CN112818835A - Method for rapidly identifying and analyzing two-dimensional material by using machine learning method - Google Patents
Method for rapidly identifying and analyzing two-dimensional material by using machine learning method Download PDFInfo
- Publication number
- CN112818835A CN112818835A CN202110124919.1A CN202110124919A CN112818835A CN 112818835 A CN112818835 A CN 112818835A CN 202110124919 A CN202110124919 A CN 202110124919A CN 112818835 A CN112818835 A CN 112818835A
- Authority
- CN
- China
- Prior art keywords
- neural network
- layer
- rgb
- hsv
- analyzing
- 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
- 239000000463 material Substances 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000010801 machine learning Methods 0.000 title claims abstract description 9
- 238000013528 artificial neural network Methods 0.000 claims abstract description 37
- 230000003287 optical effect Effects 0.000 claims abstract description 24
- 210000002569 neuron Anatomy 0.000 claims abstract description 17
- 230000006870 function Effects 0.000 claims abstract description 12
- 230000004913 activation Effects 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 13
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 6
- 229910021389 graphene Inorganic materials 0.000 claims description 6
- ITRNXVSDJBHYNJ-UHFFFAOYSA-N tungsten disulfide Chemical compound S=[W]=S ITRNXVSDJBHYNJ-UHFFFAOYSA-N 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 229910052582 BN Inorganic materials 0.000 claims description 3
- PZNSFCLAULLKQX-UHFFFAOYSA-N Boron nitride Chemical compound N#B PZNSFCLAULLKQX-UHFFFAOYSA-N 0.000 claims description 3
- 238000001237 Raman spectrum Methods 0.000 claims description 3
- ROUIDRHELGULJS-UHFFFAOYSA-N bis(selanylidene)tungsten Chemical compound [Se]=[W]=[Se] ROUIDRHELGULJS-UHFFFAOYSA-N 0.000 claims description 3
- HITXEXPSQXNMAN-UHFFFAOYSA-N bis(tellanylidene)molybdenum Chemical compound [Te]=[Mo]=[Te] HITXEXPSQXNMAN-UHFFFAOYSA-N 0.000 claims description 3
- MHWZQNGIEIYAQJ-UHFFFAOYSA-N molybdenum diselenide Chemical compound [Se]=[Mo]=[Se] MHWZQNGIEIYAQJ-UHFFFAOYSA-N 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 2
- 239000010410 layer Substances 0.000 description 26
- 238000001069 Raman spectroscopy Methods 0.000 description 4
- CWQXQMHSOZUFJS-UHFFFAOYSA-N molybdenum disulfide Chemical compound S=[Mo]=S CWQXQMHSOZUFJS-UHFFFAOYSA-N 0.000 description 4
- 238000012512 characterization method Methods 0.000 description 3
- 229910052982 molybdenum disulfide Inorganic materials 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 229910016001 MoSe Inorganic materials 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004630 atomic force microscopy Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052961 molybdenite Inorganic materials 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000000985 reflectance spectrum Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 229910052814 silicon oxide Inorganic materials 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 238000004627 transmission electron microscopy Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/693—Acquisition
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a method for quickly identifying and analyzing materials by using a machine learning method, which comprises the following steps: step 1, shooting an optical microscope picture of a two-dimensional material; step 2, extracting RGB & HSV information of pixel points of the picture 50 multiplied by 50, and sending the RGB & HSV information into a built neural network for analyzing components and layers; in the step 2, the built neural network comprises an input layer, a hidden layer and an output layer which are sequentially connected from input to output, the input data is six parameters of RGB & HSV, the number of neurons in the four hidden layers is 256,1024,1024,256, each neuron in each hidden layer is a result obtained by adding the weights of all neurons in the previous layer and then performing a nonlinear activation function, and then the result is transmitted to the next layer for operation. The method for rapidly identifying and analyzing the two-dimensional materials by utilizing the machine learning method can realize accurate and rapid identification of various two-dimensional materials.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence application, and particularly relates to a method for rapidly identifying and analyzing a two-dimensional material by using a machine learning method.
