CN113191452B - Coal ash content online detection system based on deep learning and detection method thereof - Google Patents
Coal ash content online detection system based on deep learning and detection method thereof Download PDFInfo
- Publication number
- CN113191452B CN113191452B CN202110558063.9A CN202110558063A CN113191452B CN 113191452 B CN113191452 B CN 113191452B CN 202110558063 A CN202110558063 A CN 202110558063A CN 113191452 B CN113191452 B CN 113191452B
- Authority
- CN
- China
- Prior art keywords
- model
- ash content
- coal
- image
- ash
- 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
Images
Classifications
-
- 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
- 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
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a coal ash content online detection system based on deep learning and a detection method thereof, wherein the system comprises a classification model and a regression model, and the model is deployed to embedded equipment after training is finished so as to realize distributed deployment of the coal ash content online detection function; and the real-time detection of ash content is finished by depending on a matched hardware image acquisition system. The online detection method comprises the steps of obtaining a coal microscopic image and corresponding ash content through an image acquisition device, establishing a coal microscopic image database, constructing a feature extraction network based on a deep learning method to automatically extract features of the coal ash partial image, designing a classification model and a regression model to complete final decision making, and obtaining an accurate coal ash content prediction result. Compared with other coal ash content detection methods, the method has the advantages of high detection precision, high speed and the like.
Description
Technical Field
The invention relates to the technical field of coal quality detection in mineral processing, in particular to a coal ash content online detection system based on deep learning and a detection method thereof.
Technical Field
The coal reserves in China are abundant, the rapid development of economy drives the consumption of energy sources in various industries, and the online rapid detection of the coal quality has important significance for ensuring the quality of coal products. In the coal preparation process, the conventional coal quality measurement method needs manual sampling and sample preparation, and has the problems of multiple interference factors, long assay period, lagged assay result, poor real-time performance and the like. The coal quality on-line detection technology comprises a double-energy gamma ray projection method, a natural ray method, a neutron activation method, an X-ray fluorescence method and the like, but has radioactive sources, potential radiation pollution, complex procedures such as purchase handling, acceptance inspection and the like, long period and easily influenced by foreign matters in coal, so that the coal quality on-line rapid detection method which is rapid in detection, safe, pollution-free and low in operation cost is urgently sought. With the rapid development of the artificial intelligence technology, the machine vision is possible to replace the manual labor, and the mode can generate great benefit in the industrial production process, and has important significance for building an intelligent coal preparation plant and leading the intelligent upgrading of the coal industry.
With the rapid development of the artificial intelligence technology, the use of machine vision to replace manual labor becomes possible, so in order to solve the limitation of the traditional method, an ash online ash detection method based on deep learning is provided, and the method has the advantages of high sensitivity and accurate measurement, avoids the conditions of non-uniformity of manual sampling and serious lag of test results, can timely guide and optimize the production process and the coal blending process, and ensures the product quality. The most important advantage of deep learning is that more expressive features can be automatically extracted, the end-to-end requirement in practical application is met, and the model conforms to the cognition of human brain on things, so that a high-resolution and multi-scale microscopic image database is constructed, the relation between coal ash and coal surface reflection spectrum is explored by using a deep learning training model, and a green and efficient integrated ash detection device is further developed. However, the application of the method in the technical field of coal ash measurement is not seen at present.
Disclosure of Invention
The invention aims to provide a coal ash content online detection system based on deep learning and a detection method thereof, and the method and a hardware detection system can be used for realizing online accurate measurement of coal ash content.
The purpose of the invention is realized by the following technical scheme:
in a first aspect, the invention provides a coal ash content online detection system based on deep learning, which comprises a classification model and a regression model.
Further, the classification model and the regression model train data sources: the coal sample microscopic image and the ash content corresponding to the image are collected through the high-definition industrial camera and the microscope lens matched with the auxiliary light source, and a high-resolution and multi-scale microscopic image database is established for model training.
