CN111160490A - Deep learning dangerous rock deformation prediction method and device based on multiple time sequences - Google Patents
Deep learning dangerous rock deformation prediction method and device based on multiple time sequences Download PDFInfo
- Publication number
- CN111160490A CN111160490A CN202010114086.6A CN202010114086A CN111160490A CN 111160490 A CN111160490 A CN 111160490A CN 202010114086 A CN202010114086 A CN 202010114086A CN 111160490 A CN111160490 A CN 111160490A
- Authority
- CN
- China
- Prior art keywords
- dangerous rock
- data
- deformation
- learning
- unit
- 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
- 239000011435 rock Substances 0.000 title claims abstract description 118
- 238000013135 deep learning Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000005516 engineering process Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012806 monitoring device Methods 0.000 claims abstract description 9
- 238000012800 visualization Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 238000005096 rolling process Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 2
- 230000002265 prevention Effects 0.000 abstract description 2
- 230000008901 benefit Effects 0.000 description 6
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009440 infrastructure construction Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000001314 paroxysmal effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for predicting deep learning dangerous rock deformation based on multiple time sequences, which are mainly characterized in that: the method comprises the steps of obtaining dangerous rock images according to a monitoring device of dangerous rock masses, extracting dangerous rock image features by using a Caffe visualization tool, training an AlexNet model built by dangerous rock images and feature labels, identifying distributed features of the dangerous rocks, building a dangerous rock deformation multi-time sequence by using collected dangerous rock image feature data, building a plurality of data samples, performing fitting learning on the learning samples by using a deep learning technology, finally performing optimized comparison on prediction data of the time sequences by using a screening program compiled by Matlab software, and outputting a prediction result of the time sequence with the minimum error. The apparatus of the present invention is also described in detail. The embodiment of the invention can reflect the accuracy and flexibility of dangerous rock prediction, and can provide a basis for dangerous rock instability prediction and dangerous rock collapse prediction and prevention.
Description
Technical Field
The invention discloses a method and a device for predicting deformation of a deep learning dangerous rock based on multiple time sequences, and relates to the deformation prediction problem of the dangerous rock under monitoring, in addition, the invention also relates to the related fields of performing fitting learning calculation on big data by using a network learning prediction model established by a deep learning technology, and the like.
Background
The dangerous rock is a rock mass which can be unstably divided by various structural surfaces on a steep slope or a steep cliff, and although collapse does not occur, the dangerous rock actually has main conditions for collapse, so that the collapse phenomenon can occur in a short time in the future. China is a mountainous country, particularly in the southwest mountainous area, and dangerous rock deformation and collapse occur at the same time, which brings great harm to people and objects around the occurrence area and often causes serious disasters. In recent years, with further investment in infrastructure construction in China, the number of railways and highways in mountainous areas is increasing year by year, and in addition, the traffic volume on each traffic line is rapidly increasing, and in order to improve the safety of roads, especially the problem of collapse of both sides of roads caused by deformation of dangerous rocks, prediction of deformation of dangerous rocks becomes more important.
In fact, in nature, the occurrence of a dangerous rock collapse is a slow process. The rock exposed on the earth surface gradually cracks under various weathering and erosion actions, finally falls down and rolls from the initial position and falls on the toe. Because the occurrence of dangerous rock collapse is greatly uncertain, and the complexity and the paroxysmal nature of collapse movement can cause the consequence of difficult estimation after the occurrence, in order to avoid the occurrence of the collapse phenomenon and reduce unnecessary personnel and economic losses caused by the dangerous rock collapse, the method is extremely important for predicting and quantitatively predicting the deformation trend of the dangerous rock according to the deformation rule of the dangerous rock.
With the rapid development of computer technology, the research and application of the artificial intelligence deep learning algorithm is also rapidly developed. In recent years, the deep learning technique has been widely applied to the fields of image recognition, language recognition, video analysis, text analysis, and big data analysis, with great success. The deep learning is essentially a process for describing features, so the outstanding advantages of image processing and big data analysis are realized by using the principle of deep learning, the dangerous rock deformation prediction is feasible by using the deep learning technology, and the technology can be developed into a novel prediction method.
A dangerous rock deformation prediction method based on a multi-time-series deep learning technology is characterized in that deformation images of dangerous rock bodies at various time stages are analyzed and processed to obtain sample data, a prediction model is built, and therefore dangerous rock deformation prediction is automatically and rapidly conducted. Therefore, the dangerous rock deformation prediction method and device based on the deep learning technology of the multiple time series can accurately and quickly obtain the prediction result, and the benefit brought to the actual engineering is huge.