Background
Since the 2004 professor geom and professor Novoselov used tape to mechanically exfoliate graphene (graphene), researchers began to develop endless studies of the unique properties of two-dimensional materials. Such materials, which can be as thin as an atomic layer, exhibit excellent optical, electrical, thermal, and mechanical properties, and they often need to be integrated to be applied to optoelectronic devices, and thus, their effective identification and characterization on a substrate is of great importance. Conventional methods for characterizing two-dimensional materials are time consuming, and the instruments for characterization are expensive, such as raman spectroscopy, atomic force microscopy, transmission electron microscopy, etc., and need improvement.
Disclosure of Invention
The invention aims to provide a method for quickly identifying and analyzing two-dimensional materials by using a machine learning method, which can realize accurate and quick identification of various two-dimensional materials.
In order to achieve the above purpose, the solution of the invention is:
a method for rapidly identifying and analyzing materials using a machine learning method, comprising the steps of:
step 1, shooting an optical microscope picture of a two-dimensional material by using a 100-time lens of an optical microscope;
step 2, extracting RGB & HSV information of pixel points of the picture 50 multiplied by 50 by using an open source matplotlib module of Python 3.7, and sending the RGB & HSV information into a built neural network for analyzing components and layer number;
in the step 2, the built neural network comprises an input layer, a hidden layer and an output layer which are sequentially connected from input to output, the input data is RGB & HSV six parameters, the hidden layer is provided with four layers, the number of neurons of the four layers is 256,1024,1024,256, each neuron of each hidden layer is a result obtained by adding the weights of all neurons of the previous layer and then passing through a nonlinear activation function, and then the result is transmitted to the next layer for operation.
In the step 2, the analysis result output by the neural network contains one of the following 8 components: tungsten disulfide, molybdenum ditelluride, tungsten diselenide, molybdenum diselenide, graphene, hexagonal boron nitride (h-BN), Bi2Sr2CaCu2O8。
In the step 2, the specific process of building the neural network is as follows:
s1, taking N groups of optical microscope pictures of 8 two-dimensional materials with the light intensity from the darkest to the brightest and the same interval change by using a 100-time lens of an optical microscope, wherein N is a natural number;
s2, determining the type and the layer number of each two-dimensional material by using Raman spectrum;
s3, extracting RGB & HSV information of pixel points of 50 x 50 of each material by using an open source matplotlib module of Python 3.7, sending the RGB & HSV information into a neural network, and training the neural network until a loss function is minimum;
and S4, testing the accuracy of the neural network by using the pictures in the test set, and adjusting the training times and the training rate until the accuracy meets the preset requirements, thereby obtaining the constructed neural network structure.
In step S1, N is set to 11.
In the step S1, when taking the optical microscope photograph, 8 kinds of two-dimensional materials are placed in N sets of light intensities varying at equal intervals from the darkest to the brightest to take the photograph.
In step S3, the open source matplotlib module of Python 3.7 is used to extract RGB & HSV information of a pixel point of 50 × 50 for each material, and the RGB & HSV information is calculated according to a formula of 4: 1, constructing a training set and a testing set, and sending information obtained from the training set into a neural network to train the neural network.
In the above step S3, the expression of the loss function L is as follows:
wherein,
representing the predicted value of the neural network, YiRepresenting the value represented by the true result.
After the scheme is adopted, the artificial neural network is combined with material science, the color space information (RGB & HSV) related to the reflectance spectrum in the pixel of the two-dimensional material optical microscope picture is used as input data, and a stable neural network which can be used for classifying different two-dimensional material optical microscope pictures is obtained through the training and learning of a computer, so that the average accuracy rate is up to 91%; in particular, even under different shooting light intensities, various two-dimensional materials can be accurately identified. The invention improves the representation efficiency of the two-dimensional material and is greatly helpful for promoting the manufacture of devices for basic research of the two-dimensional material.
