CN110501741A - A kind of useful signal detection method and system - Google Patents
A kind of useful signal detection method and system Download PDFInfo
- Publication number
- CN110501741A CN110501741A CN201910648806.4A CN201910648806A CN110501741A CN 110501741 A CN110501741 A CN 110501741A CN 201910648806 A CN201910648806 A CN 201910648806A CN 110501741 A CN110501741 A CN 110501741A
- Authority
- CN
- China
- Prior art keywords
- useful signal
- rlunet
- network model
- data set
- data
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 23
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims abstract description 16
- 239000000284 extract Substances 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000000926 separation method Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000009826 distribution Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 239000004744 fabric Substances 0.000 claims 1
- 238000005457 optimization Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 4
- 238000000034 method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/288—Event detection in seismic signals, e.g. microseismics
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
- G01V1/364—Seismic filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Acoustics & Sound (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The present invention relates to microseism data processing field more particularly to a kind of useful signal detection methods and system.Include: acquisition microseism initial data, makes data set;The data set is input in RluNet network model and is trained;The RluNet network model is addition Residulblock residual block on the basis of UNet++ network;The feature of training result useful signal is extracted by the RluNet network model, obtains useful signal Onset point.The present invention provides a kind of useful signal detection method, and by building RluNet network model, Residulblock residual block is added on the basis of UNet++ network, accurate to obtain network deep layer and shallow-layer feature, improves accuracy of identification.
Description
Method field
The present invention relates to microseism data processing field more particularly to a kind of useful signal detection methods and system.
Background method
Microseism detection method engineering construction, in terms of have important role, while microseism
Signal has signal energy weak again, vulnerable to the characteristic of ambient noise interference, causes the first arrival of microseismic signals not pick up accurately, causes
The position inaccurate of micro-seismic event, thus microseism useful signal detection method be microseism data processing field emphasis it
One.
Existing U-Net network is since shallower the extracted feature of the number of plies is not enough precisely and on extracting further feature
Show drawback;Nested-U-Net network (A Nested U-Net Architecture for Medical Image
Segmentation, hereinafter referred to as UNet++) on the basis of U-Net network jumped floor structure is added, it can effectively extract difference
Dimensional characteristics.
But the existing UNet++ network number of plies itself is shallower, and further feature is brought to identify unobvious problem, at the beginning of useful signal
It need to be improved on to the Accurate Prediction and accuracy of identification of point.
Summary of the invention
The embodiment of the present invention provides a kind of useful signal detection method and system, the new network model of RluNet, be
Residulblock residual block is added in UNet++ network, deepens network structure, keeps the useful signal feature extracted richer,
Improve the precision of useful signal Onset point detection.
The one side embodiment of the present invention provides a kind of useful signal detection method, comprising the following steps:
S1 acquires microseism initial data, makes data set;
The data set is input in RluNet network model and is trained by S2;The RluNet network model be
Residulblock residual block is added on the basis of UNet++ network;
S3 extracts the feature of training result useful signal by the RluNet network model, obtains useful signal first arrival
Point.
Wherein, step S1 includes:
S11 reads microseism initial data with matlab, and draws the sectional view of signal;
S12 extracts useful signal road, with the matlab to the useful signal analysis in the useful signal road, mistake
Filter and separation;
S13, by observing the sectional view and signal sampling point, determine respectively extracted useful signal in signal road just
To position and it is made into label, extracted each valid trace is made into text file as data set.
Wherein, step S2 includes that the data set that will be made is input in RluNet network model and is trained, and is adjusted
The whole RluNet network model parameter optimizes training result, and inputs the standard of the RluNet network model test data set
Exactness.
Wherein, step S3 includes carrying out two classification by softmax function to the characteristic feature of the useful signal to obtain
Probability value qi (x), the position where determining maximum probability distribution are the Onset point of the useful signal of prediction;Probability value formula:
When wherein the probability value qi (x) is 0, non-Onset point is indicated, the value of i is 2;When the probability value qi (x) is 1
Indicate Onset point, the value of i is 1;K (x) represents classification, k=1, and 2 representatives respectively represent two class of Onset point and non-Onset point;i
(x) each specific classification is represented, i.e. i (x) is any value of k (x) intermediate value.
