CN114654315A - Machine vision detection system and method for poor grinding of tapered roller base surface - Google Patents
Machine vision detection system and method for poor grinding of tapered roller base surface Download PDFInfo
- Publication number
- CN114654315A CN114654315A CN202210144693.6A CN202210144693A CN114654315A CN 114654315 A CN114654315 A CN 114654315A CN 202210144693 A CN202210144693 A CN 202210144693A CN 114654315 A CN114654315 A CN 114654315A
- Authority
- CN
- China
- Prior art keywords
- image
- base surface
- neural network
- grinding
- roller base
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000227 grinding Methods 0.000 title claims abstract description 62
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000013528 artificial neural network Methods 0.000 claims abstract description 25
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 25
- 230000011218 segmentation Effects 0.000 claims abstract description 11
- 238000002372 labelling Methods 0.000 claims abstract description 6
- 230000007547 defect Effects 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 8
- 210000002569 neuron Anatomy 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 5
- 230000002950 deficient Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 230000000694 effects Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B5/00—Machines or devices designed for grinding surfaces of revolution on work, including those which also grind adjacent plane surfaces; Accessories therefor
- B24B5/02—Machines or devices designed for grinding surfaces of revolution on work, including those which also grind adjacent plane surfaces; Accessories therefor involving centres or chucks for holding work
- B24B5/14—Machines or devices designed for grinding surfaces of revolution on work, including those which also grind adjacent plane surfaces; Accessories therefor involving centres or chucks for holding work for grinding conical surfaces, e.g. of centres
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B24—GRINDING; POLISHING
- B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
- B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
- B24B49/12—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation involving optical means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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]
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)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Mechanical Engineering (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a machine vision detection system and a method for poor grinding of a tapered roller base surface, belonging to the technical field of tapered roller grinding, wherein the system comprises a parallel light source, an area-array camera and an upper computer; performing region segmentation and size adjustment on the image in the original data set; labeling the processed data set to obtain a training data set; establishing a convolutional neural network comprising a multilayer neural network structure, and training the neural network based on a training data set; collecting an image of a roller base plane by using an area-array camera during detection; carrying out region segmentation and size adjustment on the original image; and inputting the compliance sub-image into the trained convolution neural network to obtain an output value of the last layer of neural network structure, and judging poor grinding. The invention has the advantages of high detection efficiency and high detection precision, can solve the problems of low efficiency, high labor cost and easy omission of manual detection means, and improves the production efficiency and the production quality.
Description
Technical Field
The invention relates to the technical field of tapered roller grinding, in particular to a machine vision detection system and method for poor grinding of a tapered roller base surface.
Background
When the tapered roller is subjected to base surface grinding, the rollers are pushed into a grinding area one by one according to a fixed posture (such as a small end is in front of a large end and is behind the large end). If the tapered roller does not enter the grinding area in the correct posture, the defects of the base surface such as under grinding, missing grinding and the like can be caused, and the defects are collectively called as poor grinding. The poor grinding of the base surface of the tapered roller belongs to a serious defect, which can affect the working performance of the roller and cause safety accidents in serious cases. Therefore, detection of the grinding defect of the tapered roller base surface is a very important requirement.
Currently, the detection of the grinding defect of the tapered roller base surface is mainly manual detection. For more obvious and larger area poor grinding, the grinding can be directly observed by naked eyes under the irradiation of a conventional visible light source; however, the grinding of the base surface with a weak and small area is poor, and the base surface is difficult to distinguish by human eyes under the conventional condition, and a high-resolution detection instrument is required for observation. The problems of low efficiency, low precision and low reliability exist in manual detection of poor base surface grinding.
Therefore, how to improve the detection accuracy rate of the grinding defects of the tapered roller base surface is a problem to be solved urgently.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a machine vision detection system and a machine vision detection method for poor grinding of a tapered roller base surface, so as to solve the problem of how to improve the detection accuracy rate of the poor grinding of the tapered roller base surface.