Drawings
FIG. 1 is a flow chart of the present invention for training and testing ANNs;
FIG. 2 is a diagram of a neural network architecture in the present invention;
FIG. 3 is a graphical representation of the results of an experiment according to an embodiment of the present invention;
wherein (a) is an optical microscope picture of a two-dimensional material for testing, (b) is a result of identification of the ANN to fig. 3(a), (c) is an identification accuracy of the ANN to various two-dimensional materials;
FIG. 4 is a schematic diagram of the present invention for identifying two-dimensional material heterojunctions;
wherein (a) is the optical microscope picture of the heterojunction used for testing and ANN prediction result, and (b) is WS2Characteristic peak and MoS2Raman intensity integral of characteristic peaks.
Detailed Description
The technical solution and the advantages of the present invention will be described in detail with reference to the accompanying drawings.
The invention provides a method for quickly identifying and analyzing a two-dimensional material by using a machine learning method, which mainly adopts a deep learning algorithm, namely an artificial neural network and material scienceIn combination, chromatic values Red (R), Green (G), blue (B), hue (H), Saturration (S) and value (V) of pixel points of a two-dimensional material optical microscope picture are used as input neurons, nearly 1000 optical microscope pictures of different two-dimensional materials are used as data sets, 8 typical single-layer and double-layer two-dimensional materials (tungsten disulfide (WS) with different functions and prepared on a silicon substrate plated with 285 nanometer silicon oxide and prepared by using the optical microscope pictures are trained2) Molybdenum disulfide (MoS)2) Molybdenum ditelluride (2H-MoTe)2) Tungsten diselenide (2H-WSe)2) Molybdenum diselenide (2H-MoSe)2) Graphene (Graphene), hexagonal boron nitride (h-BN), Bi2Sr2CaCu2O8(BSCCO-2212) to achieve rapid and accurate identification and classification of two-dimensional materials.
The most central point of the invention lies in the construction of the artificial neural network structure, the construction process is shown in figure 1, and the method comprises the following steps:
s1, taking optical microscope pictures of 8 two-dimensional materials under N groups of light intensities by using 100 times lens of optical microscope (WITEC 300R Alpha), in this embodiment, 11 groups of light intensities are adopted, and taking optical microscope pictures of 8 two-dimensional materials under light intensities which are quantitatively changed from darkest to brightest at equal intervals;
s2, determining the type and the layer number of each two-dimensional material by using Raman spectrum;
s3, extracting RGB & HSV information of a pixel point of 50 × 50 for each material by using an open source matplotlib module of Python 3.7, where 2000 points form a training set, 500 points form a test set, the training information is sent to a neural network shown in fig. 2, the neural network is built by using a pytorch module, and includes an input layer, a hidden layer, and an output layer, which are sequentially connected from input to output, the input data are six parameters of RGB & HSV, the hidden layer is provided with four layers, the number of neurons in the four hidden layers is 256,1024,1024,256, each neuron in each hidden layer is the sum of the weights of all neurons in the previous layer, and then a result is obtained by a nonlinear activation function (ReLu function), and the formula is as follows:
f(x)＝max(0，Aj)
wherein f (x) is the value of a neuron;
Ajthe expression is the sum of the weights of all neurons in the layer before the neuron, and is as follows:
wherein x isiThe ith neuron of the upper layer, wjiThe weight of the ith neuron is represented, and the initial value of the weight can be any value.
After several layers of operation, an output value can be finally obtained, after a neural network is trained in a large number, a predicted value is consistent with a true value by adjusting weight parameters in the network by self, and the predicted value and the true value can be observed through a loss function (mean square error), wherein the formula is as follows:
wherein,
representing the predicted value of the neural network, YiRepresenting the value represented by the true result. When the loss function reaches a minimum, the network is proven to be stable and usable.
And S4, testing the accuracy of the network by using the test set, and if the accuracy is not high, repeating the experiment by adjusting the training times and the training speed of the network to improve the accuracy until the accuracy meets the preset requirement, thereby obtaining the neural network structure adopted by the invention.