On the other hand, the embodiment of the present invention provides a kind of useful signal detection system, including
Data preprocessing module: acquisition microseism initial data makes data set;
Data training module: the data set is input in RluNet network model and is trained;The RluNet net
Network model is addition Residulblock residual block on the basis of UNet++ network;
Data outputting module: extracting the feature of training result useful signal by the RluNet network model, and acquisition has
Imitate signal Onset point.
Wherein, the data preprocessing module includes: to read microseism initial data with matlab, and draw signal
Sectional view;Useful signal road is extracted, the useful signal in the useful signal road is analyzed, filter and is divided with the matlab
From;By observing the sectional view and signal sampling point, determination has respectively extracted the first arrival position of useful signal in signal road simultaneously
It is made into label, extracted each valid trace is made into text file as data set.
Wherein, the data set made is input in RluNet network model and carries out by the data training module
Training adjusts the RluNet network model parameter, optimizes training result, and input the RluNet network model test data
The accuracy of collection.
Wherein, the data outputting module carries out two points by softmax function to the characteristic feature of the useful signal
Class acquisition probability value qi (x), the position where determining maximum probability distribution are the Onset point of the useful signal of prediction;Probability value is public
Formula:
When wherein the probability value qi (x) is 0, non-Onset point is indicated, the value of i is 2;When the probability value qi (x) is 1
Indicate Onset point, the value of i is 1;K (x) represents classification, k=1, and 2 representatives respectively represent two class of Onset point and non-Onset point;i
(x) each specific classification is represented, i.e. i (x) is any value of k (x) intermediate value.
The present invention provides a kind of useful signal detection method and system, by building RluNet network model, in UNet++
Residulblock residual block is added on the basis of network, it is accurate to obtain network deep layer and shallow-layer feature, improve accuracy of identification.
Detailed description of the invention
It, below will be to needed in the method for the present invention description in order to illustrate more clearly of technical solution of the present invention
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the commonsense method personnel of domain, without any creative labor, it can also be obtained according to these attached drawings others
Attached drawing.
Fig. 1 is a kind of useful signal detection method flow diagram of the embodiment of the present invention;
Fig. 2 is a kind of useful signal detection method RluNet network architecture schematic diagram of the embodiment of the present invention;
Fig. 3 is a kind of useful signal detection method sub-process schematic diagram of the embodiment of the present invention;
Fig. 4 is a kind of sub- raw-data map of useful signal detection method of the embodiment of the present invention;
Fig. 5 is a kind of useful signal detection method original signal sectional view of the embodiment of the present invention;
Fig. 6 is a kind of useful signal detection method Onset point location map of the embodiment of the present invention;
Fig. 7 is a kind of useful signal detection system structure of the embodiment of the present invention;
Appended drawing reference:
- 2 data outputting module -3 of -1 data training module of data preprocessing module.
Specific embodiment
It is described below in conjunction with principle and feature of the attached drawing to the embodiment of the present invention, example is served only for explaining this hair
It is bright, it is not intended to limit the scope of the present invention.
Fig. 1 is a kind of useful signal detection method flow diagram of the embodiment of the present invention;Fig. 2 is that the embodiment of the present invention is a kind of
Useful signal detection method RluNet network architecture schematic diagram;As shown in Figure 1 and Figure 2, a kind of useful signal detection method,
The following steps are included:
S1 acquires microseism initial data, makes data set;
The data set is input in RluNet network model and is trained by S2;The RluNet network model be
Residulblock residual block is added on the basis of UNet++ network;Specifically, the data set made is input to RluNet
Training in network model is added three residual blocks in the bottom of U-Net++, passes through a volume between adjacent residual block in Fig. 2
Lamination and pond layer link together.Data set is input to the input layer of U-Net++ network, which will pass through U-
The processing of the bottom, residual block network, U-Net++ expansion network of Net++ network carries out the extraction of shallow-layer, further feature;
S3 extracts the feature of training result useful signal by the RluNet network model, obtains useful signal first arrival
Point.
