CN111105396A - Printed matter quality detection method and system based on artificial intelligence - Google Patents
Printed matter quality detection method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN111105396A CN111105396A CN201911274249.0A CN201911274249A CN111105396A CN 111105396 A CN111105396 A CN 111105396A CN 201911274249 A CN201911274249 A CN 201911274249A CN 111105396 A CN111105396 A CN 111105396A
- Authority
- CN
- China
- Prior art keywords
- detection model
- data set
- printed matter
- quality detection
- training
- 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
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
- G06F18/21—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/30144—Printing quality
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Inking, Control Or Cleaning Of Printing Machines (AREA)
Abstract
The invention discloses a presswork quality detection method and system based on artificial intelligence, and belongs to the technical field of image processing. The printing quality detection method based on artificial intelligence trains a printing quality detection model by using a collected and preprocessed data set to obtain a mature printing quality detection model, detects the quality of the printing by using the printing quality detection model, and displays a detection result. The method for detecting the quality of the printed matter based on the artificial intelligence can be used for detecting the quality problem of the printed matter more efficiently, further ensuring the quality of the printed matter and having good popularization and application values.
Description
Technical Field
The invention relates to the technical field of image processing, and particularly provides a presswork quality detection method and system based on artificial intelligence.
Background
With the continuous progress of society, the rapid development of social economy drives the continuous development of various social technologies, and the requirements are higher and higher. The printing technology is an ancient technology in China, and with the continuous progress of society, the requirements of people on the printing technology are higher and higher, and the quality requirements on printed matters are also stricter and stricter, so that the quality detection of the printed matters becomes a necessary and critical step after finished products are produced. In the prior art, the product detection of a printing factory is manually finished, the efficiency is low, the cost is high, errors can occur in the detection standard, and the quality of printed matters is uneven. Therefore, how to develop a print quality detection technology to solve the problem becomes a technical problem which needs to be solved urgently.
Disclosure of Invention
The technical task of the invention is to provide a printing quality detection method based on artificial intelligence, which can more efficiently detect the quality problem of the printed matter and further ensure the quality of the printed matter.
A further technical task of the present invention is to provide a printing quality detection system based on artificial intelligence.
In order to achieve the purpose, the invention provides the following technical scheme:
a printing quality detection method based on artificial intelligence is characterized in that a collected and preprocessed data set is used for training a printing quality detection model to obtain a mature printing quality detection model, the printing quality detection model is used for detecting the quality of a printed matter, and a detection result is displayed.
According to the printing quality detection method based on artificial intelligence, a camera is used for collecting a large number of printed product result pictures.
And detecting the quality of the printed matter by using the printed matter quality detection model, and displaying the detection result on a display.
Preferably, the method specifically comprises the following steps:
s1, establishing a data set: collecting a large number of printed product result pictures, forming a data set according to the qualification of the collected pictures, and labeling;
s2, preprocessing of the data set: preprocessing pictures in the data set;
s3, establishing a printed matter detection model;
s4, training a presswork detection model by using the data set marked in the step S1 to obtain a mature printing quality detection model;
and S5, reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
Preferably, in step S2, the preprocessing of the data set includes denoising, binarizing, character segmenting, and normalizing the processed pictures in the data set.
Preferably, the presswork detection model is a deep convolutional neural network model, the pictures are subjected to feature extraction by using multilayer convolution and pooling, the pictures are classified by using three fully-connected layers, and the catch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
The printed matter detection model is a deep convolutional neural network model and has the following structure: the first layer consists of two 3x3x64 convolutional layers and one 3x3/2 maximum pooling layer, the second layer consists of two 3x3x128 convolutional layers and one 3x3/2 maximum pooling layer, the third layer consists of four 3x3x256 convolutional layers and one 3x3/2 maximum pooling layer, the fourth layer consists of four 3x3x512 convolutional layers and one 3x3/2 maximum pooling layer, and the fifth layer consists of three fully-connected layer sets.
Preferably, in step S4, the data set is divided by a cross-validation method, the print detection model is trained by using a small-batch gradient descent algorithm, the obtained print detection model calculates a cost according to a cost function, and the print detection model is modified by back propagation according to the cost until a mature print quality detection model is obtained, wherein softmax is used as the cost function.
