CN116481461B - Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile - Google Patents
Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile Download PDFInfo
- Publication number
- CN116481461B CN116481461B CN202211486394.7A CN202211486394A CN116481461B CN 116481461 B CN116481461 B CN 116481461B CN 202211486394 A CN202211486394 A CN 202211486394A CN 116481461 B CN116481461 B CN 116481461B
- Authority
- CN
- China
- Prior art keywords
- hole
- sound
- roughness
- automobile
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000009413 insulation Methods 0.000 title claims description 25
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 24
- 238000011156 evaluation Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 230000030808 detection of mechanical stimulus involved in sensory perception of sound Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000010030 laminating Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 9
- 238000012545 processing Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000011900 installation process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
- G01B11/306—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces for measuring evenness
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- 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
-
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8883—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges involving the calculation of gauges, generating models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Abstract
The invention provides a method for detecting roughness of a hole forming incision of an automobile sound-proof and heat-proof spare and accessory part, which is suitable for the field of automobile production, and has the advantages of high identification speed, high accuracy, great reduction of cost and the like compared with the traditional manual method by carrying out image acquisition and tag adding on a spare and accessory part hole site, carrying out image preprocessing on a tag, constructing an improved Alexnet network hole forming incision roughness detection model, carrying out model training and model prediction, and obtaining a detection model meeting the requirements by taking the accuracy as a control index.
Description
Technical Field
The invention relates to a method for detecting roughness of a hole forming and cutting of an automobile sound and heat insulation part, which is suitable for the field of automobile part production.
Background
In the automobile production process, due to the requirements of comfort and functionality, a large number of sound and heat insulation parts exist, holes are cut on the sound and heat insulation parts to meet the requirements of different forms, but the roughness of a hole-forming cut of the automobile sound and heat insulation parts has great influence on the subsequent installation process due to the hole-forming process of the cut, so that the automobile sound and heat insulation parts need to be detected, the conventional detection mode is a manual method, namely, the roughness of an empty cut is identified by adopting the naked eye mode, and the mode has high cost and low efficiency and cannot meet the requirements of automobile assembly line production.
Therefore, it is necessary to develop a method for detecting the roughness of the notch hole of the automobile spare part, which has high efficiency, high speed, high accuracy and low cost.
Disclosure of Invention
The invention aims to solve the problems that the existing automobile sound and heat insulation part hole forming incision roughness detection method is high in cost and low in efficiency, cannot meet the requirements of automobile assembly line production and the like, and provides the automobile sound and heat insulation part hole forming incision roughness detection method which is applicable to the automobile production field and has the advantages of being high in efficiency, high in speed, high in accuracy, low in cost and the like, and can be widely applied to the automobile production field.
The aim of the invention can be achieved by adopting the following technical scheme:
s101, placing the automobile sound and heat insulation spare and accessory parts on an operation table.
S102, numbering holes of the sound-and-heat-insulation parts of the automobile to obtain a hole position number set { K } 1 ，K 2 ...K n For each hole site K by an image acquisition device at the upper part of the operation table n Image acquisition with fixed size and position is carried out, a large amount of data acquisition work is carried out in the early stage, and each hole site K n Obtaining an image set { T ] 1 ，T 2 ...T m }。
S103, for each hole site K n Image set { T } 1 ，T 2 ...T m Evaluating the roughness of hole-forming notch according to the past sample, classifying according to the evaluation result to obtain each hole site K n Is a label-containing image set {0OR 1=t ] 1 ，0OR1=T 2 ...0OR1=T m }。
S104, each hole site K n Is { T } of the image set containing tags 1 ，T 2 ...T m The method comprises the steps of dividing a training set and a testing set into different folders, and respectively utilizing program scripts to respectively divide the training set and the testing setEach picture of the set is combined into a form of a picture path plus tag and stored in different text files for model reading.
S105, preprocessing the images of the training set and the testing set to obtain each hole site K n And processing the image set.
S106, aiming at hole site K n And constructing an improved Alexnet network pore-forming incision roughness detection model.
S107, using hole site K n Model training is performed on the improved Alexnet network model by the training set data of the (A).
S108, using hole site K n Model test is carried out on the improved Alexnet network model according to the test set data of (1), and each hole site K is aimed at n All obtain pore-forming notch roughness detection model M meeting the requirements n The method is applied to hole forming and notch roughness detection of sound and heat insulation spare and accessory parts of the automobile.