The classification model divides the image database into classes according to the ash designated interval, and randomly divides the image data with the label information into a training set, a verification set and a test set according to a certain proportion.
The regression model slowly burns labels of images in the image database to obtain ash contents in one-to-one correspondence, and randomly divides image data with label information into a training set, a verification set and a test set according to a certain proportion.
And (3) feeding the training set and the verification set in the image data into a feature extraction network for automatic feature extraction, wherein the feature extraction network comprises but is not limited to feature extractors such as a convolutional neural network, a cyclic neural network, a self-coding network, a Transformer and the like, and obtaining the feature description of the coal sample microscopic image.
Further, the classification model in the classification model includes, but is not limited to, a support vector machine, a Softmax function, and the like, and is connected in series with the feature extractor to construct the ash online detection classification model.
Further, a regression model in the regression model includes, but is not limited to, support vector regression, linear regression, polynomial regression, etc., and is connected in series with the feature extractor to obtain the ash online detection regression model.
An ensemble learning method is adopted to fuse the prediction results of the classification model and the regression model to obtain a final ash content prediction model, and the final ash content prediction model can be deployed into embedded equipment to realize the distributed deployment of the coal ash content online detection function; the real-time detection of ash content is finished by depending on a matched hardware image acquisition system; the matched hardware image acquisition system mainly comprises an experiment table, a high-definition industrial camera, a microscope lens with adjustable magnification, a light source, an object stage with adjustable height, an embedded AI platform and a coarse-focus and fine-focus screw with the height of the camera. The fine focus-aligning screw of the camera is matched with the high depth-of-field microscope lens capable of adjusting the magnification factor to realize clear focusing.
The invention also provides a coal ash content online detection method based on deep learning, which adopts the coal ash content online detection system based on deep learning;
the method comprises the steps of completing ash content prediction of a coal microscopic image by a deep learning classification model and a regression model;
the classification model divides image data of the coal microscopic image database into data sets according to the designated interval of ash content, performs data enhancement on each category, performs feature selection by adopting a feature extraction network, and finally obtains an ash content classification model of the coal through the classification model;
the regression model is characterized in that the coal microscopic image is used as model input, ash content is used as model input, and the regression model is used to obtain the ash content regression model of the coal after the characteristic extraction network.
And the characteristic extraction network in the model comprises a deep convolutional neural network, a self-encoder and a transform characteristic extraction network, the microscopic characteristics of the coal are extracted, and the classification and regression model is integrated to construct an ash content online detection model of the coal.
The method specifically comprises the following steps of carrying out real-time online detection:
step S1: taking a plurality of grams of coal samples to be detected, uniformly sampling, pouring into a sample container, and flatly paving the sample container;
step S2: collecting microscopic images with different magnification factors and ash contents corresponding to the coal sample, and establishing a coal sample microscopic image database;
step S3: eliminating abnormal images of individual models, distortion and defocusing in a database, dividing a data set according to a specific deviation interval of ash content, and adding labels to form a training data set of a classification model; forming a training data set of a regression model according to the ash content as a label; expanding a data set in a data enhancement mode, and randomly dividing the data set into a training set, a verification set and a test set according to a certain proportion;
step S4: inputting the training set and the verification set into a feature extraction network for training to obtain feature representation of the ash content of the coal sample, and performing final decision on the obtained feature vectors by using a classification model and a regression model respectively to obtain an ash content prediction result;
step S5: and fusing ash prediction results obtained by the classification model and the regression model by adopting an ensemble learning method to obtain a final coal sample ash prediction result, and providing guidance for field industrial production.
Through the technical scheme, the invention has the following advantages:
1. according to the method, a high-resolution and multi-scale coal sample microscopic image database is established, an image pyramid is formed by microscopic images with different magnification factors, so that a feature extractor can conveniently mine more expressive features, the end-to-end requirement in practical application is realized, some problems caused by result lag are avoided, and the popularization of the method and a detection system is improved.