The technical scheme of the invention is as follows: a machine vision detection method for poor grinding of a tapered roller base surface comprises the following steps:
step 1: selecting a qualified roller and a defective roller containing poor grinding of a base surface, respectively placing the qualified roller and the defective roller under an area array camera, and acquiring an original image of the base surface of the roller under the irradiation of a parallel light source, wherein the image resolution is 1920 multiplied by 1080 pixels to obtain an original data set;
and 2, step: carrying out region segmentation on the image in the original data set to obtain a base plane region sub-image; carrying out size adjustment on the sub-images, and converting the sub-images into 300 x 300 pixels to obtain compliant sub-images; labeling the processed data set to obtain a training data set;
and step 3: establishing a convolutional neural network comprising a multilayer neural network structure aiming at poor grinding data and qualified data, and training the neural network based on a training data set;
and 4, step 4: aiming at a roller to be detected, acquiring an image of a roller base plane by using the area array camera; performing region segmentation and size adjustment on the original image to obtain a compliance sub-image of a basal plane region;
and 5: and inputting the compliance sub-image into the trained convolutional neural network to obtain an output value of the last layer of neural network structure, and further judging poor grinding.
The region segmentation in step 1 and step 4 aims to extract a minimum horizontal circumscribed rectangular region including a roller base surface region from an original image including a roller base surface and a background region, so as to obtain a base surface region sub-image, specifically:
1) due to the irradiation of the parallel light source, the brightness of the roller base surface area is greater than that of the background area, binaryzation is carried out on the original image I according to a formula 1 to obtain a binaryzation image Ib, wherein h is a binaryzation threshold value, and the binaryzation threshold value is selected according to the actual situation of the image;
2) searching a connected domain Cmax with the largest area in a binary image Ib, wherein a region corresponding to the connected domain is a roller base surface region;
3) calculating a minimum horizontal bounding rectangle R of Cmax that satisfies the following condition: a horizontal rectangle including all pixels of the connected component Cmax having the smallest area;
4) and extracting a region surrounded by the rectangle R in the original image I, namely a roller base plane region sub-image.
The marking in the step 2 refers to manually classifying the images according to the roller types (qualified rollers or poor grinding rollers) corresponding to the images, and the images with known types form a data set for neural network training in the subsequent step.
The convolutional neural network comprises 14 layers of neurons, and is composed of 1 input convolutional layer, 3 residual connecting modules and 1 output full connecting layer. The convolution kernel size of the input convolution layer is 7 × 7, the step length is 2, and the activation function adopts a ReLU (reconstructed Linear Units, modified Linear Units); each residual connecting module comprises 4 convolutional layers, the sizes of the convolutional cores are 3 multiplied by 3, the step length is 1, and the activation function adopts ReLU; the number of the neurons of the output full-connection layer is 2, and the classification result is qualified and poor grinding.
The cross entropy loss function is adopted during the convolutional neural network training, the random gradient descent method is used as an optimization algorithm, the learning rate of each training is smaller than or equal to the learning rate of the previous training, the convolutional neural network is trained for multiple times by adopting the training data set during the training, and the parameters of the neural network are adjusted according to the set learning rate, so that the trained neural network is obtained.
And aiming at the marked poor grinding and normal images of the base surface, a multilayer convolutional neural network is established, and the convolutional neural network is trained based on a training data set, so that the output of the neural network accords with the actual marking result, and the poor grinding of the base surface can be detected by using the trained convolutional neural network.
A machine vision detection system for poor grinding of a tapered roller base surface comprises a parallel light source, an area array camera and an upper computer; the parallel light source emits a plurality of beams of blue light rays with high parallelism; the device is used for illuminating the roller base surface area and enabling the roller base surface area and the background area to form obvious bright-dark contrast; the area array camera is used for shooting the base plane of the tapered roller and sending the base plane image of the tapered roller to the upper computer; the upper computer is used for operating an extraction and compliance processing algorithm of the roller base plane sub-image and detecting poor grinding of the tapered roller base plane through a deep learning model.
The method adopts the deep learning model to detect the poor grinding of the tapered roller base surface, has the remarkable advantages of high detection efficiency and high detection precision, can effectively solve the problems of low efficiency, high labor cost and easy omission of the conventional manual detection means, and improves the production efficiency and the production quality.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is an original image of poor grinding;
fig. 3 is a sub-image of the basal plane area.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, a machine vision inspection method for poor grinding of a tapered roller base surface is characterized by comprising the following steps:
step 1: selecting a qualified roller and a defective roller containing poor grinding of a base surface, respectively placing the qualified roller and the defective roller under the area array camera, and acquiring an original image of the base surface of the roller under the irradiation of the parallel light source, wherein the image resolution is 1920 x 1080 pixels to obtain an original data set;
step 2: carrying out region segmentation on the image in the original data set to obtain a base plane region sub-image; carrying out size adjustment on the sub-images, and converting the sub-images into 300 x 300 pixels to obtain compliant sub-images; labeling the processed data set to obtain a training data set;
and 3, step 3: establishing a convolutional neural network comprising a multilayer neural network structure aiming at poor grinding data and qualified data, and training the neural network based on a training data set;
and 4, step 4: aiming at a roller to be detected, acquiring an image of a roller base plane by using the area array camera; performing region segmentation and size adjustment on the original image to obtain a compliance sub-image of a basal plane region;
and 5: and inputting the compliance sub-image into the trained convolutional neural network to obtain an output value of the last layer of neural network structure, and further judging poor grinding.