The specific process of identifying and classifying the unknown two-dimensional material comprises the following steps:
(1) taking an optical microscope picture of the two-dimensional material with a 100-fold lens of an optical microscope (WITEC 300R Alpha);
(2) RGB & HSV information of pixel points of the picture 50 multiplied by 50 is extracted by an open source matplotlib module of Python 3.7, and is sent into the built neural network for analyzing components and layer numbers.
Example (b): an optical microscope picture is taken for several two-dimensional materials for testing, for example, as shown in fig. 3(a), the picture is identified by using a neural network with stable training, and as a result, as shown in fig. 3(b), it can be seen that each pixel point of the optical picture is accurately classified, and the identification time is very short, which proves that the method is accurate and rapid for representing the two-dimensional materials. Meanwhile, fig. 3(c) shows the classification accuracy of each material, the diagonal elements are the accuracy of each material being accurately identified, the off-diagonal elements are the probability of the material being confused with other materials, and it can be seen that the accuracy of most materials is over 85%.
In addition, we can also use the existing data set to identify the composition information of the two-dimensional material heterojunction, as shown in fig. 4, fig. 4(a) includes two-dimensional material heterojunction molybdenum disulfide/tungsten disulfide (MoS)2/WS2) Fig. 4(b) is a real composition analysis of the heterojunction obtained by raman spectroscopy, and it can be seen that the analysis of the ANN and the raman spectroscopy are substantially identical. The method for rapidly identifying the heterojunction components has great significance for the application of the two-dimensional material heterojunction, shortens the time required by sample characterization, and can rapidly screen out the heterostructure required by device manufacturing.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the protection scope of the present invention.
Claims (7)
1. A method for rapidly identifying and analyzing materials using a machine learning method, comprising the steps of:
step 1, taking an optical microscope picture of a two-dimensional material by using a 100-time lens of an optical microscope;
step 2, extracting RGB & HSV information of pixel points of a picture 50 multiplied by 50 by using an open source matplotlib module of Python 3.7, and sending the RGB & HSV information into a built neural network for analyzing components and layer number;
in the step 2, the built neural network comprises an input layer, a hidden layer and an output layer which are sequentially connected from input to output, the input data is RGB & HSV six parameters, the hidden layer is provided with four layers, the number of neurons of the four layers is 256,1024,1024,256, each neuron of each hidden layer is a result obtained by adding the weights of all neurons of the previous layer and then passing through a nonlinear activation function, and then the result is transmitted to the next layer for operation.
2. The method of claim 1, wherein: in the step 2, the analysis result output by the neural network comprises one of the following 8 components: tungsten disulfide, molybdenum ditelluride, tungsten diselenide, molybdenum diselenide, graphene, hexagonal boron nitride (h-BN), Bi2Sr2CaCu2O8。
3. The method of claim 1, wherein: in the step 2, the specific process of building the neural network is as follows:
s1, taking N groups of light microscope photos of 8 two-dimensional materials under the light intensity from the darkest to the brightest with the same interval change by using a 100-time lens of the light microscope, wherein N is a natural number;
s2, determining the type and the layer number of each two-dimensional material by using Raman spectrum;
s3, extracting RGB & HSV information of pixel points of 50 x 50 of each material by using an open source matplotlib module of Python 3.7, sending the RGB & HSV information into a neural network, and training the neural network until a loss function is minimum;
and S4, testing the accuracy of the neural network by using the photos in the test set, and adjusting the training times and the training rate until the accuracy meets the preset requirements, thereby obtaining the constructed neural network structure.
4. The method of claim 3, wherein: in step S1, N takes the value of 11.
5. The method of claim 3, wherein: in step S1, when taking the optical microscope photograph, 8 two-dimensional materials are placed in N sets of light intensities varying at equal intervals from the darkest to the brightest for photographing.