2. According to the method, the microscopic image characteristics can be automatically acquired through the characteristic extractor based on deep learning, the coal quality online detection model is constructed, the model precision is improved by adopting an integrated learning method, and the online rapid detection of the coal quality is more accurate and intelligent. Compared with other ash content measuring methods, the method and the detection system are rapid in detection, safe, pollution-free and low in operation cost.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is a schematic diagram of a training process of an ash on-line detection method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a hardware detection system installation of an embodiment of the present invention;
FIG. 3 is a flow chart of a ash classification model architecture in an example of the present invention;
FIG. 4 is a flow chart of a regression model of ash in an example of the present invention;
FIG. 5 is a flow chart of operation and use of the ash on-line detection system in an embodiment of the present invention.
Detailed Description
In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention is described below in detail and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are some embodiments of the present invention, and not all embodiments should not be understood as limiting the present invention.
As shown in figure 1, the main equipment of the image acquisition device matched with the on-line detection of the ash content of the coal comprises: the device comprises a light source 1, a glass plate 2, an experiment table 3, a camera height coarse focusing screw 4, a computer 5, a camera height fine focusing screw 6, an industrial camera 7, a sample container 8, a lens fixing button 9, a magnification-adjustable microscope lens 10 and an objective table height adjuster 11. The two light sources respectively irradiate the glass plate from the left and right 45 degrees, samples are uniformly spread in the sample container and placed in the center of the glass plate, the camera is used for focusing the lens roughly, and the fine focus is matched with the magnification regulator in a spiral mode to achieve clear focusing.
The image acquisition is carried out according to the following steps:
s11: the image acquisition system sets an acquisition time interval of 5 s;
s12: putting a coal sample to be detected on an objective table, shaking uniformly, putting the coal sample to be detected in the center of a glass plate, opening a camera to capture a picture, and starting an image acquisition program after detecting that the coal sample appears in a visual field range;
s13: after each image is shot, the magnification of the microscope lens is manually adjusted until the image is focused clearly, and microscopic images with the magnification of 0.3, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4 and 1.6 are respectively collected to form an image pyramid;
s14: after the collection is complete, the program is suspended for 10S, the sample is shaken up, the objective table is placed at the right center of the glass plate again, the operation of S13 is repeated, and the cumulative repetition is carried out for 5 times;
s15: and (3) slowly burning the coal sample on the objective table for 2h, testing to obtain accurate ash content of the coal sample, and storing the accurate ash content as a label to a database.
As shown in FIG. 2, the method for detecting ash content of coal on line in the present invention is divided into two parts of microscopic image processing and model training. The method comprises the following steps:
step S21: screening microscopic images in a database, and removing abnormal images of individual models, distortion and defocusing in the database;
step S22: scaling the original image to 576 multiplied by 576 pixels by a cubic spline interpolation mode, and then expanding the data set by using a data enhancement method;
step 23: dividing the augmented data set obtained in the step S22 into data sets according to the interval of +/-0.5 of ash content, and adding labels to form a training data set of a classification model;
step 24: using the images in the augmented data set obtained in the step S22 as the input of a regression model, and using the ash content as the output of the model to form a training data set of the regression model;
s25: randomly dividing a training data set into a training set, a verification set and a test set according to the ratio of 6:2: 2;
s26: inputting the training set and the verification set into a feature extraction network to obtain feature representation of the coal sample microscopic image;
s27: and respectively using and carrying out loss calculation on the feature vectors obtained by the feature extraction network, then carrying out back propagation, and carrying out repeated iterative training to obtain a final ash content prediction model.
As shown in fig. 3, the classification model performs global average pooling on the feature map obtained by the feature extractor to obtain a one-dimensional feature vector, then uses a dropout layer to discard certain neurons at random to prevent overfitting of the model, finally uses a softmax function to calculate the probability value of each ash category, uses a cross entropy loss function to calculate a training error, and repeats iteration until the loss values of the training set and the verification set do not decrease any more.
