CN110806444B - Seabed sediment recognition and classification method based on shallow stratum profiler and SVM - Google Patents

Seabed sediment recognition and classification method based on shallow stratum profiler and SVM Download PDF

Info

Publication number
CN110806444B
CN110806444B CN201911111407.0A CN201911111407A CN110806444B CN 110806444 B CN110806444 B CN 110806444B CN 201911111407 A CN201911111407 A CN 201911111407A CN 110806444 B CN110806444 B CN 110806444B
Authority
CN
China
Prior art keywords
data
substrate
training
svm
ping
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911111407.0A
Other languages
Chinese (zh)
Other versions
CN110806444A (en
Inventor
罗宇
郑旭
施剑
吴逸凡
徐辉
李斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University of Science and Technology
Original Assignee
Shandong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong University of Science and Technology filed Critical Shandong University of Science and Technology
Priority to CN201911111407.0A priority Critical patent/CN110806444B/en
Publication of CN110806444A publication Critical patent/CN110806444A/en
Application granted granted Critical
Publication of CN110806444B publication Critical patent/CN110806444B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/04—Analysing solids
    • 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
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • 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
    • G06F18/24133—Distances to prototypes
    • G06F18/24137—Distances to cluster centroïds
    • G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30—Assessment of water resources

Abstract

The invention discloses a seabed sediment recognition and classification method based on a shallow stratum profiler and an SVM, which comprises the following specific steps: step 1, acquiring shallow stratum echo acoustic signal data of a specified place by using a shallow stratum profiler; step 2, preprocessing the data in the step 1 to obtain a training set and a test set required by SVM training; step 3, modeling the RBF kernel function model, and training the model to obtain a water bottom substrate classification model; step 4, obtaining the optimal parameter C and parameter through multiple tests by using a cross validation methodγFurther obtaining an optimal substrate classification model; the method utilizes the SVM bottom material classification model of the data measured by the shallow stratum profiler, can process large batch of data after training, can efficiently and accurately identify the type of the seabed bottom material, and has important application value.