In the S101 further, the operation panel is provided with a correction and leveling device and an image acquisition device, and the image acquisition device acquires the image of the hole-forming incision position after the position adjustment by the correction and adjustment device.
Further, in S103, the evaluation categories are passed and failed, and the corresponding labels are 1 and 0.
In the step S105, the preprocessing includes size normalization, image graying, histogram equalization and graphic enhancement of the test set image, and size normalization, image graying and histogram equalization of the test set image.
In the further step S106, the improved Alexnet network model is totally divided into 8 layers, and is mainly composed of a convolution layer, a LeakRelu layer, a pooling layer and a full-connection superposition.
In further step S106, the improved Alexnet network model uses a linear leak-corrected relu activation function, expression (1),
(1)
Where leak is a small constant.
In the further step S106, the modified Alexnet network model adopts dropout regularization to perform elimination processing on all neurons according to the probability of 20%.
In further S106, the main parameters of the modified Alexnet network model are set as follows: the battsize was 64, the momentum was set to 0.95, the ω decay rate was set to 0.0005, and the learning rate was 0.001.
In the step S108, the network training effect adopts Accuracy (acc) as an evaluation index, the expression is shown in the formula (2), generally acc is more than or equal to 90%, the model is considered to meet the requirements,
(2)
Wherein: x is x 1 Representing the number of label qualified samples predicted to be qualified; x is x 2 Representing the number of label qualified samples predicted to be unqualified; y is 1 Representing the number of failed label samples predicted to be failed; y is 2 The number of label reject samples predicted to be acceptable is indicated.
The beneficial effects of the invention are as follows: the problems that an existing method for detecting the roughness of a hole forming and cutting of a sound-proof and heat-proof spare part of an automobile is high in cost and low in efficiency, cannot meet the requirements of automobile assembly line production and the like, and a displacement prediction model of a foundation pit supporting structure is poor in prediction data selection rule and poor in prediction model effect are solved; the method for detecting the roughness of the hole forming and cutting of the sound and heat insulating spare and accessory parts of the automobile has the advantages of high efficiency, high speed, high accuracy, low cost and the like, and can be widely applied to the field of automobile production.
Drawings
Fig. 1: the invention relates to a flow chart of a method for detecting roughness of a hole forming and cutting of an automobile sound and heat insulation spare and accessory part;
fig. 2: the improved Alexnet network model structure is an improved Alexnet network model structure of the automobile sound and heat insulation spare and accessory part pore-forming incision roughness detection method;
in the figure: 1 is a convolution layer, 101 is 48 convolution channels, 201 is 128 convolution channels, 301 is 192 convolution channels, 2 is a BN layer, 3 is a LeakRelu layer, 4 is a pooling layer, 5 is a fully connected layer, 6 is a Dropout regularization layer, 7 is a softmax, 8 is an input image, 100 is a first layer, 200 is a second layer, 300 is a third layer, 400 is a fourth layer, 500 is a fifth layer, 600 is a sixth layer, 700 is a seventh layer, and 800 is an eighth layer.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings; it should be understood that the particular embodiments presented herein are illustrative and explanatory only and are not restrictive of the invention.
The following is a specific example of a method for detecting roughness of a hole forming and cutting of a sound and heat insulating spare and accessory part of an automobile.
Referring to FIG. 1, a flow chart of a method for detecting roughness of a hole forming and cutting of an automobile sound and heat insulation part according to the invention is shown.
Step S101, placing the automobile sound and heat insulation spare and accessory parts on an operation table. Further, the operation table is provided with a correction leveling device and an image acquisition device, and after the position adjustment is carried out through the correction adjusting device, the image acquisition is carried out on the position of the pore-forming incision through the image acquisition device.
Step S102, numbering the hole sites of the sound and heat insulation spare and accessory parts of the automobile to obtain a hole site number set { K } 1 ，K 2 ...K n For each hole site K by an image acquisition device at the upper part of the operation table n Image acquisition with fixed size and position is carried out, a large amount of data acquisition work is carried out in the early stage, and each hole site K n Obtaining an image set { T ] 1 ，T 2 ...T m }。
Step S103, for each hole site K n Image set { T } 1 ，T 2 ...T m Evaluating the roughness of hole-forming notch according to the past sample, classifying according to the evaluation result to obtain each hole site K n Is a label-containing image set {0OR 1=t ] 1 ，0OR1=T 2 ...0OR1=T m }. Further, the evaluation categories are passed and failedThe corresponding labels are 1 and 0.