6. The method of claim 3, wherein: in step S3, the open source matplotlib module of Python 3.7 is used to extract RGB & HSV information of a pixel point of 50 × 50 for each material, and the RGB & HSV information is calculated according to a formula of 4: 1, constructing a training set and a testing set, and sending information obtained from the training set into a neural network to train the neural network.
7. The method of claim 3, wherein: in step S3, the expression of the loss function L is as follows:
wherein,
representing the predicted value of the neural network, YiRepresenting the value represented by the true result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110124919.1A CN112818835A (en) | 2021-01-29 | 2021-01-29 | Method for rapidly identifying and analyzing two-dimensional material by using machine learning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110124919.1A CN112818835A (en) | 2021-01-29 | 2021-01-29 | Method for rapidly identifying and analyzing two-dimensional material by using machine learning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112818835A true CN112818835A (en) | 2021-05-18 |
Family
ID=75860157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110124919.1A Pending CN112818835A (en) | 2021-01-29 | 2021-01-29 | Method for rapidly identifying and analyzing two-dimensional material by using machine learning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112818835A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906540A (en) * | 2021-02-08 | 2021-06-04 | 北京科技大学 | Identification method of triangular molybdenum disulfide based on growth on substrate |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110299720A1 (en) * | 2010-03-18 | 2011-12-08 | The Regents Of The University Of California | Systems and methods for material layer identification through image processing |
| US20180107896A1 (en) * | 2016-10-16 | 2018-04-19 | Samsung Electronics Co., Ltd. | Method and apparatus to perform material recognition and training for material recognition |
| CN110288661A (en) * | 2019-06-19 | 2019-09-27 | 深圳市睿智龙电子科技有限公司 | Location regulation method, device, computer equipment and the storage medium of operating lamp |
| US10509466B1 (en) * | 2011-05-11 | 2019-12-17 | Snap Inc. | Headwear with computer and optical element for use therewith and systems utilizing same |
| CN110619355A (en) * | 2019-08-28 | 2019-12-27 | 武汉科技大学 | Automatic steel material microstructure identification method based on deep learning |
| CN111192471A (en) * | 2020-02-25 | 2020-05-22 | 佛山科学技术学院 | Intelligent public transportation system based on image processing |
| CN111188021A (en) * | 2020-01-19 | 2020-05-22 | 南京大学 | Pretreatment method of graphene growth substrate |
| CN112069900A (en) * | 2020-08-06 | 2020-12-11 | 常熟理工学院 | Bill character recognition method and system based on convolutional neural network |
| WO2020248513A1 (en) * | 2019-06-11 | 2020-12-17 | 苏州玖物互通智能科技有限公司 | Ocr method for comprehensive performance test |
-
2021
- 2021-01-29 CN CN202110124919.1A patent/CN112818835A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110299720A1 (en) * | 2010-03-18 | 2011-12-08 | The Regents Of The University Of California | Systems and methods for material layer identification through image processing |
| US10509466B1 (en) * | 2011-05-11 | 2019-12-17 | Snap Inc. | Headwear with computer and optical element for use therewith and systems utilizing same |
| US20180107896A1 (en) * | 2016-10-16 | 2018-04-19 | Samsung Electronics Co., Ltd. | Method and apparatus to perform material recognition and training for material recognition |
| WO2020248513A1 (en) * | 2019-06-11 | 2020-12-17 | 苏州玖物互通智能科技有限公司 | Ocr method for comprehensive performance test |
| CN110288661A (en) * | 2019-06-19 | 2019-09-27 | 深圳市睿智龙电子科技有限公司 | Location regulation method, device, computer equipment and the storage medium of operating lamp |
| CN110619355A (en) * | 2019-08-28 | 2019-12-27 | 武汉科技大学 | Automatic steel material microstructure identification method based on deep learning |
| CN111188021A (en) * | 2020-01-19 | 2020-05-22 | 南京大学 | Pretreatment method of graphene growth substrate |