The region segmentation in step 1 and step 4 aims to extract a minimum horizontal circumscribed rectangular region including a roller base surface region from an original image including a roller base surface and a background region, so as to obtain a base surface region sub-image, specifically:
1) due to the irradiation of the parallel light source, the brightness of the roller base surface area is greater than that of the background area, binaryzation is carried out on the original image I according to a formula 1 to obtain a binaryzation image Ib, wherein h is a binaryzation threshold value, and the binaryzation threshold value is selected according to the actual situation of the image;
2) searching a connected domain Cmax with the largest area in a binary image Ib, wherein a region corresponding to the connected domain is a roller base surface region;
3) calculating a minimum horizontal bounding rectangle R of Cmax that satisfies the following condition: a horizontal rectangle in which the area of all pixels including the connected component Cmax is the smallest;
4) and extracting a region surrounded by the rectangle R in the original image I, namely a roller base plane region sub-image.
The labeling in the step 2 refers to manually classifying the images according to the roller types (qualified rollers or poor grinding rollers) corresponding to the images, and the images with known types form a data set for neural network training in the subsequent step.
The convolutional neural network described in step 3 comprises 14 layers of neurons, and each neuron is composed of 1 input convolutional layer, 3 residual error connection modules and 1 output full connection layer. The convolution kernel size of the input convolution layer is 7 × 7, the step length is 2, and the activation function adopts a ReLU (Rectified Linear Unit); each residual connecting module comprises 4 convolutional layers, the sizes of the convolutional cores are 3 multiplied by 3, the step length is 1, and the activation function adopts ReLU; the number of the neurons of the output full-connection layer is 2, and the classification result is qualified and poor grinding.
And a cross entropy loss function is adopted during the convolutional neural network training, and a random gradient descent method is used as an optimization algorithm. The learning rate of each training is less than or equal to the learning rate of the previous training. And when the convolutional neural network is trained, the training data set is adopted for training for multiple times, and the parameters of the neural network are adjusted according to the set learning rate, so that the trained neural network is obtained.
And aiming at the marked poor grinding and normal images of the base surface, a multilayer convolutional neural network is established, and the convolutional neural network is trained based on a training data set, so that the output of the neural network accords with the actual marking result, and the poor grinding of the base surface can be detected by using the trained convolutional neural network.
A machine vision detection system for poor grinding of a tapered roller base surface comprises a parallel light source, an area array camera and an upper computer;
the parallel light source emits a plurality of beams of blue light rays with high parallelism; the device is used for illuminating the roller base surface area and enabling the roller base surface area and the background area to form obvious bright-dark contrast, and a good illuminating effect can be formed;
the area array camera is used for shooting the base plane of the tapered roller and sending the base plane image of the tapered roller to an upper computer;
the upper computer is used for operating an extraction and compliance processing algorithm of the roller base plane sub-image and detecting poor grinding of the tapered roller base plane through a deep learning model.
The method adopts the deep learning model to detect the poor grinding of the tapered roller base surface, has the remarkable advantages of high detection efficiency and high detection precision, can effectively solve the problems of low efficiency, high labor cost and easy omission of the conventional manual detection means, and improves the production efficiency and the production quality.
Claims (9)
1. A machine vision detection method for poor grinding of a tapered roller base surface is characterized by comprising the following steps:
step 1: selecting a qualified roller and a defective roller containing poor grinding of a base surface, respectively placing the rollers under an area array camera, and acquiring an original image of the base surface of the roller under the irradiation of a parallel light source to obtain an original data set;
step 2: carrying out region segmentation on the image in the original data set to obtain a base plane region sub-image; adjusting the size of the sub-image to obtain a compliant sub-image; labeling the processed data set to obtain a training data set;
and 3, step 3: establishing a convolutional neural network comprising a multilayer neural network structure aiming at poor grinding data and qualified data, and training the neural network based on a training data set;
and 4, step 4: aiming at a roller to be detected, acquiring an image of a roller base plane by using the area array camera; performing region segmentation and size adjustment on the original image to obtain a compliance sub-image of a basal plane region;
and 5: and inputting the compliance sub-image into the trained convolutional neural network to obtain an output value of the last layer of neural network structure, and further judging poor grinding.
2. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: the region segmentation in step 1 and step 4 aims to extract a minimum horizontal circumscribed rectangular region including a roller base surface region from an original image including a roller base surface and a background region, so as to obtain a base surface region sub-image, specifically:
1) due to the irradiation of the parallel light source, the brightness of the roller base surface area is greater than that of the background area, binaryzation is carried out on the original image I according to a formula 1 to obtain a binaryzation image Ib, wherein h is a binaryzation threshold value, and the binaryzation threshold value is selected according to the actual situation of the image;
2) searching a connected domain Cmax with the largest area in a binary image Ib, wherein a region corresponding to the connected domain is a roller base surface region;
3) calculating a minimum horizontal bounding rectangle R of Cmax that satisfies the following condition: a horizontal rectangle in which the area of all pixels including the connected component Cmax is the smallest;
4) and extracting an area surrounded by the rectangle R in the original image I, namely a roller basal plane area sub-image.
3. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: and 2, the labeling means that the images are manually classified according to the roller types corresponding to the images, and the images with known types form a data set for neural network training in the subsequent step.
4. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: the convolutional neural network comprises 14 layers of neurons, and the neurons respectively consist of 1 input convolutional layer, 3 residual error connection modules and 1 output full connection layer, wherein the size of a convolutional core of the input convolutional layer is 7 multiplied by 7, the step length is 2, and an activation function adopts ReLU; each residual connecting module comprises 4 convolutional layers, the sizes of the convolutional cores are 3 multiplied by 3, the step length is 1, and the activation function adopts ReLU; the number of the neurons of the output full-connection layer is 2, and the classification result is qualified and poor grinding.
5. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: the cross entropy loss function is adopted during the convolutional neural network training, the random gradient descent method is used as an optimization algorithm, the learning rate of each training is smaller than or equal to the learning rate of the previous training, the convolutional neural network is trained for multiple times by adopting the training data set during the training, and the parameters of the neural network are adjusted according to the set learning rate, so that the trained neural network is obtained.
6. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: and aiming at the marked poor grinding and normal images of the base surface, establishing a multilayer convolutional neural network, training the convolutional neural network based on a training data set, and enabling the output of the neural network to accord with the actual marking result, so that the poor grinding of the base surface is detected by utilizing the trained convolutional neural network.
7. The machine vision inspection method for the grinding defect of the tapered roller base surface according to claim 1, characterized in that: the resolution of the original image of the roller floor is 1920 × 1080 pixels and the resolution of the compliant sub-image is 300 × 300 pixels.
8. The utility model provides a poor machine vision detecting system of tapered roller base plane grinding which characterized in that: the system comprises a parallel light source, an area array camera and an upper computer;
the parallel light source is used for illuminating the roller base surface area and enabling the roller base surface area and the background area to form obvious bright-dark contrast;
the area array camera is used for shooting the base plane of the tapered roller and sending the base plane image of the tapered roller to an upper computer;
the upper computer is used for operating an extraction and compliance processing algorithm of the roller base plane sub-image and detecting poor grinding of the tapered roller base plane through a deep learning model.