As shown in fig. 4, the regression model classification model performs global average pooling on the feature map obtained by the feature extractor to obtain a one-dimensional feature vector, performs regression prediction on ash by using a linear activation function after passing through two full-connected layers, calculates a training error by using a smoothL1 loss function, and repeatedly iterates until loss values of a training set and a verification set do not decrease.
S28: and fusing the classification result and the regression result by adopting an integrated learning mode to obtain a final ash content online detection model.
FIG. 5 is a flow chart of operation and use of the ash on-line detection system in the embodiment of the present invention, which measures ash content of a coal sample according to the following steps:
step S31: acquiring a coal sample microscopic image to be detected through a high-definition industrial camera and a high-depth-of-field lens;
step S32: transmitting the obtained image to an AI embedded platform;
step S33: integrating the results of the classification model and the regression model to obtain a final ash content prediction model;
step S34: in the initial stage of field deployment, manual assay is required to be carried out, the artificial assay is compared with a model prediction result, and if the error is large, an image needs to be transmitted to a microscopic image database;
step S35: if the number of the error images exceeds a certain number, updating the model and guiding the model into an AI embedded platform;
step S36: and stopping the manual test after the model prediction result is stable.
It is noted that those skilled in the art will recognize that embodiments of the present invention are not described in detail herein.
The embodiments and technical parameters mentioned in the description are only for the purpose of helping the reader to understand the principle of the present invention and to explain the advantages of the present invention, and do not represent the best case, and those skilled in the art can optimize the parameters of the present invention to obtain better effects. It should be understood by those skilled in the art that various equivalents and modifications may be made based on the present invention and that they are within the scope of the present invention as claimed.
Claims (1)
1. A coal ash online detection method based on deep learning adopts a coal ash online detection system based on deep learning, which comprises a classification model and a regression model, the model is deployed to an embedded device after training is completed, and real-time detection of ash is completed by depending on a matched hardware image acquisition system; the hardware image acquisition system comprises an experiment table, a high-definition industrial camera, a microscope lens capable of adjusting magnification, a light source, an object stage capable of adjusting height, an embedded AI platform and a coarse-focus and fine-focus screw for the height of the camera;
it is characterized in that the preparation method is characterized in that,
the method comprises the steps of completing ash content prediction of a coal microscopic image by a deep learning classification model and a regression model;
the classification model divides image data of the coal microscopic image database into data sets according to the designated interval of ash content, performs data enhancement on each category, performs feature selection by adopting a feature extraction network, and finally obtains an ash content classification model of the coal through the classification model;
the regression model is characterized in that the coal microscopic image is used as model input, ash content is used as model input, and the regression model is used to obtain the ash content regression model of the coal after the characteristic extraction network;
extracting a characteristic extraction network in the model, wherein the characteristic extraction network comprises a deep convolutional neural network, a self-encoder and a transform characteristic extraction network, extracting the microscopic characteristic of the coal, and integrating a classification model and a regression model to construct an ash content online detection model of the coal;
the image acquisition is carried out according to the following steps:
s11: the image acquisition system sets an acquisition time interval of 5 s;
s12: putting a coal sample to be detected on an objective table, shaking uniformly, putting the coal sample to be detected in the center of a glass plate, opening a camera to capture a picture, and starting an image acquisition program after detecting that the coal sample appears in a visual field range;
s13: after each image is shot, the magnification of the microscope lens is manually adjusted until the image is focused clearly, and microscopic images with the magnification of 0.3, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4 and 1.6 are respectively collected to form an image pyramid;