| CN111192471A (en) * | 2020-02-25 | 2020-05-22 | 佛山科学技术学院 | Intelligent public transportation system based on image processing |
| CN112069900A (en) * | 2020-08-06 | 2020-12-11 | 常熟理工学院 | Bill character recognition method and system based on convolutional neural network |
Non-Patent Citations (4)
| Title |
|---|
| BINGNAN HAN ET AL., 《ADV. MATER.》DEEP-LEARNING-ENABLED FAST OPTICAL IDENTIFICATION AND CHARACTERIZATION OF 2D MATERIALS, vol. 32, 9 June 2020 (2020-06-09), pages 1 - 10 * |
| LI ZHU ET AL., 《ACS NANO》 ARTIFICIAL NEURON NETWORKS ENABLED IDENTIFICATION AND CHARACTERIZATIONS OF 2D MATERIALS AND VAN DER WAALS HETEROSTRUCTURES, vol. 16, 18 January 2022 (2022-01-18), pages 2721 * |
| LUKAS MENNEL ET AL., 《NATURE》 ULTRAFAST MACHINE VISION WITH 2D MATERIAL NEURAL NETWORK IMAGE SENSORS, vol. 579, 4 March 2020 (2020-03-04), pages 1 - 16 * |
| 孙禧亭 等, 《分析测试学报》 "动态"近红外光谱结合深度学习图像识别和迁移学习的模式识别方法研究, no. 10, 10 November 2020 (2020-11-10), pages 1247 - 1253 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906540A (en) * | 2021-02-08 | 2021-06-04 | 北京科技大学 | Identification method of triangular molybdenum disulfide based on growth on substrate |
| CN112906540B (en) * | 2021-02-08 | 2022-02-01 | 北京科技大学 | Identification method of triangular molybdenum disulfide based on growth on substrate |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Karimi et al. | Generalized and mechanistic PV module performance prediction from computer vision and machine learning on electroluminescence images | |
| Audebert et al. | Generative adversarial networks for realistic synthesis of hyperspectral samples | |
| US8290275B2 (en) | Effective pigment identification method, identification system, identification program, and recording medium therefor | |
| US11555996B2 (en) | Method and system for analyzing 2D material thin film | |
| CN107563999A (en) | A kind of chip defect recognition methods based on convolutional neural networks | |
| CN111914696A (en) | Hyperspectral remote sensing image classification method based on transfer learning | |
| Pan et al. | Siamese network based metric learning for SAR target classification | |
| CN112818835A (en) | Method for rapidly identifying and analyzing two-dimensional material by using machine learning method | |
| Demant et al. | Visualizing material quality and similarity of mc-Si wafers learned by convolutional regression networks | |
| CN107545281B (en) | Single harmful gas infrared image classification and identification method based on deep learning | |
| CN116523905A (en) | Defect detection method and system for lamp beads | |
| Orillo et al. | Rice plant nitrogen level assessment through image processing using artificial neural network | |
| WO2024082796A1 (en) | Spectral cross-domain transfer super-resolution reconstruction method for multi-domain image | |
| Chen et al. | An improved GMM-based algorithm with optimal multi-color subspaces for color difference classification of solar cells | |
| Aggarwal et al. | Optimized sequential model for plant recognition in keras | |
| Chen et al. | Convolutional neural network-based pure paint pigment identification using hyperspectral images | |
| CN103218626A (en) | Color separation method of solar cell chip | |
| CN115829042B (en) | Structural performance evaluation method and system for quantum dot diffusion plate | |
| CN108052981B (en) | Image classification method based on nonsubsampled Contourlet transformation and convolutional neural network | |
| Cheng et al. | A novel approach to color normalization using neural network | |
| CN110569889A (en) | Convolutional neural network image classification method based on L2 normalization | |
| CN109934281A (en) | A kind of unsupervised training method of two sorter networks | |
| CN111639465B (en) | AlGaAs photocathode structure design method based on deep learning | |
| CN112990333A (en) | Deep learning-based weather multi-classification identification method | |
| Durai Arun et al. | An image based microtiter plate reader system for 96-well format fluorescence assays |
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 |