * | 2022-03-18 | 2022-06-03 | 重庆科技学院 | Industrial part no-load rate calculation method based on trunk edge feature extraction |
| GB2604991A (en) * | 2021-01-14 | 2022-09-21 | Nvidia Corp | Performing non-maximum suppression in parallel |
| CN116109639A (en) * | 2023-04-13 | 2023-05-12 | 常州微亿智造科技有限公司 | Defect detection method and device based on multi-task learning network |
| CN117152577A (en) * | 2023-10-30 | 2023-12-01 | 北京机械工业自动化研究所有限公司 | Casting defect detection method, casting defect detection device, electronic equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110660040A (en) * | 2019-07-24 | 2020-01-07 | 浙江工业大学 | Industrial product irregular defect detection method based on deep learning |
-
2020
- 2020-02-20 CN CN202010105860.7A patent/CN111415329B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110660040A (en) * | 2019-07-24 | 2020-01-07 | 浙江工业大学 | Industrial product irregular defect detection method based on deep learning |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114076794A (en) * | 2020-08-19 | 2022-02-22 | 宝山钢铁股份有限公司 | Automatic detection device and detection method for near-surface defects of small strip steel square billet |
| CN112419237A (en) * | 2020-11-03 | 2021-02-26 | 中国计量大学 | Automobile clutch master cylinder groove surface defect detection method based on deep learning |
| CN112419237B (en) * | 2020-11-03 | 2023-06-30 | 中国计量大学 | Deep learning-based automobile clutch master cylinder groove surface defect detection method |
| CN112329861A (en) * | 2020-11-06 | 2021-02-05 | 北京工业大学 | Layered feature fusion method for multi-target detection of mobile robot |
| CN112686843A (en) * | 2020-12-21 | 2021-04-20 | 福建新大陆软件工程有限公司 | Wood board defect detection method and system based on neural network |
| CN112686843B (en) * | 2020-12-21 | 2023-09-15 | 福建新大陆软件工程有限公司 | Board defect detection method and system based on neural network |
| GB2604991B (en) * | 2021-01-14 | 2023-08-02 | Nvidia Corp | Performing non-maximum suppression in parallel |
| GB2604991A (en) * | 2021-01-14 | 2022-09-21 | Nvidia Corp | Performing non-maximum suppression in parallel |
| CN112950584B (en) * | 2021-03-01 | 2022-07-29 | 哈尔滨工程大学 | Coating surface defect identification method based on deep learning |
| CN112950584A (en) * | 2021-03-01 | 2021-06-11 | 哈尔滨工程大学 | Coating surface defect identification method based on deep learning |
| CN113111719A (en) * | 2021-03-16 | 2021-07-13 | 苏州海宸威视智能科技有限公司 | Vehicle detection method under unmanned aerial vehicle visual angle based on reconstructed brain-like convolution |
| CN112734760B (en) * | 2021-03-31 | 2021-08-06 | 高视科技(苏州)有限公司 | Semiconductor bump defect detection method, electronic device, and storage medium |
| CN112734760A (en) * | 2021-03-31 | 2021-04-30 | 惠州高视科技有限公司 | Semiconductor bump defect detection method, electronic device, and storage medium |
| CN113096098A (en) * | 2021-04-14 | 2021-07-09 | 大连理工大学 | Casting appearance defect detection method based on deep learning |
| CN113096098B (en) * | 2021-04-14 | 2024-01-05 | 大连理工大学 | Casting appearance defect detection method based on deep learning |
| CN114581806A (en) * | 2022-03-18 | 2022-06-03 | 重庆科技学院 | Industrial part no-load rate calculation method based on trunk edge feature extraction |
| CN114581806B (en) * | 2022-03-18 | 2024-03-19 | 重庆科技学院 | Industrial part empty rate calculation method based on trunk edge feature extraction |
| CN116109639A (en) * | 2023-04-13 | 2023-05-12 | 常州微亿智造科技有限公司 | Defect detection method and device based on multi-task learning network |
| CN116109639B (en) * | 2023-04-13 | 2023-08-18 | 常州微亿智造科技有限公司 | Defect detection method and device based on multi-task learning network |
| CN117152577A (en) * | 2023-10-30 | 2023-12-01 | 北京机械工业自动化研究所有限公司 | Casting defect detection method, casting defect detection device, electronic equipment and storage medium |
| CN117152577B (en) * | 2023-10-30 | 2024-02-02 | 北京机械工业自动化研究所有限公司 | Casting defect detection method, casting defect detection device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111415329B (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111415329B (en) | Workpiece surface defect detection method based on deep learning | |
| Ren et al. | State of the art in defect detection based on machine vision | |
| CN108562589B (en) | Method for detecting surface defects of magnetic circuit material | |
| CN112967243B (en) | Deep learning chip packaging crack defect detection method based on YOLO | |
| CN111862064B (en) | Silver wire surface flaw identification method based on deep learning | |
| CN109724984B (en) | Defect detection and identification device and method based on deep learning algorithm | |
| CN111784633B (en) | Insulator defect automatic detection algorithm for electric power inspection video | |
| CN111681232B (en) | Industrial welding image defect detection method based on semantic segmentation | |
| CN109191421B (en) | Visual detection method for pits on circumferential surface of cylindrical lithium battery | |
| CN109215009B (en) | Continuous casting billet surface image defect detection method based on deep convolution neural network | |
| CN109615604B (en) | Part appearance flaw detection method based on image reconstruction convolutional neural network | |
| CN112037219A (en) | Metal surface defect detection method based on two-stage convolution neural network | |
| CN112614105B (en) | Depth network-based 3D point cloud welding spot defect detection method | |
| CN112308826A (en) | Bridge structure surface defect detection method based on convolutional neural network | |
| CN114972316A (en) | Battery case end surface defect real-time detection method based on improved YOLOv5 | |
| CN111986145A (en) | Bearing roller flaw detection method based on fast-RCNN | |
| CN114663382A (en) | Surface defect detection method for electronic component based on YOLOv5 convolutional neural network | |
| Han et al. | SSGD: A smartphone screen glass dataset for defect detection | |
| CN113516652A (en) | Battery surface defect and adhesive detection method, device, medium and electronic equipment | |
| CN111179278B (en) | Image detection method, device, equipment and storage medium | |
| CN112257849A (en) | Intelligent detection method, system and device based on deep learning and application thereof | |
| CN115731228B (en) | Gold-plated chip defect detection system and method | |
| CN116385364A (en) | Multi-level ground lead defect identification method based on parallax auxiliary semantic segmentation | |
| Chen et al. | Real-time defect detection of TFT-LCD displays using a lightweight network architecture | |
| CN113192018B (en) | Water-cooled wall surface defect video identification method based on fast segmentation convolutional neural 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 |