Step S104, each hole site K n Is { T } of the image set containing tags 1 ，T 2 ...T m The method comprises the steps of dividing a training set and a testing set, respectively placing the training set and the testing set in different folders, respectively combining each picture of the training set and each picture of the testing set into a picture path and label form by utilizing a program script, and storing the picture path and label form into different text files for reading by a model.
Step S105, preprocessing the images of the training set and the testing set to obtain each hole site K n And processing the image set. Further, the preprocessing comprises size normalization, image graying, histogram equalization and graphic enhancement on the test set image, and size normalization, image graying and histogram equalization on the test set image.
Step S106, aiming at hole site K n And constructing an improved Alexnet network pore-forming incision roughness detection model. Further, the improved Alexnet network model is totally divided into 8 layers, and is mainly composed of a convolution layer, a LeakRelu layer, a pooling layer and full-connection superposition. The improved Alexnet network model adopts a linear LeakRelu activation function with leakage correction, the expression is shown in the formula (1),
(1)
Where leak is a small constant.
The improved Alexnet network model adopts Droput regularization to eliminate all neurons according to 20% probability. The main parameters of the improved Alexnet network model are set as follows: the battsize was 64, the momentum was set to 0.95, the ω decay rate was set to 0.0005, and the learning rate was 0.001.
Step S107, using hole site K n Model training is performed on the improved Alexnet network model by the training set data of the (A).
Step S108, using hole site K n Model test is carried out on the improved Alexnet network model according to the test set data of (1), and each hole site K is aimed at n All obtain the pore-forming notch Mao Caodu meeting the requirementsDetection model M n The method is applied to hole forming and notch roughness detection of sound and heat insulation spare and accessory parts of the automobile. Furthermore, the network training effect adopts Accuracy (acc) as an evaluation index, the expression is shown in the formula (2), generally acc is more than or equal to 90 percent, the model is considered to meet the requirements,
(2)
Wherein: x is x 1 Representing the number of label qualified samples predicted to be qualified; x is x 2 Representing the number of label qualified samples predicted to be unqualified; y is 1 Representing the number of failed label samples predicted to be failed; y is 2 The number of label reject samples predicted to be acceptable is indicated.
The method comprises the steps of carrying out image acquisition and tag adding on the hole sites of the parts, carrying out image preprocessing on the tag, constructing an improved Alexnet network hole forming incision roughness detection model, carrying out model training and model prediction, and obtaining a detection model meeting the requirements by taking the accuracy as a control index to be applied to the hole roughness detection work.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the particular embodiments disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (5)
1. The method for detecting the roughness of the pore-forming and notch of the sound-proof and heat-proof spare and accessory part of the automobile is characterized by comprising the following steps of:
1) Placing the automobile sound and heat insulation spare and accessory parts on an operation desk;
2) For the automobile sound and heat insulation spare and accessory parts, the hole positions are numbered to obtain a hole position number set { K } 1 ，K 2 ...K n For each hole site K by an image acquisition device at the upper part of the operation table n Performing fixed size andimage acquisition of the position, a large amount of data acquisition work is carried out in the early stage, and each hole site K n Obtaining an image set { T ] 1 ，T 2 ...T m }；
3) For each hole site K n Image set { T } 1 ，T 2 ...T m Evaluating the roughness of hole-forming notch according to the past sample, classifying according to the evaluation result to obtain each hole site K n Is a label-containing image set {0OR 1=t ] 1 ，0OR1=T 2 ...0OR1=T m }；
4) Each hole site K n Is {0OR 1=t } 1 ，0OR1=T 2 ...0OR1=T m Dividing the training set and the testing set into a training set and a testing set, respectively placing the training set and the testing set in different folders, respectively combining each picture of the training set and the testing set into a picture path and label form by utilizing a program script, and storing the picture path and label form into different text files for reading by a model;
5) Preprocessing the images of the training set and the testing set to obtain each hole site K n A processed image set;
6) For hole site K n Constructing an improved Alexnet network pore-forming incision roughness detection model, wherein the improved Alexnet network model is totally divided into 8 layers and is formed by laminating a convolution layer, a LeakRelu layer, a pooling layer and a full-connection layer, the improved Alexnet network model adopts a linear LeakRelu activation function with leakage correction, the expression is shown in the formula (1),
(1)
In the formula, the leak is a very small constant, the modified Alexnet network model adopts Droput regularization, all neurons are eliminated according to the probability of 20%, and the parameters of the modified Alexnet network model are set as follows: the battsize is 64, the momentum is set to 0.95, the omega decay rate is set to 0.0005, and the learning rate is set to 0.001;
7) Using hole sites K n Model training is carried out on the improved Alexnet network model by the training set data of the (2);
8) Using hole sites K n Model test is carried out on the improved Alexnet network model according to the test set data of (1), and each hole site K is aimed at n All obtain pore-forming notch roughness detection model M meeting the requirements n The method is applied to hole forming and notch roughness detection of sound and heat insulation spare and accessory parts of the automobile.
2. The method for detecting the roughness of the hole forming and cutting of the sound and heat insulation spare and accessory parts of the automobile according to claim 1, which is characterized in that: in the step 1), an operation table is provided with a correction and leveling device and an image acquisition device, and after the position adjustment is carried out through the correction and adjustment device, the image acquisition device is used for carrying out image acquisition on the position of the pore-forming incision.
3. The method for detecting the roughness of the hole forming and cutting of the sound and heat insulation spare and accessory parts of the automobile according to claim 1, which is characterized in that: in the 3), the evaluation categories are qualified and unqualified, and the corresponding labels are 1 and 0.
4. The method for detecting the roughness of the hole forming and cutting of the sound and heat insulation spare and accessory parts of the automobile according to claim 1, which is characterized in that: in the step 5), the preprocessing comprises size normalization, image graying, histogram equalization and graphic enhancement on the test set image, and size normalization, image graying and histogram equalization on the test set image.
5. The method for detecting the roughness of the hole forming and cutting of the sound and heat insulation spare and accessory parts of the automobile according to claim 1, which is characterized in that: in the 8), the network training effect adopts Accuracy (acc) as an evaluation index, the expression is shown in the formula (2), acc is more than or equal to 90 percent, the model is considered to meet the requirements,
(2)
Wherein: x is x 1 Representing label passing sample predictionsIs qualified number; x is x 2 Representing the number of label qualified samples predicted to be unqualified; y is 1 Representing the number of failed label samples predicted to be failed; y is 2 The number of label reject samples predicted to be acceptable is indicated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211486394.7A CN116481461B (en) | 2022-11-24 | 2022-11-24 | Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211486394.7A CN116481461B (en) | 2022-11-24 | 2022-11-24 | Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116481461A CN116481461A (en) | 2023-07-25 |
| CN116481461B true CN116481461B (en) | 2023-09-22 |
Family
ID=87220153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211486394.7A Active CN116481461B (en) | 2022-11-24 | 2022-11-24 | Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116481461B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104764712A (en) * | 2015-04-29 | 2015-07-08 | 浙江工业大学 | Method for detecting quality of inner wall of via hole of PCB |
| CN107703146A (en) * | 2017-09-30 | 2018-02-16 | 北京得华机器人技术研究院有限公司 | A kind of auto-parts vision detection system and method |
| CN108776808A (en) * | 2018-05-25 | 2018-11-09 | 北京百度网讯科技有限公司 | A kind of method and apparatus for detecting ladle corrosion defect |
| CN110009614A (en) * | 2019-03-29 | 2019-07-12 | 北京百度网讯科技有限公司 | Method and apparatus for output information |
| CN110175614A (en) * | 2019-04-23 | 2019-08-27 | 浙江工业大学 | A kind of detection method of printed circuit board via hole inner wall quality |
| CN110930347A (en) * | 2018-09-04 | 2020-03-27 | 京东方科技集团股份有限公司 | Convolutional neural network training method, and method and device for detecting welding spot defects |
| CN111724358A (en) * | 2020-06-11 | 2020-09-29 | 华中科技大学 | Concrete quality detection method and system based on image and convolutional neural network |
| 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 |
| JP2021135903A (en) * | 2020-02-28 | 2021-09-13 | 武蔵精密工業株式会社 | Defective product image generation program and quality determination device |
| CN114399484A (en) * | 2022-01-11 | 2022-04-26 | 华电电力科学研究院有限公司 | Photovoltaic module defect detection method, device, equipment and medium |
| CN114841937A (en) * | 2022-04-21 | 2022-08-02 | 燕山大学 | Detection method for detecting surface defects of automobile hub |
| CN115082392A (en) * | 2022-06-14 | 2022-09-20 | 西安邮电大学 | Silver plating defect detection method and system for semiconductor lead frame |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3739513A1 (en) * | 2019-05-13 | 2020-11-18 | Fujitsu Limited | Surface defect identification method and apparatus |
-
2022