9. The system of claim 8 for machine vision inspection of poor grinding of a tapered roller base surface, wherein: the parallel light source emits a plurality of blue light rays with high parallelism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210144693.6A CN114654315A (en) | 2022-02-17 | 2022-02-17 | Machine vision detection system and method for poor grinding of tapered roller base surface |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210144693.6A CN114654315A (en) | 2022-02-17 | 2022-02-17 | Machine vision detection system and method for poor grinding of tapered roller base surface |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114654315A true CN114654315A (en) | 2022-06-24 |
Family
ID=82026892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210144693.6A Pending CN114654315A (en) | 2022-02-17 | 2022-02-17 | Machine vision detection system and method for poor grinding of tapered roller base surface |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114654315A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103264019A (en) * | 2013-04-25 | 2013-08-28 | 洛阳久德轴承模具技术有限公司 | Tapered roller surface defect detection system |
| CN104655635A (en) * | 2013-11-25 | 2015-05-27 | 刘晶 | Automatic surface defect detector for bearing dust cap |
| CN110927171A (en) * | 2019-12-09 | 2020-03-27 | 中国科学院沈阳自动化研究所 | Bearing roller chamfer surface defect detection method based on machine vision |
| CN110974214A (en) * | 2019-12-20 | 2020-04-10 | 华中科技大学 | Automatic electrocardiogram classification method, system and equipment based on deep learning |
| CN111402196A (en) * | 2020-02-10 | 2020-07-10 | 浙江工业大学 | Bearing roller image generation method based on countermeasure generation network |
| CN111507990A (en) * | 2020-04-20 | 2020-08-07 | 南京航空航天大学 | Tunnel surface defect segmentation method based on deep learning |
| CN111660147A (en) * | 2020-06-28 | 2020-09-15 | 上海理工大学 | Conical roller spherical base surface grinding technological parameter optimization method |
| CN111861990A (en) * | 2020-06-10 | 2020-10-30 | 宜通世纪物联网研究院(广州)有限公司 | Method, system and storage medium for detecting bad appearance of product |
-
2022
- 2022-02-17 CN CN202210144693.6A patent/CN114654315A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103264019A (en) * | 2013-04-25 | 2013-08-28 | 洛阳久德轴承模具技术有限公司 | Tapered roller surface defect detection system |
| CN104655635A (en) * | 2013-11-25 | 2015-05-27 | 刘晶 | Automatic surface defect detector for bearing dust cap |
| CN110927171A (en) * | 2019-12-09 | 2020-03-27 | 中国科学院沈阳自动化研究所 | Bearing roller chamfer surface defect detection method based on machine vision |
| CN110974214A (en) * | 2019-12-20 | 2020-04-10 | 华中科技大学 | Automatic electrocardiogram classification method, system and equipment based on deep learning |
| CN111402196A (en) * | 2020-02-10 | 2020-07-10 | 浙江工业大学 | Bearing roller image generation method based on countermeasure generation network |
| CN111507990A (en) * | 2020-04-20 | 2020-08-07 | 南京航空航天大学 | Tunnel surface defect segmentation method based on deep learning |
| CN111861990A (en) * | 2020-06-10 | 2020-10-30 | 宜通世纪物联网研究院(广州)有限公司 | Method, system and storage medium for detecting bad appearance of product |
| CN111660147A (en) * | 2020-06-28 | 2020-09-15 | 上海理工大学 | Conical roller spherical base surface grinding technological parameter optimization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106875373B (en) | Mobile phone screen MURA defect detection method based on convolutional neural network pruning algorithm | |
| CN111275679B (en) | Image-based solar cell defect detection system and method | |
| CN101424645B (en) | Soldered ball surface defect detection device and method based on machine vision | |
| CN108492291B (en) | CNN segmentation-based solar photovoltaic silicon wafer defect detection system and method | |
| WO2023168972A1 (en) | Linear array camera-based copper surface defect detection method and apparatus | |
| CN110930390B (en) | Chip pin missing detection method based on semi-supervised deep learning | |
| CN107966454A (en) | A kind of end plug defect detecting device and detection method based on FPGA | |
| CN109919908B (en) | Method and device for detecting defects of light-emitting diode chip | |
| CN113469953B (en) | Transmission line insulator defect detection method based on improved YOLOv4 algorithm | |
| CN109142383B (en) | Character defect detection method based on morphology | |
| CN113554631B (en) | Chip surface defect detection method based on improved network | |
| CN104881665A (en) | Chip character identification and verification method and apparatus | |
| CN112381175A (en) | Circuit board identification and analysis method based on image processing | |
| CN112215821A (en) | Photovoltaic cell panel grid line defect detection method and system based on artificial intelligence | |
| CN112014407A (en) | Method for detecting surface defects of integrated circuit wafer | |
| CN112200790A (en) | Cloth defect detection method, device and medium | |
| CN105277560A (en) | Fabric pilling grade evaluation and analysis method and device | |
| CN117455917B (en) | Establishment of false alarm library of etched lead frame and false alarm on-line judging and screening method | |
| KR20210086303A (en) | Pattern inspection apparatus based on deep learning and inspection method using the same | |
| CN114549414A (en) | Abnormal change detection method and system for track data | |
| CN107545565B (en) | Solar screen plate detection method | |
| CN113962929A (en) | Photovoltaic cell assembly defect detection method and system and photovoltaic cell assembly production line | |
| CN114654315A (en) | Machine vision detection system and method for poor grinding of tapered roller base surface | |
| CN117422970A (en) | Improved defect PCB detection method based on YOLOv8 | |
| CN114120318B (en) | Dark field image target point accurate extraction method based on integrated decision tree |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220624 |