The present invention provides a kind of useful signal detection method, by building RluNet network model, in UNet++ network
On the basis of Residulblock residual block is added, it is accurate to obtain network deep layer and shallow-layer feature, improve accuracy of identification.
Further, Fig. 3 is a kind of useful signal detection method sub-process schematic diagram of the embodiment of the present invention;Fig. 4 is this hair
A kind of bright sub- raw-data map of useful signal detection method of embodiment;Fig. 5 is a kind of useful signal detection side of the embodiment of the present invention
Method original signal sectional view;Step S1 includes: as shown in Figure 3
S11 reads microseism initial data with matlab, and draws the sectional view of signal;Specifically, such as figure, Fig. 5 institute
Show, according to different original signals, is read with matlab and draw original signal picture and sectional view;
S12 extracts useful signal road, with the matlab to the useful signal analysis in the useful signal road, mistake
Filter and separation;Specifically, from original signal picture and sectional view observation signal sampled point, therefrom extract useful signal
Non valid trace is rejected in road, then recycle matlab to carry out time frequency analysis to signal, signal is realized by low pass or high-pass filtering and
The separation of noise, abundant data collection.
S13, by observing the sectional view and signal sampling point, determine respectively extracted useful signal in signal road just
To position and it is made into label, extracted each valid trace is made into text file as data set.Specifically, production label is people
Work observation signal sectional view finds out useful signal road position, and the useful signal road extracted is amplified, useful signal is found
These position coordinates, that is, labels of the position coordinates of Onset point.
Further, step S2 includes that the data set that will be made is input in RluNet network model and instructs
Practice, adjust the RluNet network model parameter, optimizes training result, and input the RluNet network model test data set
Accuracy.Specifically, the data set made is input in RluNet network model and is trained, is arranged 1000 times
Repetitive exercise, training is completed to carry out the adjustment of network model parameter every time;When network model when loss value levels off to 0, accuracy
Close to 1, accuracy can achieve 80% in present networks model;When loss value and accuracy value level off to constant, generation
Table network training has been saturated, and stops iteration.
Further, Fig. 6 is a kind of useful signal detection method Onset point location map of the embodiment of the present invention;Such as Fig. 6
Shown, step S3 includes carrying out two classification acquisition probability value qi by softmax function to the characteristic feature of the useful signal
(x), the position where determining maximum probability distribution is the Onset point of the useful signal of prediction;Probability value formula:
When wherein the probability value qi (x) is 0, non-Onset point is indicated, the value of i is 2;When the probability value qi (x) is 1
Indicate Onset point, the value of i is 1;K (x) represents classification, k=1, and 2 representatives respectively represent two class of Onset point and non-Onset point;i
(x) each specific classification is represented, i.e. i (x) is any value of k (x) intermediate value;In Fig. 6, horizontal axis represents useful signal and position occurs
It sets, the longitudinal axis represents the probability occurred in the point.
The present invention provides a kind of useful signal detection method, by building RluNet network model, in UNet++ network
On the basis of Residulblock residual block is added, it is accurate to obtain network deep layer and shallow-layer feature, improve accuracy of identification.
Based on the above embodiment, Fig. 7 is a kind of useful signal detection system structure of the embodiment of the present invention, and data are pre-
Processing module 1: acquisition microseism initial data makes data set;
Data training module 2: the data set is input in RluNet network model and is trained;The RluNet net
Network model is addition Residulblock residual block on the basis of UNet++ network;Data outputting module 3:
The feature of training result useful signal is extracted by the RluNet network model, obtains useful signal Onset point.