Description

Seabed sediment recognition and classification method based on shallow stratum profiler and SVM
Technical Field
The invention relates to a method for recognizing and classifying seabed sediments, in particular to a method for recognizing and classifying the seabed sediments based on a shallow stratum profiler and an SVM, and belongs to the technical field of marine surveying and mapping.
Background
The classification research of the seabed sediments has very broad prospect and significance in civil use and military use, and the currently common seabed classification methods comprise a direct sampling method and an indirect detection method. The direct sampling can realize accurate substrate classification, but the cost is high, the realization difficulty of deep water is high, the indirect detection method is to detect by means of optics, acoustics, biology and the like, the principle of remotely measuring the type of the sediment by using the acoustics method is to transmit sound waves to the seabed by using an acoustic transducer, and the attributes of the seabed sediment are known by recording and analyzing echo signals, so the method has the characteristics of high efficiency and economy.
Among various ocean exploration instruments, a shallow Profiler (Sub-bottom Profiler) has lower emission frequency and strong sound wave signal penetration capability compared with other acoustic exploration instruments, and can detect the structure and the construction condition of a shallow stratum below the sea bottom. The acoustic data acquired by the shallow stratum profiler is particularly suitable for identifying and classifying the submarine geology.
The purpose of machine learning is to evaluate the dependency between the inputs and outputs of a system based on a given training sample, so that it can predict the unknown output as accurately as possible. A Support Vector Machine (SVM) is a Machine learning algorithm based on a statistical theory, classification is carried out according to a structure risk minimum principle, an optimal boundary is searched in a training sample set, and data are separated to the maximum extent. The SVM has good effect of classifying the multi-dimensional small sample data, the submarine data is difficult to calibrate and sample, the number of samples is small, and the SVM is suitable for classifying and identifying the substrate.
How to efficiently and accurately acquire sound wave signals of different substrates and select a proper classifier for classification is a key problem for substrate classification. If the substrate can be classified by combining the basic data measured by the shallow stratum profiler, the method is more accurate than the traditional direct sampling method and the traditional indirect detection method, and the working efficiency is higher.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a seabed sediment identification and classification method based on a shallow stratum profiler and an SVM (support vector machine), which is used for solving the technical problems of low efficiency and low accuracy of the conventional method.
In order to solve the problems in the background art, the method for identifying and classifying the seabed sediment based on the shallow stratum profiler and the SVM comprises the following steps:
step 1: and acquiring the shallow stratum echo acoustic signal data of the designated place by using a shallow stratum profiler.
Step 2: and (3) preprocessing the data in the step (1) to obtain a training set and a test set required by SVM training.
2.1, specific identifiers exist among every ping data of the original data, the data are segmented through the identifiers, and the data are read and stored in a ping mode; after reading and writing, we can see that each ping has N data picture original data representing the characteristics of the sound signals.
2.2 according to the thickness and the water depth of the geology, useful data are selected from the data of each ping, and the formula is as follows:
S＝N*(X/M)
P＝N*(H/M)
wherein S is the data volume of water bottom substrate which can be reflected by each ping, N is the data volume of each ping, X is the default thickness of the substrate, M is the maximum measuring range set by the shallow profile profiler, P is the effective data starting sequence number, and H is the water depth.
2.3, carrying out normalization processing on the sorted data: the result values are mapped between [0-1] using dispersion Normalization (Min-Max Normalization). The transfer function is as follows:
Figure BDA0002272819640000021
wherein r isiA value, max { r, representing the shallow formation echo acoustic signal data corresponding to the ith data arrayiThe maximum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays is represented, and min { r }iAnd expressing the minimum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays.
2.4 according to the above steps, selecting the better 4 groups of 5 groups of data of each substrate type, each group selecting 200ping data and calibrating the substrate type of the data, wherein 2 groups are used as training sets, 1 group is used as verification sets for cross-verification, and 1 group is used as test sets for verifying the effect after training.
And 3, step 3: and modeling by using the RBF kernel function model, and training the model to obtain a water bottom substrate classification model.
And 3.1, counting the basic characteristics of the echo acoustic signals of different superficial strata, modeling according to the number of the effective data of each ping obtained in the step 2 as a characteristic vector, wherein the adopted kernel function models are Radial Basis Function (RBF) functions respectively. The RBF kernel can map samples in a nonlinear high-dimensional space, has a wider convergence domain and wider adaptability, has only one parameter gamma and small calculation difficulty and complexity, and is an ideal mapping kernel function.
Radial Basis (RBF) function: k (x, x)i)＝exp(-γ||xi-xj||2)，γ＞0。
And 3.2, according to the training set obtained in the step 2, taking the acoustic signal data of the echo of the superficial stratum as input, outputting the acoustic signal data corresponding to the type of the bottom material, and training the model to obtain a classification model of the bottom material.
And 4, step 4: and (3) obtaining an optimal parameter C and parameter gamma by multiple tests by using a Cross-validation (Cross-validation)) method, and further obtaining an optimal substrate classification model.
And 4.1, testing the substrate classification model obtained in the step 3 by using the verification set of the calibrated substrate sampled in the step 2, inputting the test set data, outputting the predicted classification result, and comparing the result with the known substrate to obtain the classification accuracy which is used as the performance index of the evaluation classifier.
4.2 changing the radius C of the kernel function and the parameter gamma, repeating the step 3 and the step 4.1, and selecting the optimal parameter C and the optimal parameter gamma to obtain the optimal substrate classification model.