s14: after the collection is complete, the program is suspended for 10S, the sample is shaken up, the objective table is placed at the right center of the glass plate again, the operation of S13 is repeated, and the cumulative repetition is carried out for 5 times;
s15: slowly burning the coal sample on the objective table for 2h, testing to obtain accurate ash content of the coal sample, and storing the accurate ash content as a label to a database;
two parts of microscopic image processing and model training are carried out, wherein each part is carried out according to the following steps:
s21: screening microscopic images in a database, and removing abnormal images of individual models, distortion and defocusing in the database;
s22: scaling the original image to 576 multiplied by 576 pixels by a cubic spline interpolation mode, and then expanding the data set by using a data enhancement method;
23: dividing the augmented data set obtained in the step S22 into data sets according to the interval of +/-0.5 of ash content, and adding labels to form a training data set of a classification model;
24: using the images in the augmented data set obtained in the step S22 as the input of a regression model, and using the ash content as the output of the model to form a training data set of the regression model;
s25: randomly dividing a training data set into a training set, a verification set and a test set according to the ratio of 6:2: 2;
s26: inputting the training set and the verification set into a feature extraction network to obtain feature representation of the coal sample microscopic image;
s27: respectively using and carrying out loss calculation on the feature vectors obtained by the feature extraction network, then carrying out back propagation, and carrying out repeated iterative training to obtain a final ash content prediction model;
s28: fusing the classification result and the regression result by adopting an integrated learning mode to obtain a final ash online detection model;
the ash content of the coal sample is measured according to the following steps:
step S31: acquiring a coal sample microscopic image to be detected through a high-definition industrial camera and a high-depth-of-field lens;
step S32: transmitting the obtained image to an AI embedded platform;
step S33: integrating the results of the classification model and the regression model to obtain a final ash content prediction model;
step S34: in the initial stage of field deployment, manual assay is required to be carried out, the artificial assay is compared with a model prediction result, and if the error is large, an image needs to be transmitted to a microscopic image database;
step S35: if the number of the error images exceeds a certain number, updating the model and guiding the model into an AI embedded platform;
step S36: stopping the manual assay after the model prediction result is stable;
the real-time online detection is carried out according to the following steps:
step S1: taking a plurality of grams of coal samples to be detected, uniformly sampling, pouring into a sample container, and flatly paving the sample container;
step S2: collecting microscopic images with different magnification factors and ash contents corresponding to the coal sample, and establishing a coal sample microscopic image database;
step S3: eliminating abnormal images of individual models, distortion and defocusing in a database, dividing a data set according to a specific deviation interval of ash content, and adding labels to form a training data set of a classification model; forming a training data set of a regression model according to the ash content as a label; expanding a data set in a data enhancement mode, and randomly dividing the data set into a training set, a verification set and a test set according to a certain proportion;
step S4: inputting the training set and the verification set into a feature extraction network for training to obtain feature representation of the ash content of the coal sample, and performing final decision on the obtained feature vectors by using a classification model and a regression model respectively to obtain an ash content prediction result;
step S5: and fusing ash prediction results obtained by the classification model and the regression model by adopting an ensemble learning method to obtain a final coal sample ash prediction result, and providing guidance for field industrial production.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110558063.9A CN113191452B (en) | 2021-05-21 | 2021-05-21 | Coal ash content online detection system based on deep learning and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110558063.9A CN113191452B (en) | 2021-05-21 | 2021-05-21 | Coal ash content online detection system based on deep learning and detection method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113191452A CN113191452A (en) | 2021-07-30 |
| CN113191452B true CN113191452B (en) | 2022-03-01 |
Family