- 2022-11-24 CN CN202211486394.7A patent/CN116481461B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104764712A (en) * | 2015-04-29 | 2015-07-08 | 浙江工业大学 | Method for detecting quality of inner wall of via hole of PCB |
| CN107703146A (en) * | 2017-09-30 | 2018-02-16 | 北京得华机器人技术研究院有限公司 | A kind of auto-parts vision detection system and method |
| CN108776808A (en) * | 2018-05-25 | 2018-11-09 | 北京百度网讯科技有限公司 | A kind of method and apparatus for detecting ladle corrosion defect |
| US11276158B2 (en) * | 2018-05-25 | 2022-03-15 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for inspecting corrosion defect of ladle |
| CN110930347A (en) * | 2018-09-04 | 2020-03-27 | 京东方科技集团股份有限公司 | Convolutional neural network training method, and method and device for detecting welding spot defects |
| CN110009614A (en) * | 2019-03-29 | 2019-07-12 | 北京百度网讯科技有限公司 | Method and apparatus for output information |
| CN110175614A (en) * | 2019-04-23 | 2019-08-27 | 浙江工业大学 | A kind of detection method of printed circuit board via hole inner wall quality |
| JP2021135903A (en) * | 2020-02-28 | 2021-09-13 | 武蔵精密工業株式会社 | Defective product image generation program and quality determination device |
| CN111724358A (en) * | 2020-06-11 | 2020-09-29 | 华中科技大学 | Concrete quality detection method and system based on image and convolutional neural network |
| 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 |
| CN114399484A (en) * | 2022-01-11 | 2022-04-26 | 华电电力科学研究院有限公司 | Photovoltaic module defect detection method, device, equipment and medium |
| CN114841937A (en) * | 2022-04-21 | 2022-08-02 | 燕山大学 | Detection method for detecting surface defects of automobile hub |
| CN115082392A (en) * | 2022-06-14 | 2022-09-20 | 西安邮电大学 | Silver plating defect detection method and system for semiconductor lead frame |
Non-Patent Citations (1)
| Title |
|---|
| 岩质隧道炮孔图像识别算法及光面爆破参数优化研究;张万志;《中国博士学位论文全文数据库（电子期刊）》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116481461A (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8239189B2 (en) | Method and system for estimating a sentiment for an entity | |
| CN114240878A (en) | Routing inspection scene-oriented insulator defect detection neural network construction and optimization method | |
| CN109473123A (en) | Voice activity detection method and device | |
| CN104966097A (en) | Complex character recognition method based on deep learning | |
| CN106846961B (en) | Electronic test paper processing method and device | |
| CN108053545B (en) | Certificate verification method and device, server and storage medium | |
| CN110110768A (en) | Fault Diagnosis of Roller Bearings based on Concurrent Feature study and multi-categorizer | |
| CN110334187A (en) | Burmese sentiment analysis method and device based on transfer learning | |
| CN103500216B (en) | Method for extracting file information | |
| CN111222330B (en) | Chinese event detection method and system | |
| CN112767423B (en) | Remote sensing image building segmentation method based on improved SegNet | |
| CN111400446A (en) | Standard text duplicate checking method and system | |
| CN113672931B (en) | Software vulnerability automatic detection method and device based on pre-training | |
| CN102087657A (en) | IO (Input/output) to point method and IO (Input/output) to point for nuclear power plant simulation platform | |
| CN115100668A (en) | Method and device for identifying table information in image | |
| CN113989487A (en) | Fault defect detection method and system for live-action scheduling | |
| CN116481461B (en) | Method for detecting roughness of hole forming and notch of sound and heat insulation spare and accessory parts of automobile | |
| CN115526234A (en) | Cross-domain model training and log anomaly detection method and device based on transfer learning | |
| Heeringa et al. | Computational dialectology | |
| CN117197057A (en) | Automatic detection method for corrosion degree of steel material based on deep learning | |
| CN111767919A (en) | Target detection method for multi-layer bidirectional feature extraction and fusion | |
| CN115249329A (en) | Apple leaf disease detection method based on deep learning | |
| Yan et al. | Yolo-pd: Abnormal signal detection in gas pipelines based on improved yolov7 | |
| CN113869194A (en) | Variable parameter milling process signal marking method and system based on deep learning | |
| CN114239539A (en) | English composition off-topic detection method and device |
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 |