The present invention provides a kind of useful signal detection system execution above method, builds RluNet net in data training module
Residulblock residual block is added in network model on the basis of UNet++ network, accurate to obtain network deep layer and shallow-layer feature,
Improve accuracy of identification;The fining of signal characteristic is extracted simultaneously, by the skip floor in UNet++ network, beta pruning structure, is being mentioned
While taking out signal main and imperceptibility feature, avoid occur feature pile up, over-fitting the problem of.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations, although referring to before
Stating embodiment, invention is explained in detail, and the commonsense method personnel of this field are it is understood that it still can be to preceding
It states technical solution documented by each embodiment to modify, or part of method characteristic is equivalently replaced;And these
It modifies or replaces, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (8)
1. a kind of useful signal detection method, which comprises the following steps:
S1 acquires microseism initial data, makes data set;
The data set is input in RluNet network model and is trained by S2;The RluNet network model is in UNet+
Residulblock residual block is added on the basis of+network;
S3 extracts the feature of training result useful signal by the RluNet network model, obtains useful signal Onset point.
2. a kind of useful signal detection method according to claim 1, which is characterized in that step S1 includes:
S11 reads microseism initial data with matlab, and draws the sectional view of signal;
S12, extract useful signal road, with the matlab in the useful signal road useful signal analysis, filtering and
Separation;
S13 determines the first arrival position for respectively having extracted useful signal in signal road by observing the sectional view and signal sampling point
Label is set and be made into, extracted each valid trace is made into text file as data set.
3. a kind of useful signal detection method according to claim 1, which is characterized in that step S2 includes that will make
The data set is input in RluNet network model and is trained, and adjusts the RluNet network model parameter, optimization training
As a result, and inputting the accuracy of the RluNet network model test data set.
4. a kind of useful signal detection method according to claim 1, which is characterized in that step S3 includes to described effective
The characteristic feature of signal carries out two classification acquisition probability value qi (x) by softmax function, determines maximum probability distribution place
Position is the Onset point of the useful signal of prediction;Probability value formula:
When wherein the probability value qi (x) is 0, non-Onset point is indicated, the value of i is 2;The probability value qi (x) indicates when being 1
Onset point, the value of i are 1;K (x) represents classification, k=1, and 2 representatives respectively represent two class of Onset point and non-Onset point;I (x) generation
Each specific classification of table, i.e. i (x) are any value of k (x) intermediate value.
5. a kind of useful signal detection system, which is characterized in that
Data preprocessing module (1): acquisition microseism initial data makes data set;
Data training module (2): the data set is input in RluNet network model and is trained;The RluNet network
Model is addition Residulblock residual block on the basis of UNet++ network;
Data outputting module (3): extracting the feature of training result useful signal by the RluNet network model, obtains effective
Signal Onset point.
6. a kind of useful signal detection system according to claim 5, which is characterized in that the data prediction mould (1)
Block includes: to read microseism initial data with matlab, and draw the sectional view of signal;Useful signal road is extracted, with institute
Matlab is stated to useful signal analysis, filtering and the separation in the useful signal road;By observing the sectional view and letter
Number sampled point, determination has respectively extracted the first arrival position of useful signal in signal road and has been made into label, by extracted each valid trace
Text file is made into as data set.
7. a kind of useful signal detection system according to claim 5, which is characterized in that the data training module, it will
The data set made, which is input in RluNet network model, to be trained, and the RluNet network model parameter is adjusted,
Optimize training result, and inputs the accuracy of the RluNet network model test data set.