Disclosure of Invention
The embodiment of the invention adopts a method and a device for predicting the deformation of dangerous rocks by deep learning based on multiple time sequences, the method and the device can be used for rapidly and accurately predicting the deformation of the dangerous rocks and establishing an analysis processing model of self-adaptive deep learning meeting engineering requirements, and the method and the device for predicting the deformation of the dangerous rocks by deep learning based on the multiple time sequences provided by the invention comprise the following steps:
1. the prediction method principle comprises the following steps: step one, obtaining images according to a monitoring device of a dangerous rock mass, such as: the method can acquire a deformation image of the dangerous rock once by taking a month as a unit, and can acquire a plurality of time series images by analogy; extracting image characteristics of the dangerous rock by using a Caffe visualization tool; training an AlexNet model established by the dangerous rock deformation image and the characteristic label, and identifying distributed characteristics such as joints, cracks, crushing degree, roughness and section form of the dangerous rock;
fourthly, constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of the dangerous rock image, and establishing a plurality of data samples; step five, a learning model is established by utilizing a deep learning technology to perform fitting learning on the learning sample, so that an accurate prediction effect is achieved; step six, optimizing and comparing the prediction data of the plurality of time sequences by using a screening program compiled by Matlab software, and outputting the prediction data of the time sequence with the minimum error; and seventhly, in order to quantitatively predict the deformation data of the current dangerous rock mass, the data change of each influence factor can be intelligently analyzed according to a learning model established by a deep learning technology, and finally, the deformation result can be predicted according to the provided main influence factors. The device mainly comprises: the device comprises a data collecting unit, a processing unit, a sample establishing unit, a processing unit, a training unit, a learning unit, an optimizing unit, a result unit and an analyzing unit.
2. The prediction method comprises the following steps: the method comprises the steps of carrying out Caffe visualization on dangerous rock deformation time sequence images collected by a dangerous rock monitoring device to extract image characteristics of the images, training an AlexNet model built by the dangerous rock deformation images and characteristic labels, carrying out identification analysis, building a dangerous rock deformation multi-time sequence by an image characteristic data set, and building a plurality of samples to be output.
3. The prediction method comprises the following steps: the purpose of constructing the time sequence is to establish a nonlinear mapping relation between an image characteristic data value at n moments and image characteristic data values at p moments, in practical application, a first time sequence can acquire dangerous rock image characteristics once by taking half a month as a unit and acquire p times, the first 1-p measured values are used as output sequences, and the p +1 th measured value is used as an output value, so that 1 st sample data is formed; then, replacing the oldest value of the original sequence with the (p + 1) th measured value, and taking the (p + 2) th measured value as an output value to further form 2 nd sample data; by analogy, rolling to construct a plurality of dangerous rock deformation sequence sample data; the second time sequence can obtain the dangerous rock image characteristics once in a month unit, a plurality of dangerous rock deformation sequence sample data of the second time sequence can be obtained by referring to the first time sequence, a plurality of time sequence sample data can be constructed accordingly, deep learning is carried out on the dangerous rock deformation sequence sample data of the plurality of time sequences to establish a model, after the learning sample is learned by adopting a deep learning technology learning model, the learning sample is not constructed in a rolling mode, and the future time dangerous rock deformation result is directly forecasted.
4. The prediction method comprises the following steps: the method comprises the steps of constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of dangerous rock images, establishing a plurality of data samples, utilizing a deep learning technology to establish a learning model to perform fitting learning on the learning samples, achieving an accurate prediction effect, finally utilizing a screening program compiled by Matlab software to perform optimization comparison on prediction data of the plurality of time sequences, outputting prediction data performed by the time sequence with the minimum error, in order to quantitatively predict deformation data of the current dangerous rock, intelligently analyzing data changes of various influence factors according to the learning model established by the deep learning technology, and finally predicting a to-be-known deformation result according to the provided main influence factors.
5. The prediction apparatus further includes: a data collecting unit: acquiring an image according to a monitoring device of the dangerous rock mass; a processing unit: extracting image features of the dangerous rock by using a Caffe visualization tool, training an AlexNet model established by using a dangerous rock deformation image and a feature label, and identifying distributed features such as joints, cracks, crushing degree, roughness, section morphology and the like of the dangerous rock; establishing a sample unit: constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of the dangerous rock image, and establishing a plurality of data samples; a training unit: training the established data samples by utilizing a deep learning technology; a learning unit: a model is established through a deep learning technology to carry out fitting learning on the learning sample, so that the effect of accurate prediction is achieved; an optimization unit: finally, optimizing and comparing the predicted data of the time sequences by using a screening program compiled by Matlab software; a result unit: outputting prediction data performed by the time series with the minimum error; an analysis unit: in order to quantitatively predict the deformation data of the current dangerous rock mass, the data change of each influence factor can be intelligently analyzed according to a learning model established by a deep learning technology, and finally, the deformation result to be known can be predicted according to the provided main influence factors.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a flow chart of an embodiment of the present invention;
FIG. 3 is a flow chart of various stages of the present invention;
FIG. 4 is a schematic diagram of the operation between the units of the present invention;
FIG. 5 is an overall workflow diagram of the present invention;
Detailed Description
The embodiment of the invention provides a multi-time-sequence-based dangerous rock deformation deep learning prediction method and device, and the embodiment of the invention carries out deep learning based on the dangerous rock image characteristics of multiple time sequences, so that a model is established by the big data image characteristics to predict dangerous rock deformation, and the method has the advantages of high precision and strong applicability in predicting dangerous rock deformation. In order to make the technical problems, technical solutions and advantages solved by the present invention more clear, the following statements further describe the present invention in detail, wherein the related contents of the sample data of the multi-time series of the deformation of the dangerous rock are described in detail in the summary of the invention, and are omitted in this embodiment.
The method for predicting the deformation of the deep learning dangerous rock based on the multiple time sequences comprises the following steps: the method adopted according to the invention comprises the following steps: acquiring a required dangerous rock image through a monitoring device of a dangerous rock body, extracting image characteristics of the dangerous rock by using a Caffe visualization tool, then training an AlexNet model established by using a dangerous rock deformation image and a characteristic label, further identifying distributed characteristics of the dangerous rock, such as joints, cracks, crushing degrees, roughness degrees, section forms and the like, constructing a dangerous rock deformation multi-time sequence from the collected data of various distributed characteristics of the dangerous rock image, establishing a plurality of data samples, establishing a learning model by using a deep learning technology to carry out fitting learning on the learning sample to achieve the effect of accurate prediction, finally carrying out optimized comparison on the prediction data of the plurality of time sequences by using a screening program compiled by Matlab software, outputting the prediction data carried out by the time sequence with the minimum error, and in addition, in order to quantitatively predict the deformation data of the current dangerous rock body, carrying out intelligent analysis on the data change of each influence factor according to the learning model established by the deep learning technology, finally, the deformation result to be known can be predicted according to the provided main influence factors.
The dangerous rock deformation prediction device based on the multi-time series deep learning technology in the present invention is described in detail below, wherein: a data collecting unit: acquiring an image according to a monitoring device of the dangerous rock mass; a processing unit: extracting image features of the dangerous rock by using a Caffe visualization tool, training an AlexNet model established by using a dangerous rock deformation image and a feature label, and identifying distributed features such as joints, cracks, crushing degree, roughness, section morphology and the like of the dangerous rock; establishing a sample unit: constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of the dangerous rock image, and establishing a plurality of data samples; a training unit: training the established data samples by utilizing a deep learning technology; a learning unit: a model is established through a deep learning technology to carry out fitting learning on the learning sample, so that the effect of accurate prediction is achieved; an optimization unit: finally, optimizing and comparing the predicted data of the time sequences by using a screening program compiled by Matlab software; a result unit: outputting prediction data performed by the time series with the minimum error; an analysis unit: in order to quantitatively predict the deformation data of the current dangerous rock mass, the data change of each influence factor can be intelligently analyzed according to a learning model established by a deep learning technology, and finally, the deformation result to be known can be predicted according to the provided main influence factors.
The present invention provides an apparatus, which is merely schematic, the unit division is a division of logic functions, and the above embodiments are merely illustrative of the method and apparatus of the present invention. In addition, the deep learning dangerous rock deformation prediction method and device based on the multiple time sequences have the following advantages: the method and the device for predicting the deformation of the deep learning dangerous rock based on the multiple time sequences have the advantages of being conservative in result, excellent in fitting performance, high in accuracy, high in precision, strong in applicability and the like, basically achieve the purposes of being systematic, comprehensive and practical, and can effectively guide the work of dangerous rock investigation evaluation, monitoring, early warning, prevention and control and the like.