Compared with the prior art, the invention has the beneficial effects that: the method has the advantages that the shallow stratum profiler with low working frequency and certain penetrating power is used for collecting the seabed echo data, the change laws of different substrate echoes are different, the characteristics of different types of seabed substrates can be accurately reflected, a substrate classification model is built based on the SVM, a good classification effect can be obtained under smaller sample data, various substrate types can be accurately classified and recognized, the method can be used for classifying by simply processing the collected original data, the classification speed is high, the efficiency is high, and the method has important practical value.
Drawings
FIG. 1 is a flow chart of the operation of the present invention.
Fig. 2 is a per ping data characterization of raw data collected by a shallow profiler in an embodiment of the present invention.
FIG. 3 is a summary of data conditions in an embodiment of the invention.
FIG. 4 is a bottom sample data box plot in an embodiment of the invention.
FIG. 5 is a two-dimensional coordinate distribution diagram of the substrate sample data in an embodiment of the invention.
Fig. 6 is a schematic diagram of the training principle of the present invention.
FIG. 7 is a diagram of the actual classification and predicted classification of a test set in an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
In order to prove the accuracy of the method for identifying and classifying the seabed sediment based on the shallow stratum profiler and the SVM, a certain laboratory water pool is selected below, the types of the sediment are all known and fixed sampling sites are fixed, 5 groups of data are randomly measured in areas with different sediments, and the fixed-point measurement time is 5 minutes each time.
Step 1: and acquiring the shallow stratum echo acoustic signal data of the designated place by using a shallow stratum profiler. The shallow stratum profiler who this uses is the SBP220 wide band shallow stratum profiler of Shandong science and technology university institute ocean intelligent navigation and perception laboratory development, and the operating frequency range is 2-20 KHz, and power is great, and the penetrability is stronger, can effectively survey various geology, selects to test in the certain experiment basin, and three kinds of bases of cement, earth, sand are total in this basin, and 5 group data are measured at random in the region of different bases, and fixed point measuring time is 5 minutes at every turn.
Step 2: and (3) preprocessing the data in the step (1) to obtain a training set and a test set required by SVM training.
2.1 the original data has specific identifier between every ping data, and the data is divided by the identifier and read and stored in ping mode. After reading and writing, we can see that there are N data picture raw data representing the characteristics of the acoustic signal per ping, as shown in fig. 2.
2.2 according to the thickness and the water depth of the geology, useful data are selected from the data of each ping, and the formula is as follows:
S＝N*(X/M)
P＝N*(H/M)
wherein S is the data volume of each ping capable of reflecting the underwater substrate, N is the data volume of each ping, X is the default thickness of the substrate, M is the maximum range set by the shallow profiler, P is the effective data starting sequence number, H is the water depth, the experimental data interval is 8, namely the effective shallow profile data obtained in each ping is 8 bits.
And after the data processing is finished, drawing a two-dimensional coordinate distribution diagram of the bottom sample data box diagram in the figure 4 and the two-dimensional coordinate distribution diagram of the bottom sample data in the figure 5 to check the samples, and preventing the training effect from being influenced by the abnormal values.
2.3, carrying out normalization processing on the sorted data: the result values are mapped between [0-1] using dispersion Normalization (Min-Max Normalization). The transfer function is as follows:
Figure BDA0002272819640000041
wherein r isiA value, max { r, representing the shallow formation echo acoustic signal data corresponding to the ith data arrayiThe maximum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays is represented, and min { r }iAnd expressing the minimum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays.
2.4 according to the above steps, select the better 4 groups of 5 groups of data of each substrate type, each group selects 200ping data and calibrates the substrate type of the data, wherein 2 groups are used as training set, 1 group is used as validation set for cross validation, 1 group is used as test set for verifying the effect after training, and the data condition is shown in fig. 3.
And step 3: and modeling by using the RBF kernel function model, and training the model to obtain a water bottom substrate classification model.
The method comprises the following steps:
3.1, counting the basic characteristics of echo acoustic signals of different superficial stratums, modeling according to the number of effective data of each ping obtained in the step 2 as a characteristic vector, wherein the adopted kernel function models are Radial Basis Function (RBF) functions respectively, and RBF kernels can map samples in a nonlinear high-dimensional space, have a wider convergence domain and wider adaptability, and only one parameter gamma of the kernel function has small calculation difficulty and complexity, so that the method is an ideal mapping kernel function.
Radial Basis (RBF) function: k (x, x)i)＝exp(-γ||xi-xj||2)，γ＞0。
And 3.2, according to the training set obtained in the second step, outputting corresponding substrate types by taking the acoustic signal data of the echo of the superficial stratum as input, and training the model to obtain a water bottom substrate classification model.
And 4, step 4: and (3) obtaining an optimal parameter C and parameter gamma by multiple tests by using a Cross-validation (Cross-validation)) method, and further obtaining an optimal substrate classification model.
And 4.1, testing the substrate classification model obtained in the step 3 by using the verification set of the calibrated substrate sampled in the step 2, inputting the test set data, outputting the predicted classification result, and comparing the result with the known substrate to obtain the classification accuracy which is used as the performance index of the evaluation classifier.
4.2 changing the radius C of the kernel function and the parameter gamma, repeating the step 3 and the step 4.1, and selecting the optimal parameter C and the optimal parameter gamma to obtain the optimal substrate classification model.
To verify the accuracy of the method: and (3) testing the substrate classification model by using the test set in the step (3) according to the optimal substrate classification model generated in the step (4), inputting the test set in the step (3) into the substrate classification model, comparing a result output by the substrate classification model with the real substrate type marked by the model, and representing and predicting the superposition of the result obtained after training and the actually marked substrate accurately, wherein the test is carried out in three substrate types, namely, sand soil of a label 1, cement of a label 2 and soil of a label 3 as shown in figure 7. The test set comprises three types of substrates, wherein each type of substrate is 200ping data, a prediction result is obtained after the substrate classification model, most of the substrates are the same as the known substrate in type, namely, the classification is correct, and only a few 'flying spots' are classified as wrong parts.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.