Preferably, step S4 specifically includes the following processes:
s401, marking the collected presswork pictures according to the condition of being qualified, and dividing a training set into a training test set and a verification set according to the ratio of 1: 5;
s402, dividing and verifying a training test set by using a ten-fold cross verification method, wherein one tenth of the training test set is used as the test set, and nine tenths are used as the training set;
s403, training a presswork detection model by using a training set test set in the training process;
s404, obtaining a prediction result by using a voting method, and calculating the cost of the printed matter detection model according to a cost function by using a verification set;
and S405, selecting a mature presswork quality detection model according to the cost function.
The utility model provides a printing quality detects system based on artificial intelligence, this system includes dataset building module, dataset preprocessing module, printed matter detection model building module, printed matter detection model training module and printing quality testing result display module:
the data set establishing module is used for establishing a data set, collecting a large number of printed product result pictures, forming the data set according to whether the collected pictures are qualified or not, and marking the data set;
the data set preprocessing module is used for preprocessing the data set: preprocessing pictures in the data set;
the printed matter detection model establishing module is used for establishing a printed matter detection model;
the printed matter detection model training module is used for training a printed matter detection model by using the marked data set to obtain a mature printed matter quality detection model;
the printing quality detection result display module is used for reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
Preferably, the data set preprocessing module preprocesses the data set into the processes of denoising, binaryzation, character segmentation and normalization of the image processed in the data set.
Preferably, the presswork detection model established by the presswork detection model establishing module is a deep convolutional neural network model, the pictures are subjected to feature extraction by using multilayer convolution and pooling, three full-connection layers are used for classification, and the batch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
Preferably, in the printed matter detection model training module, the data set is divided by a ten-fold cross validation method, the printed matter detection model is trained by a small-batch gradient descent algorithm, the obtained printed matter detection model calculates cost according to a cost function, and the printed matter detection model is modified by back propagation according to the cost until a mature printed matter quality detection model is obtained, wherein softmax is used as the cost function.
Compared with the prior art, the artificial intelligence-based presswork quality detection method has the following outstanding beneficial effects: the artificial intelligence based printed matter quality detection method can be used for detecting the quality problem of the printed matter more efficiently and further ensuring the quality of the printed matter, meanwhile, the problems of overfitting, gradient disappearance, gradient explosion and the like in the neural network training process are avoided by using the technologies such as batch-normal, dropout and the like, and the method has good popularization and application values.
Drawings
FIG. 1 is a flow chart of an artificial intelligence based print quality detection method of the present invention;
FIG. 2 is a topology diagram of an artificial intelligence based print quality detection system according to the present invention.
Detailed Description
The artificial intelligence based printed matter quality detection method and system of the present invention will be further described in detail with reference to the accompanying drawings and embodiments.
Examples
The printing quality detection method based on artificial intelligence trains a printing quality detection model by using a collected and preprocessed data set to obtain a mature printing quality detection model, detects the quality of the printing by using the printing quality detection model, and displays a detection result.
Wherein a large number of print result pictures are collected using a camera. And detecting the quality of the printed matter by using the printed matter quality detection model, and displaying the detection result on a display.
As shown in fig. 1, the method for detecting the printing quality based on artificial intelligence specifically includes the following steps:
s1, establishing a data set: collecting a large number of printed product result pictures, forming a data set according to the qualification of the collected pictures, and labeling.
S2, preprocessing of the data set: and preprocessing the pictures in the data set.
The data set preprocessing comprises the steps of picture noise reduction, binarization, character segmentation and normalization after data set processing.
And S3, establishing a printed matter detection model.
The presswork detection model is a deep convolutional neural network model, the characteristics of the pictures are extracted by using multilayer convolution and pooling, the pictures are classified by using three full-connection layers, and the batch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
The printed matter detection model is a deep convolutional neural network model and has the following structure: the first layer consists of two 3x3x64 convolutional layers and one 3x3/2 maximum pooling layer, the second layer consists of two 3x3x128 convolutional layers and one 3x3/2 maximum pooling layer, the third layer consists of four 3x3x256 convolutional layers and one 3x3/2 maximum pooling layer, the fourth layer consists of four 3x3x512 convolutional layers and one 3x3/2 maximum pooling layer, and the fifth layer consists of three fully-connected layer sets.
And S4, training a presswork detection model by using the data set marked in the step S1 to obtain a mature printing quality detection model.
Dividing a data set by using a ten-fold cross validation method, training a presswork detection model by using a small-batch gradient descent algorithm, calculating cost of the presswork detection model according to a cost function, and performing back propagation to modify the presswork detection model according to the cost until a mature presswork quality detection model is obtained, wherein softmax is used as the cost function. The method specifically comprises the following steps:
s401, marking the collected presswork pictures according to the condition of being qualified, and dividing a training set into a training test set and a verification set according to the ratio of 1: 5;
s402, dividing and verifying a training test set by using a ten-fold cross verification method, wherein one tenth of the training test set is used as the test set, and nine tenths are used as the training set;
s403, training a presswork detection model by using a training set test set in the training process;
s404, obtaining a prediction result by using a voting method, and calculating the cost of the printed matter detection model according to a cost function by using a verification set;
and S405, selecting a mature presswork quality detection model according to the cost function.
And S5, reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
As shown in fig. 2, the printing quality detection system based on artificial intelligence of the present invention includes a data set establishing module, a data set preprocessing module, a printed product detection model establishing module, a printed product detection model training module, and a printing quality detection result display module.
The data set establishing module is used for establishing a data set, collecting a large number of printed product result pictures, forming the data set according to whether the collected pictures are qualified or not, and marking.
The data set preprocessing module is used for preprocessing the data set: and preprocessing the pictures in the data set.
The data set preprocessing module preprocesses the data set into the processes of noise reduction, binaryzation, character segmentation and normalization of the image processed in the data set.
The printed matter detection model establishing module is used for establishing a printed matter detection model.
The presswork detection model established by the presswork detection model establishing module is a deep convolutional neural network model, the characteristics of the pictures are extracted by using multilayer convolution and pooling, the three full-connection layers are used for classification, and the batch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
The printed matter detection model is a deep convolutional neural network model and has the following structure: the first layer consists of two 3x3x64 convolutional layers and one 3x3/2 maximum pooling layer, the second layer consists of two 3x3x128 convolutional layers and one 3x3/2 maximum pooling layer, the third layer consists of four 3x3x256 convolutional layers and one 3x3/2 maximum pooling layer, the fourth layer consists of four 3x3x512 convolutional layers and one 3x3/2 maximum pooling layer, and the fifth layer consists of three fully-connected layer sets.
And the printed matter detection model training module is used for training the printed matter detection model by using the marked data set to obtain a mature printed matter quality detection model.
In the printed matter detection model training module, a data set is divided by using a ten-fold cross validation method, a small-batch gradient descent algorithm is used for training a printed matter detection model, the calculated cost of the printed matter detection model according to a cost function is obtained, the printed matter detection model is modified by back propagation according to the cost until a mature printed matter quality detection model is obtained, and softmax is used as the cost function.
The printing quality detection result display module is used for reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
The above-described embodiments are merely preferred embodiments of the present invention, and general changes and substitutions by those skilled in the art within the technical scope of the present invention are included in the protection scope of the present invention.
Claims (10)
1. A printing quality detection method based on artificial intelligence is characterized by comprising the following steps: the method comprises the steps of training a presswork detection model by using a collected and preprocessed data set to obtain a mature presswork quality detection model, detecting the quality of presswork by using the presswork quality detection model, and displaying a detection result.
2. The artificial intelligence based print quality detection method according to claim 1, wherein: the method specifically comprises the following steps:
s1, establishing a data set: collecting a large number of printed product result pictures, forming a data set according to the qualification of the collected pictures, and labeling;
s2, preprocessing of the data set: preprocessing pictures in the data set;
s3, establishing a printed matter detection model;
s4, training a presswork detection model by using the data set marked in the step S1 to obtain a mature printing quality detection model;
and S5, reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
3. The artificial intelligence based print quality detecting method according to claim 2, wherein: in step S2, the data set preprocessing includes image denoising, binarization, character segmentation and normalization after the data set processing.
4. The artificial intelligence based print quality detection method according to claim 3, wherein: the presswork detection model is a deep convolutional neural network model, the characteristics of the pictures are extracted by using multilayer convolution and pooling, the pictures are classified by using three full-connection layers, and the batch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
5. The artificial intelligence based print quality detection method according to claim 4, wherein: in step S4, the data set is divided by a cross-fold cross validation method, a print detection model is trained by a small-batch gradient descent algorithm, the obtained print detection model calculates a cost according to a cost function, and the print detection model is modified by back propagation according to the cost until a mature print quality detection model is obtained, wherein softmax is used as the cost function.
6. The artificial intelligence based print quality detecting method according to claim 5, wherein: step S4 specifically includes the following processes:
s401, marking the collected presswork pictures according to the condition of being qualified, and dividing a training set into a training test set and a verification set according to the ratio of 1: 5;
s402, dividing and verifying a training test set by using a ten-fold cross verification method, wherein one tenth of the training test set is used as the test set, and nine tenths are used as the training set;
s403, training a presswork detection model by using a training set test set in the training process;
s404, obtaining a prediction result by using a voting method, and calculating the cost of the printed matter detection model according to a cost function by using a verification set;
and S405, selecting a mature presswork quality detection model according to the cost function.
7. A printing quality detection system based on artificial intelligence is characterized in that: the system comprises a data set establishing module, a data set preprocessing module, a presswork detection model establishing module, a presswork detection model training module and a printing quality detection result display module:
the data set establishing module is used for establishing a data set, collecting a large number of printed product result pictures, forming the data set according to whether the collected pictures are qualified or not, and marking the data set;
the data set preprocessing module is used for preprocessing the data set: preprocessing pictures in the data set;
the printed matter detection model establishing module is used for establishing a printed matter detection model;
the printed matter detection model training module is used for training a printed matter detection model by using the marked data set to obtain a mature printed matter quality detection model;
the printing quality detection result display module is used for reading the picture of the printed matter with the quality to be detected, preprocessing the picture, inputting the preprocessed picture into a mature printing quality detection model to obtain a printing quality detection result, and displaying the detection result.
8. The artificial intelligence based print quality detection system of claim 7, wherein: the data set preprocessing module preprocesses the data set into the processes of noise reduction, binaryzation, character segmentation and normalization of the image processed in the data set.
9. The artificial intelligence based print quality detection system of claim 8, wherein: the presswork detection model established by the presswork detection model establishing module is a deep convolutional neural network model, the characteristics of the pictures are extracted by using multilayer convolution and pooling, the three full-connection layers are used for classification, and the batch-normal, the prelu activation function and the dropout are used for preventing overfitting, gradient disappearance and gradient explosion.
10. The artificial intelligence based print quality detection system of claim 9, wherein: in the printed matter detection model training module, a data set is divided by using a ten-fold cross validation method, a small-batch gradient descent algorithm is used for training a printed matter detection model, the calculated cost of the printed matter detection model according to a cost function is obtained, the printed matter detection model is modified by back propagation according to the cost until a mature printed matter quality detection model is obtained, and softmax is used as the cost function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911274249.0A CN111105396A (en) | 2019-12-12 | 2019-12-12 | Printed matter quality detection method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911274249.0A CN111105396A (en) | 2019-12-12 | 2019-12-12 | Printed matter quality detection method and system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111105396A true CN111105396A (en) | 2020-05-05 |
Family
ID=70423172
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911274249.0A Pending CN111105396A (en) | 2019-12-12 | 2019-12-12 | Printed matter quality detection method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111105396A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111753662A (en) * | 2020-05-25 | 2020-10-09 | 济南浪潮高新科技投资发展有限公司 | Traffic accident handling method, equipment and medium |
| CN111783901A (en) * | 2020-07-13 | 2020-10-16 | 征图新视(江苏)科技股份有限公司 | Printing quality classification method based on deep learning |
| CN111986274A (en) * | 2020-06-28 | 2020-11-24 | 济南浪潮高新科技投资发展有限公司 | Watermelon maturity state detection method, equipment and medium |
| CN113378478A (en) * | 2021-06-30 | 2021-09-10 | 安庆机床有限公司 | Automatic control system of ball making machine |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9836853B1 (en) * | 2016-09-06 | 2017-12-05 | Gopro, Inc. | Three-dimensional convolutional neural networks for video highlight detection |
| CN108918527A (en) * | 2018-05-15 | 2018-11-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of printed matter defect inspection method based on deep learning |
| CN109359693A (en) * | 2018-10-24 | 2019-02-19 | 国网上海市电力公司 | A kind of Power Quality Disturbance Classification Method |
| CN110032654A (en) * | 2019-04-19 | 2019-07-19 | 济南浪潮高新科技投资发展有限公司 | A kind of supermarket's commodity input method and system based on artificial intelligence |
| CN110084867A (en) * | 2019-04-24 | 2019-08-02 | 河北科技大学 | Arteriovenous image rebuilding method based on CNN and multi-electrode electromagnetic measurement |
| CN110287940A (en) * | 2019-07-03 | 2019-09-27 | 山东浪潮人工智能研究院有限公司 | A kind of palm grain identification method and system based on artificial intelligence |
| CN110335262A (en) * | 2019-06-28 | 2019-10-15 | 西安理工大学 | A kind of printed matter automatic defect classification method in machine vision |
-
2019
- 2019-12-12 CN CN201911274249.0A patent/CN111105396A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9836853B1 (en) * | 2016-09-06 | 2017-12-05 | Gopro, Inc. | Three-dimensional convolutional neural networks for video highlight detection |
| CN108918527A (en) * | 2018-05-15 | 2018-11-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of printed matter defect inspection method based on deep learning |
| CN109359693A (en) * | 2018-10-24 | 2019-02-19 | 国网上海市电力公司 | A kind of Power Quality Disturbance Classification Method |
| CN110032654A (en) * | 2019-04-19 | 2019-07-19 | 济南浪潮高新科技投资发展有限公司 | A kind of supermarket's commodity input method and system based on artificial intelligence |
| CN110084867A (en) * | 2019-04-24 | 2019-08-02 | 河北科技大学 | Arteriovenous image rebuilding method based on CNN and multi-electrode electromagnetic measurement |
| CN110335262A (en) * | 2019-06-28 | 2019-10-15 | 西安理工大学 | A kind of printed matter automatic defect classification method in machine vision |
| CN110287940A (en) * | 2019-07-03 | 2019-09-27 | 山东浪潮人工智能研究院有限公司 | A kind of palm grain identification method and system based on artificial intelligence |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111753662A (en) * | 2020-05-25 | 2020-10-09 | 济南浪潮高新科技投资发展有限公司 | Traffic accident handling method, equipment and medium |
| CN111986274A (en) * | 2020-06-28 | 2020-11-24 | 济南浪潮高新科技投资发展有限公司 | Watermelon maturity state detection method, equipment and medium |
| CN111783901A (en) * | 2020-07-13 | 2020-10-16 | 征图新视(江苏)科技股份有限公司 | Printing quality classification method based on deep learning |
| CN113378478A (en) * | 2021-06-30 | 2021-09-10 | 安庆机床有限公司 | Automatic control system of ball making machine |
| CN113378478B (en) * | 2021-06-30 | 2022-11-22 | 安庆机床有限公司 | Automatic control system of ball making machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111105396A (en) | Printed matter quality detection method and system based on artificial intelligence | |
| Saberironaghi et al. | Defect detection methods for industrial products using deep learning techniques: A review | |
| Ali et al. | Structural crack detection using deep convolutional neural networks | |
| Bayar et al. | Design principles of convolutional neural networks for multimedia forensics | |
| CN111709909B (en) | General printing defect detection method based on deep learning and model thereof | |
| Yuk et al. | Feature-learning-based printed circuit board inspection via speeded-up robust features and random forest | |
| CN107909564B (en) | Full convolution network image crack detection method based on deep learning | |
| CN109859164B (en) | Method for visual inspection of PCBA (printed circuit board assembly) through rapid convolutional neural network | |
| CN104992223A (en) | Intensive population estimation method based on deep learning | |
| CN110264445A (en) | The screen printing of battery quality determining method of piecemeal template matching combining form processing | |
| CN105405142A (en) | Edge defect detection method and system for glass panel | |
| CN110490842B (en) | Strip steel surface defect detection method based on deep learning | |
| Chang et al. | A deep learning-based weld defect classification method using radiographic images with a cylindrical projection | |
| Zhao et al. | Real‐time fabric defect detection based on multi‐scale convolutional neural network | |
| CN106874913A (en) | A kind of vegetable detection method | |
| CN111080615A (en) | PCB defect detection system and detection method based on convolutional neural network | |
| Lin et al. | Automated quality inspection of surface defects on touch panels | |
| Rigaud et al. | What do we expect from comic panel extraction? | |
| Shih et al. | Automatic reclaimed wafer classification using deep learning neural networks | |
| CA3035387C (en) | Digitization of industrial inspection sheets by inferring visual relations | |
| CN115170520A (en) | Metal mesh defect detection method based on structure contrast information lamination | |
| Zhang et al. | EGD-Net: Edge-guided and differential attention network for surface defect detection | |
| Wang et al. | Structure-priority image restoration through genetic algorithm optimization | |
| CN114187247A (en) | Ampoule bottle printing character defect detection method based on image registration | |
| Fan et al. | Automated steel bar counting and center localization with convolutional neural networks |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200505 |