ID=76984783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110558063.9A Active CN113191452B (en) | 2021-05-21 | 2021-05-21 | Coal ash content online detection system based on deep learning and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113191452B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113608510A (en) * | 2021-08-04 | 2021-11-05 | 中国矿业大学 | Control method of intelligent control system of coal preparation single-machine equipment |
| CN114758196A (en) * | 2022-05-18 | 2022-07-15 | 中国矿业大学 | Clean coal ash content on-line soft measurement method in lump coal shallow slot sorting process |
| CN115035045B (en) * | 2022-05-19 | 2023-04-25 | 西安速视光电科技有限公司 | Foreign matter intelligent detection method and system based on high-resolution X-Ray image |
| CN117074643A (en) * | 2023-08-21 | 2023-11-17 | 华院计算技术(上海)股份有限公司 | Coal quality evaluation method, system, equipment and medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723738A (en) * | 2020-06-19 | 2020-09-29 | 安徽工业大学 | Coal petrography group microscopic image classification method and system based on transfer learning |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101349661B (en) * | 2008-09-17 | 2011-10-05 | 丹东东方测控技术有限公司 | Method for on-line detecting coal ashes on belt |
| CN101382556B (en) * | 2008-09-28 | 2012-07-25 | 山西潞安矿业(集团)有限责任公司 | Soft measurement method for floating fine coal ash from slurry based on data drive |
| CN104390927B (en) * | 2014-10-24 | 2017-11-03 | 中华人民共和国黄埔出入境检验检疫局 | The quick determination method of ash content in coal sample |
| CN109269951B (en) * | 2018-09-06 | 2021-12-03 | 山西智卓电气有限公司 | Image-based flotation tailing ash content, concentration and coarse particle content detection method |
| CN111340123A (en) * | 2020-02-29 | 2020-06-26 | 韶鼎人工智能科技有限公司 | Image score label prediction method based on deep convolutional neural network |
| CN111753912A (en) * | 2020-06-28 | 2020-10-09 | 中国矿业大学 | Coal slime flotation clean coal ash content prediction method based on deep learning |
-
2021
- 2021-05-21 CN CN202110558063.9A patent/CN113191452B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723738A (en) * | 2020-06-19 | 2020-09-29 | 安徽工业大学 | Coal petrography group microscopic image classification method and system based on transfer learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113191452A (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113191452B (en) | Coal ash content online detection system based on deep learning and detection method thereof | |
| JP6981533B2 (en) | Cell image analysis device, cell image analysis system, training data generation method, training model generation method, training data generation program, and training data manufacturing method | |
| Holt et al. | Progress towards an automated trainable pollen location and classifier system for use in the palynology laboratory | |
| CN110494894A (en) | Defects detection based on profile | |
| CN107004123A (en) | Iterative defect filters out technique | |
| CN110045015A (en) | A kind of concrete structure Inner Defect Testing method based on deep learning | |
| CN111753912A (en) | Coal slime flotation clean coal ash content prediction method based on deep learning | |
| CN109754392A (en) | A kind of device and method that defect image automatically analyzes | |
| KR20220095216A (en) | BBP-assisted defect detection flow for SEM images | |
| CN113128335A (en) | Method, system and application for detecting, classifying and discovering micro-body paleontological fossil image | |
| Arshed et al. | A light weight deep learning model for real world plant identification | |
| CN114092935A (en) | Textile fiber identification method based on convolutional neural network | |
| CN108765391A (en) | A kind of plate glass foreign matter image analysis methods based on deep learning | |
| CN115482227B (en) | Machine vision self-adaptive imaging environment adjusting method | |
| CN113705531B (en) | Identification method of alloy powder inclusions based on microscopic imaging | |
| Khoroshun et al. | The Development of an Application for Microparticle Counting Using a Neural Network. | |
| CN115112509A (en) | Material surface indentation measuring method based on Mask R-CNN network | |
| Pechenin et al. | Neural network model of machine parts classification by optical scanning results | |
| CN106203462A (en) | Astronomicalc optics transition source quick automatic identification method based on machine learning and system | |
| CN114897835B (en) | Image-based real-time detection system for ash content of coal products | |
| CN115082386B (en) | Injection molding flaw detection method, device and medium based on normal sample auxiliary feature extraction | |
| CN116306875B (en) | Drainage pipe network sample increment learning method based on space pre-learning and fitting | |
| TW202034218A (en) | Method for detecting fastener by using artificial intelligent module, detection process and device thereof increasing the detection efficiency and precision by virtue of artificial intelligence and optical detection | |
| CN117274819B (en) | Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas | |
| GÜZELCE et al. | Developing a real-time pattern matching algorithm using artificial neural network for a reliable quality control in industrial applications |
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 |