Claims (2)

1. The method for recognizing and classifying the seabed sediment based on the shallow stratum profiler and the SVM is characterized by comprising the following specific steps of:
step 1: acquiring shallow stratum echo acoustic signal data of a specified place by using a shallow stratum profiler;
step 2: preprocessing the data in the step 1 to obtain a training set and a test set required by SVM training;
2.1 the original data has specific identifier between every ping data, the data is divided by the identifier, and the data is read and stored by ping; through reading and writing, the original data of the data pictures with N representative sound signal characteristics can be obtained per ping;
2.2 according to the thickness and the water depth of the geology, useful data are selected from the data of each ping, and the formula is as follows:
S＝N*(X/M)
P＝N*(H/M)
the method comprises the following steps that S is the data volume of water bottom substrate which can be reflected by each ping, N is the data volume of each ping, X is the default thickness of the substrate, M is the maximum measuring range set by a shallow stratum profiler, P is the effective data starting sequence number, and H is the water depth;
2.3, carrying out normalization processing on the sorted data: using dispersion normalization, the resulting values are mapped between [0-1], and the transfer function is as follows:
Figure FDA0002272819630000011
wherein r isiA value, max { r, representing the shallow formation echo acoustic signal data corresponding to the ith data arrayiThe maximum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays is represented, and min { r }iRepresenting the minimum value of the shallow stratum echo acoustic signal data corresponding to all the sample data arrays;
2.4 according to the steps, selecting better 4 groups from the 5 groups of data of each substrate type, selecting 200ping data in each group and calibrating the substrate type of the data, wherein 2 groups are used as a training set, 1 group is used as a verification set for cross verification, and 1 group is used as a test set for verifying the effect after training;
and 3, step 3: modeling by using an RBF kernel function model, and training the model to obtain a water bottom substrate classification model;
3.1, counting the basic characteristics of the echo acoustic signals of different superficial strata, modeling according to the number of effective data of each ping obtained in the step 2 as a characteristic vector, wherein the adopted kernel function models are Radial Basis Function (RBF) functions respectively;
radial Basis (RBF) function: k (x, x)i)＝exp(-γ||xi-xj||2)，γ＞0
3.2 according to the training set obtained in the step 2, taking the echo acoustic signal data of the superficial stratum as input, outputting corresponding to the type of the substrate, and training the model to obtain a water bottom substrate classification model;
and 4, step 4: by using a cross validation method, testing for multiple times to obtain an optimal parameter C and an optimal parameter gamma, and further obtaining an optimal substrate classification model;
4.1 testing the substrate classification model obtained in the step 3 by using the verification set of the calibrated substrate sampled in the step 2, inputting the test set data, outputting the predicted classification result, and comparing the result with the known substrate to obtain the classification accuracy serving as the performance index of the evaluation classifier;
4.2 changing the radius C of the kernel function and the parameter gamma, repeating the step 3 and the step 4.1, and selecting the optimal parameter C and the optimal parameter gamma to obtain the optimal substrate classification model.
2. The method for identifying and classifying seafloor sediments based on shallow profilers and SVM as claimed in claim 1, wherein the data collected at each designated place in step 1 cannot be less than five groups.
CN201911111407.0A 2019-11-14 2019-11-14 Seabed sediment recognition and classification method based on shallow stratum profiler and SVM Active CN110806444B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911111407.0A CN110806444B (en) 2019-11-14 2019-11-14 Seabed sediment recognition and classification method based on shallow stratum profiler and SVM

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911111407.0A CN110806444B (en) 2019-11-14 2019-11-14 Seabed sediment recognition and classification method based on shallow stratum profiler and SVM

Publications (2)

Publication Number Publication Date
CN110806444A CN110806444A (en) 2020-02-18
CN110806444B true CN110806444B (en) 2022-06-07