8. a kind of useful signal detection system according to claim 5, which is characterized in that the data outputting module is right
The characteristic feature of the useful signal carries out two classification acquisition probability value qi (x) by softmax function, determines maximum probability point
Position where cloth is the Onset point of the useful signal of prediction;Probability value formula:
When wherein the probability value qi (x) is 0, non-Onset point is indicated, the value of i is 2;The probability value qi (x) indicates when being 1
Onset point, the value of i are 1;K (x) represents classification, k=1, and 2 representatives respectively represent two class of Onset point and non-Onset point;I (x) generation
Each specific classification of table, i.e. i (x) are any value of k (x) intermediate value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910648806.4A CN110501741A (en) | 2019-07-18 | 2019-07-18 | A kind of useful signal detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910648806.4A CN110501741A (en) | 2019-07-18 | 2019-07-18 | A kind of useful signal detection method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110501741A true CN110501741A (en) | 2019-11-26 |
Family
ID=68585626
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910648806.4A Pending CN110501741A (en) | 2019-07-18 | 2019-07-18 | A kind of useful signal detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110501741A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111505706A (en) * | 2020-04-28 | 2020-08-07 | 长江大学 | Microseism P wave first arrival pickup method and device based on depth T-Net network |
CN111695413A (en) * | 2020-04-28 | 2020-09-22 | 长江大学 | Signal first arrival pickup method and device combining U-Net and Temporal encoding |
CN111796326A (en) * | 2020-07-07 | 2020-10-20 | 中海石油(中国)有限公司 | Method and system for constructing sequence stratum framework based on Unet network |
CN113534240A (en) * | 2021-07-07 | 2021-10-22 | 中国石油大学(华东) | Microseism event detection and positioning method and system |
-
2019
- 2019-07-18 CN CN201910648806.4A patent/CN110501741A/en active Pending
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111505706A (en) * | 2020-04-28 | 2020-08-07 | 长江大学 | Microseism P wave first arrival pickup method and device based on depth T-Net network |
CN111695413A (en) * | 2020-04-28 | 2020-09-22 | 长江大学 | Signal first arrival pickup method and device combining U-Net and Temporal encoding |
CN111505706B (en) * | 2020-04-28 | 2023-04-18 | 长江大学 | Microseism P wave first arrival pickup method and device based on deep T-Net network |
CN111695413B (en) * | 2020-04-28 | 2023-06-20 | 长江大学 | Signal first arrival pickup method and device combining U-Net and Temporal Ensembling |
CN111796326A (en) * | 2020-07-07 | 2020-10-20 | 中海石油(中国)有限公司 | Method and system for constructing sequence stratum framework based on Unet network |
CN111796326B (en) * | 2020-07-07 | 2022-11-22 | 中海石油(中国)有限公司 | Method and system for constructing sequence stratum framework based on Unet network |
CN113534240A (en) * | 2021-07-07 | 2021-10-22 | 中国石油大学(华东) | Microseism event detection and positioning method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110501741A (en) | A kind of useful signal detection method and system | |
CN107346420A (en) | Text detection localization method under a kind of natural scene based on deep learning | |
CN110378879A (en) | A kind of Bridge Crack detection method | |
CN101806583B (en) | Microscopic image-based fiber fineness measurement method | |
CN107633250A (en) | A kind of Text region error correction method, error correction system and computer installation | |
CN107392141B (en) | Airport extraction method based on significance detection and LSD (least squares distortion) line detection | |
CN105388162B (en) | Raw material silicon chip surface scratch detection method based on machine vision | |
CN108710919A (en) | A kind of crack automation delineation method based on multi-scale feature fusion deep learning | |
CN108596046A (en) | A kind of cell detection method of counting and system based on deep learning | |
CN108596038B (en) | Method for identifying red blood cells in excrement by combining morphological segmentation and neural network | |
CN105608454A (en) | Text structure part detection neural network based text detection method and system | |
CN104239902B (en) | Hyperspectral image classification method based on non local similitude and sparse coding | |
CN103528617A (en) | Cockpit instrument automatic recognizing and detecting method and device | |
CN101877074A (en) | Tubercle bacillus target recognizing and counting algorithm based on diverse characteristics | |
CN105913090B (en) | SAR image objective classification method based on SDAE-SVM | |
CN107292314A (en) | A kind of lepidopterous insects species automatic identification method based on CNN | |
CN105513066B (en) | It is a kind of that the generic object detection method merged with super-pixel is chosen based on seed point | |
CN106408030A (en) | SAR image classification method based on middle lamella semantic attribute and convolution neural network | |
CN111507296A (en) | Intelligent illegal building extraction method based on unmanned aerial vehicle remote sensing and deep learning | |
CN105138983B (en) | The pedestrian detection method divided based on weighting block model and selective search | |
CN103440035A (en) | Gesture recognition system in three-dimensional space and recognition method thereof | |
CN110533100A (en) | A method of CME detection and tracking is carried out based on machine learning | |
CN115601544A (en) | High-resolution image landslide detection and segmentation method | |
CN109472790A (en) | A kind of machine components defect inspection method and system | |
CN103268494B (en) | Parasite egg recognition methods based on rarefaction representation |
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 |