Claims (6)
1. A deep learning dangerous rock deformation prediction method and device based on multiple time series are characterized in that the method comprises the following steps:
step one, obtaining images according to a monitoring device of a dangerous rock mass, such as: the method can acquire a deformation image of the dangerous rock once by taking a half month as a unit, can acquire a deformation image of the dangerous rock once by taking a month as a unit, and can acquire a plurality of time series images by analogy;
extracting image characteristics of the dangerous rock by using a Caffe visualization tool;
training an AlexNet model established by the dangerous rock deformation image and the characteristic label, and identifying distributed characteristics such as joints, cracks, crushing degree, roughness and section form of the dangerous rock;
fourthly, constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of the dangerous rock image, and establishing a plurality of data samples;
step five, a learning model is established by utilizing a deep learning technology to perform fitting learning on the learning sample, so that an accurate prediction effect is achieved;
step six, optimizing and comparing the prediction data of the plurality of time sequences by using a screening program compiled by Matlab software, and outputting the prediction data of the time sequence with the minimum error;
in order to quantitatively predict the deformation data of the current dangerous rock mass, the data change of each influence factor can be intelligently analyzed according to a learning model established by a deep learning technology, and finally, the deformation result to be known can be predicted according to the provided main influence factors;
the device mainly comprises: the device comprises a data collecting unit, a processing unit, a sample establishing unit, a processing unit, a training unit, a learning unit, an optimizing unit, a result unit and an analyzing unit.
2. The prediction method and apparatus according to claim 1, wherein the prediction method is: the method comprises the steps of carrying out Caffe visualization on dangerous rock deformation time sequence images collected by a dangerous rock monitoring device to extract image characteristics of the images, training an AlexNet model built by the dangerous rock deformation images and characteristic labels, carrying out identification analysis, building a dangerous rock deformation multi-time sequence by an image characteristic data set, and building a plurality of samples to be output.
3. The prediction method and apparatus according to claim 1, wherein the prediction method is: the purpose of constructing the time sequence is to establish a nonlinear mapping relation between an image characteristic data value at n moments and image characteristic data values at p moments, in practical application, a first time sequence can acquire dangerous rock image characteristics once by taking half a month as a unit and acquire p times, the first 1-p measured values are used as output sequences, and the p +1 th measured value is used as an output value, so that 1 st sample data is formed; then, replacing the oldest value of the original sequence with the (p + 1) th measured value, and taking the (p + 2) th measured value as an output value to further form 2 nd sample data; by analogy, rolling to construct a plurality of dangerous rock deformation sequence sample data; the second time sequence can obtain the dangerous rock image characteristics once in a month unit, and a plurality of dangerous rock deformation sequence sample data of the second time sequence can be obtained by referring to the first time sequence, and a plurality of time sequence sample data can be constructed in the same way.
4. The prediction method and apparatus according to claim 1, wherein the prediction method is: and performing deep learning on the dangerous rock deformation data samples of the multiple time sequences to establish a model, and directly forecasting the dangerous rock deformation result in the future time without constructing the learning sample in a rolling manner after learning the learning sample by adopting the deep learning technology learning model.
5. The prediction method and apparatus according to claim 1, wherein the prediction method further comprises: the method comprises the steps of constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of dangerous rock images, establishing a plurality of data samples, utilizing a deep learning technology to establish a learning model to perform fitting learning on the learning samples, achieving an accurate prediction effect, finally utilizing a screening program compiled by Matlab software to perform optimization comparison on prediction data of the plurality of time sequences, outputting prediction data performed by the time sequence with the minimum error, carrying out intelligent analysis on each influence factor according to the learning model established by the deep learning technology in order to quantitatively predict deformation data of the current dangerous rock mass, and finally predicting a to-be-known deformation result according to the provided main influence factors.
6. The apparatus of claim 5, wherein the predicting means further comprises: a data collecting unit: acquiring an image according to a monitoring device of the dangerous rock mass; a processing unit: extracting image features of the dangerous rock by using a Caffe visualization tool, training an AlexNet model established by using a dangerous rock deformation image and a feature label, and identifying distributed features such as joints, cracks, crushing degree, roughness, section morphology and the like of the dangerous rock; establishing a sample unit: constructing a dangerous rock deformation multi-time sequence from various collected characteristic data of the dangerous rock image, and establishing a plurality of data samples; a training unit: training the established data samples by utilizing a deep learning technology; a learning unit: a model is established through a deep learning technology to carry out fitting learning on the learning sample, so that the effect of accurate prediction is achieved; an optimization unit: finally, optimizing and comparing the predicted data of the time sequences by using a screening program compiled by Matlab software; a result unit: outputting prediction data performed by the time series with the minimum error; an analysis unit: in order to quantitatively predict the deformation data of the current dangerous rock mass, the data change of each influence factor can be intelligently analyzed according to a learning model established by a deep learning technology, and finally, the deformation result to be known can be predicted according to the provided main influence factors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010114086.6A CN111160490A (en) | 2020-02-24 | 2020-02-24 | Deep learning dangerous rock deformation prediction method and device based on multiple time sequences |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010114086.6A CN111160490A (en) | 2020-02-24 | 2020-02-24 | Deep learning dangerous rock deformation prediction method and device based on multiple time sequences |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111160490A true CN111160490A (en) | 2020-05-15 |
Family
ID=70566362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010114086.6A Pending CN111160490A (en) | 2020-02-24 | 2020-02-24 | Deep learning dangerous rock deformation prediction method and device based on multiple time sequences |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111160490A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111965255A (en) * | 2020-08-14 | 2020-11-20 | 广西大学 | Pressure shear slip type karst dangerous rock instability early warning multi-precursor sound method and device |
| CN112182888A (en) * | 2020-09-29 | 2021-01-05 | 广西大学 | Method and device for identifying mechanical parameters of main control structural plane of small-sized sliding dangerous rock mass |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110472729A (en) * | 2019-08-02 | 2019-11-19 | 中南大学 | Rock burst trend prediction method based on comprehensive CNN-LSTM |
-
2020
- 2020-02-24 CN CN202010114086.6A patent/CN111160490A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110472729A (en) * | 2019-08-02 | 2019-11-19 | 中南大学 | Rock burst trend prediction method based on comprehensive CNN-LSTM |
Non-Patent Citations (2)
| Title |
|---|
| 柳厚祥: "基于深度学习技术的公路隧道围岩分级方法", 岩土工程学报, vol. 40, no. 10, pages 1180 - 1181 * |
| 苏国韶: "隧道围岩变形预报的高斯过程机器学习模型", 桂林理工大学学报, vol. 30, no. 04, pages 552 - 554 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111965255A (en) * | 2020-08-14 | 2020-11-20 | 广西大学 | Pressure shear slip type karst dangerous rock instability early warning multi-precursor sound method and device |
| CN112182888A (en) * | 2020-09-29 | 2021-01-05 | 广西大学 | Method and device for identifying mechanical parameters of main control structural plane of small-sized sliding dangerous rock mass |
| CN112182888B (en) * | 2020-09-29 | 2022-11-01 | 广西大学 | Method and device for identifying mechanical parameters of main control structural plane of small-sized sliding dangerous rock mass |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110222701B (en) | Automatic bridge disease identification method | |
| CN114693615A (en) | Deep learning concrete bridge crack real-time detection method based on domain adaptation | |
| CN113096088B (en) | Concrete structure detection method based on deep learning | |
| CN107480611A (en) | A kind of crack identification method based on deep learning convolutional neural networks | |
| CN110569843B (en) | Intelligent detection and identification method for mine target | |
| CN101697229B (en) | Method for extracting region of interest of medical image | |
| Enshaei et al. | Automated detection of textured-surface defects using UNet-based semantic segmentation network | |
| CN111160490A (en) | Deep learning dangerous rock deformation prediction method and device based on multiple time sequences | |
| CN116359285A (en) | Oil gas concentration intelligent detection system and method based on big data | |
| CN106855865B (en) | Water conservancy and hydropower big data architecture construction method | |
| Altabey et al. | A deep learning-based approach for pipeline cracks monitoring | |
| CN114241423A (en) | Intelligent detection method and system for river floaters | |
| CN104504265A (en) | Method for safety evaluation of monitoring information of in-service bridge | |
| CN112906593A (en) | Sluice image identification method based on fast RCNN | |
| CN116663126A (en) | Bridge temperature effect prediction method based on channel attention BiLSTM model | |
| Baskaran et al. | Steel frame structure defect detection using image processing and artificial intelligence | |
| CN115345846A (en) | Intelligent grading method and system for grain size of medium and low carbon steel | |
| Kuang et al. | Review on machine learning-based defect detection of shield tunnel lining | |
| CN114550460A (en) | Rail transit anomaly detection method and device and storage medium | |
| CN111651930B (en) | Extreme learning machine-based flow field vortex region detection method | |
| CN114648095A (en) | Air quality concentration inversion method based on deep learning | |
| CN110095204A (en) | A kind of ethylene cracking tube temperature diagnostic system | |
| CN118015839B (en) | Expressway road domain risk prediction method and device | |
| CN113470739B (en) | Protein interaction prediction method and system based on mixed membership degree random block model | |
| Xu et al. | Defect recognition method based on fusion learning of multi-layer image features |
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 |