Family

ID=69502604

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911111407.0A Active CN110806444B (en) 2019-11-14 2019-11-14 Seabed sediment recognition and classification method based on shallow stratum profiler and SVM

Country Status (1)

Country Link
CN (1) CN110806444B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111709386B (en) * 2020-06-22 2023-05-12 中国科学院空天信息创新研究院 Underwater shallow stratum profile image substrate classification method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106770640A (en) * 2016-11-25 2017-05-31 淮海工学院 With the matching used laboratory testing rig of sub-bottom profiler
CN107247291A (en) * 2017-06-21 2017-10-13 上海海洋大学 The shallow stratum sound energy attenuation model construction in seabed and two important sound energy attenuation characteristic parameter extraction method
CN108520271A (en) * 2018-03-15 2018-09-11 中国石油大学(华东) Submarine geomorphy type sorter design method based on factorial analysis
CN109086824A (en) * 2018-08-01 2018-12-25 哈尔滨工程大学 A kind of sediment sonar image classification method based on convolutional neural networks

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9128203B2 (en) * 2011-09-28 2015-09-08 Saudi Arabian Oil Company Reservoir properties prediction with least square support vector machine

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106770640A (en) * 2016-11-25 2017-05-31 淮海工学院 With the matching used laboratory testing rig of sub-bottom profiler
CN107247291A (en) * 2017-06-21 2017-10-13 上海海洋大学 The shallow stratum sound energy attenuation model construction in seabed and two important sound energy attenuation characteristic parameter extraction method
CN108520271A (en) * 2018-03-15 2018-09-11 中国石油大学(华东) Submarine geomorphy type sorter design method based on factorial analysis
CN109086824A (en) * 2018-08-01 2018-12-25 哈尔滨工程大学 A kind of sediment sonar image classification method based on convolutional neural networks

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于声纳回波波形特征的海底底质类型分类方法研究;万宏俊;《万方》;20190806;第1、18、45-53页 *
多波束测深声呐海底底质分类技术研究;徐超;《万方》;20150817;第87-99页 *
海底声呐图像智能底质分类技术研究综述;赵玉新等;《智能系统学报》(第03期);全文 *

Also Published As

Publication number Publication date
CN110806444A (en) 2020-02-18

Similar Documents

Publication Publication Date Title
CN112883564B (en) Water body temperature prediction method and prediction system based on random forest
Ji et al. Seabed sediment classification using multibeam backscatter data based on the selecting optimal random forest model
CN102109495A (en) Method for classifying types of mixed seabed sediment based on multi-beam sonar technology
CN111709386B (en) Underwater shallow stratum profile image substrate classification method and system
CN111506861B (en) Method for calculating crack strength of favorable region of target layer
CN104090033A (en) Method for establishing FDTD (Finite Difference Time Domain) ultrasonic detection simulation model of coarse crystal material based on EBSD (Electron Back-Scattered Diffraction) spectrum
CN113553763A (en) Tunnel surrounding rock rapid grading method and system
CN117198330B (en) Sound source identification method and system and electronic equipment
CN112733447A (en) Underwater sound source positioning method and system based on domain adaptive network
Stanton et al. Non-Rayleigh echoes from resolved individuals and patches of resonant fish at 2–4 kHz
CN110806444B (en) Seabed sediment recognition and classification method based on shallow stratum profiler and SVM
Nichols et al. Use of noise correlation matrices to interpret ocean ambient noise
CN117197596B (en) Mixed substrate acoustic classification method based on small sample transfer learning
CN102914797A (en) Method and device for acquiring anisotropy coefficient of stratum
CN113219531A (en) Method and device for identifying gas-water distribution of tight sandstone
CN113484907B (en) Method for predicting distribution on different types of reservoir planes
TANG et al. Application of LVQ neural network combined with the genetic algorithm in acoustic seafloor classification
Smith et al. Using topographic slope distributions to infer seafloor patterns
Hong et al. A DTW distance-based seismic waveform clustering method for layers of varying thickness
CN113221651A (en) Seafloor sediment classification method using acoustic propagation data and unsupervised machine learning
Yuan et al. A convolutional neural network for prestack fracture detection
CN113884575B (en) Surrounding rock grading method and system
CN117331125B (en) Split beam data processing method and device, electronic equipment and medium
CN112114361B (en) Vibration field space-time tomography method in shallow underground complex space
Pulliam et al. Assessing uncertainties in waveform modeling of the crust and upper mantle

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant