CN112180369B - Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer - Google Patents
Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer Download PDFInfo
- Publication number
- CN112180369B CN112180369B CN202011013771.6A CN202011013771A CN112180369B CN 112180369 B CN112180369 B CN 112180369B CN 202011013771 A CN202011013771 A CN 202011013771A CN 112180369 B CN112180369 B CN 112180369B
- Authority
- CN
- China
- Prior art keywords
- sea surface
- layer
- inversion
- wind speed
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
- G01S13/9027—Pattern recognition for feature extraction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/28—Details of pulse systems
- G01S7/285—Receivers
- G01S7/295—Means for transforming co-ordinates or for evaluating data, e.g. using computers
- G01S7/2955—Means for determining the position of the radar coordinate system for evaluating the position data of the target in another coordinate system
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Electromagnetism (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses a sea surface wind speed inversion method of a one-dimensional synthetic aperture radiometer based on deep learning, which comprises the following steps: acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content; inputting the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content and the cloud liquid water content into a radiation transmission forward model to obtain simulated bright temperature; and inputting the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on the convolutional neural network to obtain the sea surface wind speed. The inversion accuracy is improved, and a method is provided for inverting the sea surface wind speed by the one-dimensional comprehensive pore-forming microwave radiometer.
Description
Technical Field
The invention relates to the technical field of remote sensing, in particular to a sea surface wind speed inversion method of a one-dimensional synthetic aperture radiometer based on deep learning.
Background
Sea surface wind speed influences sea-air interaction and is an important physical quantity for marine environment detection. The microwave remote sensor has all-weather observation capability and certain subsurface detection capability, and is one of the main means for detecting sea surface wind speed. However, for the satellite platform, the size and weight of the antenna are strictly limited, and the spatial resolution is low. The synthetic aperture microwave radiometer is a product of applying interference technology to earth observation, and adopts a small aperture antenna array, so that the contradiction between high resolution and large antenna of the microwave radiometer is solved, and the resolution is obviously improved. However, the imaging mode of the synthetic aperture microwave radiometer is completely different from that of the real aperture microwave radiometer, so that the existing sea surface wind speed inversion algorithm of the real aperture microwave radiometer cannot be applied to the one-dimensional synthetic aperture microwave radiometer.
The sea surface wind speed inversion algorithm of the synthetic aperture microwave radiometer needs to be completed under the condition of multiple incidence angles, and the radiation transmission between the sea surface and the atmosphere is influenced by multiple factors, so that the physical inversion method is complex and is not suitable for the satellite-borne operation requirement.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a depth learning-based sea surface wind speed inversion method for a one-dimensional synthetic aperture radiometer, so as to solve the problem that the sea surface wind speed is difficult to invert in the prior art.
In order to achieve the purpose, the invention provides the following technical scheme:
a sea surface wind speed inversion method comprises the following steps:
acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content to form inversion data;
inputting the inversion data into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model in sequence to obtain simulated brightness temperature;
and inputting the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on the convolutional neural network to obtain the sea surface wind speed.
Further, the construction method of the deep learning inversion model is as follows:
acquiring historical inversion data and corresponding sea surface wind speed, and constructing an initial data set;
randomly dividing an initial data set into an initial training set and an initial verification set;
respectively and sequentially inputting historical inversion data in the initial training set and historical inversion data in the initial verification set into a radiation transmission forward modeling model and a one-dimensional synthetic aperture microwave radiometer model to obtain corresponding simulated bright temperature;
respectively putting the corresponding simulated brightness temperature back to the initial training set and the initial verification set;
and training the deep learning inversion model through the initial training set with the simulated bright temperature, and verifying the learning inversion model through the initial verification set with the simulated bright temperature to obtain the final deep learning inversion model.
Further, the convolutional neural network of the deep learning inversion model has 10 layers, wherein the first layer is a batch normalization layer; the second layer is a convolution layer, the third layer is an average pooling layer, the fourth layer to the ninth layer are convolution layers, and the tenth layer is a full-connected layer.
Further, the activation function of the second layer of the convolutional neural network is a tanh function; the activation functions of the fourth layer to the ninth layer of the convolutional neural network are Sigmoid functions.
Further, the expression of the tanh function is:
the expression of the Sigmoid function is:
further, in the verification process, a root mean square error between an inversion result and a true value is used as a standard for judging the inversion accuracy of the model, and a calculation formula of the root mean square error is as follows:
MSE＝∑∑(Ws-Ws′)/N
where MSE is the root mean square error, WSIs the sea surface true wind speed, Ws' is the inverse wind speed and N is the total number of sample points.
Further, when the deep learning inversion model is trained through an initial training set with simulated brightness temperature data, the data in the initial training set is input into a self-encoder and the following algorithm is adopted:
X″＝s(w×X′+b)
X″′＝s(w′×X″+b′)
wherein X ' is data in the initial training set, X ' is a feature expression after X ' encoding, w represents the weight from the input layer to the hidden layer, b is hidden layer bias, s is an activation function, w ' is decoding weight, b ' is decoding bias, and X ' is the result after X ' decoding.
A sea surface wind speed inversion system, the system comprising:
a first obtaining module: the system is used for acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content to form inversion data;
a second obtaining module: the system is used for sequentially inputting inversion data into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model to obtain simulated brightness temperature;
a third obtaining module: and the system is used for inputting the sea surface temperature, the sea surface salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on the convolutional neural network to obtain the sea surface wind speed.
A sea surface wind speed inversion system, the system comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate according to the instructions to perform the steps of the method described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method described above.
Compared with the prior art, the invention has the beneficial effects that:
the method breaks through the thought of the traditional real-aperture microwave radiometer, solves the problem that the sea surface wind speed is difficult to invert by multiple incidence angles of the one-dimensional synthetic aperture microwave radiometer by utilizing the characteristics of strong nonlinear fitting of deep learning and small calculation amount of a convolution network, can invert the sea surface wind speed quickly and accurately, and provides technical support for load development and application of the follow-up satellite-borne one-dimensional synthetic aperture microwave radiometer.
Drawings
FIG. 1 is a diagram of a deep learning convolutional neural network model architecture in the present invention.
FIG. 2 is a graph showing the results of the examination in the present invention.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.
As shown in fig. 1 and 2, a depth learning-based sea surface wind speed inversion method for a one-dimensional synthetic aperture radiometer includes the following steps:
acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content to form inversion data;
inputting the inversion data into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model in sequence to obtain simulated brightness temperature;
and inputting the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on the convolutional neural network to obtain the sea surface wind speed.
The method specifically comprises the following steps:
step 1: acquiring historical data of sea surface temperature, sea surface wind speed, sea surface salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content, taking the historical data as an initial data set C, and randomly dividing the initial data set C into an initial training set A and an initial verification set B, wherein the initial training set and the initial verification set respectively account for 80% and 20% of the initial data set C. And respectively and sequentially inputting the A and the B into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model, outputting simulated brightness temperature, and returning the simulated brightness temperature to the A and the B to obtain a training set A 'and a verification set B'.
The process of generating the simulated brightness temperature takes the variable of the incidence angle into consideration, and the problem of inverting the brightness temperature by the multiple incidence angles of the one-dimensional synthetic aperture microwave radiometer is solved.
Step 2: constructing a deep learning inversion model based on a convolutional neural network, wherein the convolutional neural network has 10 layers, and the first layer is a batch normalization layer; the second layer is a convolution layer, and the activation function of the layer is a tanh function; the third layer is an average pooling layer, the fourth to ninth layers are convolution layers, the activation functions of the fourth to ninth layers are Sigmoid functions, and the tenth layer is a full connection layer.
A plurality of activation functions are selected from the deep learning inversion model, so that the nonlinear fitting effect of the neural network model can be improved, and the expression corresponding to each activation function is as follows:
sigmoid function:
AdaGrad algorithm:
in the whole training process of the step 2, the algorithm is optimized by comparing the output result with the true value and predicting the input data, such as changing the number of convolution kernels and the size of the convolution kernels in the convolution layer.
It should be noted that, in step 2, the data X' in the training set is input to the self-encoder, and the following algorithm is adopted:
X″＝s(w×X′+b)
X″′＝s(w′×X″+b′)
wherein X ' is data in the initial training set, X ' is a feature expression after X ' encoding, w represents the weight from the input layer to the hidden layer, b is hidden layer bias, s is an activation function, w ' is decoding weight, b ' is decoding bias, and X ' is the result after X ' decoding.
And step 3: and (3) taking the simulated brightness temperature, the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content and the cloud liquid water content in the training set A' as training samples of the deep learning inversion model, taking the sea surface wind speed as a true value, inputting the true value into the model, training the model, and finally outputting the sea surface wind speed. The weight iteration optimizer algorithm of the whole network adopts a deep learning optimization method, namely an AdaGrad algorithm.
And 4, inputting the simulated bright temperature in the verification set B' into the trained deep learning inversion model, calculating corresponding output, performing error comparison with the sea surface wind speed in the verification set, and checking the accuracy of the method. The root mean square error MSE between the inversion result and the real value is used as the standard for judging the inversion accuracy of the model, and the calculation formula is as follows:
MSE＝∑∑(Ws-Ws′)/N
wherein, WSIs the sea surface true wind speed, Ws' is the inverse wind speed, N is the sample pointThe total number of (c).
Examples
1. Data set acquisition
Obtaining 1 degree multiplied by 1 degree sea plane mode data of 2016 year, 1 month and 1 day to 12 months and 31 days from an intermediate weather forecast center (ECMWF), wherein the sea plane mode data comprises sea surface wind speed, sea surface temperature, sea surface wind direction, sea water salinity, cloud liquid water content, atmospheric water vapor content and other factors, and screening out a data set C containing 80740 groups of data. And C is randomly divided into an initial training set A and an initial verification set B, wherein the training set and the verification set respectively account for 80% and 20% of the initial data set C. Inputting A and B into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model, outputting simulated brightness temperature, and returning to A and B to obtain a training set A 'and a verification set B'.
2. Construction of deep learning convolutional neural network
In the step, a deep learning optimization method is adopted by a weight iteration optimizer algorithm, and the idea of the algorithm is to independently adapt to each parameter of the model: parameters with larger partial derivatives correspond to a larger learning rate and parameters with smaller partial derivatives correspond to a smaller learning rate, specifically, the learning rate of each parameter is scaled by the square root of the sum of the square values of the parameters inversely proportional to their historical gradients.
In the step, the output of the self-convolution layer is used as the input of the full-connection layer, and the sea surface wind speed in the training set A' is used as an ideal value output by the full-connection layer to carry out sea surface wind speed inversion.
In this step, the sea surface wind speed inversion mainly includes: after the trained convolutional network is used for eliminating the brightness temperature transmission noise and the instrument system noise, the trained data are input into a full-connection layer, and the data used for inverting the sea surface wind speed are as follows: the one-dimensional synthetic aperture microwave radiometer simulates brightness temperature, sea surface salinity, sea surface temperature, sea surface relative wind direction, atmospheric water content, cloud liquid water content and incident angle. The total number of 7 meteorological ocean elements is 7, so that 7 neurons and 1 bias are arranged in the batch normalization layer, the number of neurons in the output layer is 1 (sea surface wind speed), the neurons in each layer are not connected with each other, and the neurons in each layer are connected with each other through different weights. And then, according to a set activation function, transmitting backwards layer by layer, generating an error signal by comparing the expectation of the output data with the expectation of the input data, reversely transmitting the error signal layer by layer, and continuously correcting the weight of each layer to obtain the sea surface wind speed.
In this embodiment, the reverse transfer phase uses the Levenberg-Marquardt minimization algorithm to obtain the exact sea surface wind speed through a large amount of training.
3. Training of deep learning inversion models
And (3) taking the simulated brightness temperature, the sea surface temperature, the sea water salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content and the cloud liquid water content in the training set A 'as a training sample X' of the deep learning inversion model, and inputting the sea surface wind speed as a true value into the model, training the model, and finally outputting the sea surface wind speed. The weight iteration optimizer algorithm of the whole network adopts a deep learning optimization method, namely an AdaGrad algorithm, and sea surface wind speed is output through training.
4. Deep learning inversion model effect verification
And inputting the simulated brightness temperature and the simulated incidence angle in the verification set B' into the trained deep learning inversion model, calculating corresponding output, performing error comparison with the sea surface wind speed in the verification set, and checking the accuracy of the method.
The one-dimensional synthetic aperture microwave radiometer sea surface wind speed inversion method based on deep learning does not need to deeply research deep physical mechanisms in microwave transmitting and transmitting processes, has strong nonlinear fitting capacity, can invert sea surface wind speed with high precision, and provides technical support for load development and application of a follow-up satellite-borne one-dimensional synthetic aperture microwave radiometer.
The method breaks through the idea of the traditional real-aperture microwave radiometer, solves the problem that the sea surface wind speed is difficult to invert by multiple incidence angles of the one-dimensional synthetic aperture microwave radiometer by utilizing the characteristics of strong nonlinear fitting of deep learning and small calculation amount of a convolution network, and can rapidly invert the sea surface wind speed with high precision.
A one-dimensional synthetic aperture microwave radiometer sea surface wind speed inversion system based on deep learning, the system comprising:
a first obtaining module: the system is used for acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content;
a second obtaining module: the radiation transmission forward modeling system is used for inputting the sea surface temperature, the sea surface salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content and the cloud liquid water content into a radiation transmission forward modeling to obtain simulated bright temperature;
a third obtaining module: and the system is used for inputting the sea surface temperature, the sea surface salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on the convolutional neural network to obtain the sea surface wind speed.
A sea surface wind speed inversion system of a one-dimensional synthetic aperture microwave radiometer based on deep learning comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate according to the instructions to perform the steps of the method described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The present invention is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are included in the scope of the claims of the present invention which are filed as the application.
Claims (5)
1. A sea surface wind speed inversion method is characterized by comprising the following steps:
acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content to form inversion data;
inputting the inversion data into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model in sequence to obtain simulated brightness temperature;
inputting the sea surface temperature, the sea surface salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on a convolutional neural network to obtain the sea surface wind speed;
the convolution neural network of the deep learning inversion model has 10 layers in total, wherein the first layer is a batch normalization layer; the second layer is a convolution layer, the third layer is an average pooling layer, the fourth layer to the ninth layer are convolution layers, and the tenth layer is a full-connection layer;
the activation function of the second layer of the convolutional neural network is a tanh function; the activation functions of the fourth layer to the ninth layer of the convolutional neural network are Sigmoid functions;
the expression of the tanh function is:
the expression of the Sigmoid function is:
2. the sea surface wind speed inversion method of claim 1, wherein the deep learning inversion model is constructed by the following method:
acquiring historical inversion data and corresponding sea surface wind speed, and constructing an initial data set;
randomly dividing an initial data set into an initial training set and an initial verification set;
respectively and sequentially inputting historical inversion data in the initial training set and historical inversion data in the initial verification set into a radiation transmission forward modeling model and a one-dimensional synthetic aperture microwave radiometer model to obtain corresponding simulated bright temperature;
respectively putting the corresponding simulated brightness temperature back to the initial training set and the initial verification set;
and training the deep learning inversion model through the initial training set with the simulated bright temperature, and verifying the learning inversion model through the initial verification set with the simulated bright temperature to obtain the final deep learning inversion model.
3. A surface wind speed inversion system, the system comprising:
a first obtaining module: the system is used for acquiring sea surface temperature, sea water salinity, sea surface relative wind direction, incident angle, atmospheric water vapor content and cloud liquid water content to form inversion data;
a second obtaining module: the system is used for sequentially inputting inversion data into a radiation transmission forward model and a one-dimensional synthetic aperture microwave radiometer model to obtain simulated brightness temperature;
a third obtaining module: the system is used for inputting the sea surface temperature, the sea surface salinity, the sea surface relative wind direction, the incident angle, the atmospheric water vapor content, the cloud liquid water content and the simulated bright temperature into a deep learning inversion model constructed based on a convolutional neural network to obtain the sea surface wind speed;
the convolutional neural network of the deep learning inversion model has 10 layers in total, wherein the first layer is a batch normalization layer; the second layer is a convolution layer, the third layer is an average pooling layer, the fourth layer to the ninth layer are convolution layers, and the tenth layer is a full-connection layer;
the activation function of the second layer of the convolutional neural network is a tanh function; the activation functions of the fourth layer to the ninth layer of the convolutional neural network are Sigmoid functions;
the expression of the tanh function is:
the expression of the Sigmoid function is:
4. a sea surface wind speed inversion system, comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any of claims 1-2.
5. Computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011013771.6A CN112180369B (en) | 2020-09-23 | 2020-09-23 | Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011013771.6A CN112180369B (en) | 2020-09-23 | 2020-09-23 | Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112180369A CN112180369A (en) | 2021-01-05 |
| CN112180369B true CN112180369B (en) | 2021-09-10 |
Family
ID=73955392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011013771.6A Active CN112180369B (en) | 2020-09-23 | 2020-09-23 | Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112180369B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113156439B (en) * | 2021-05-08 | 2022-08-30 | 中国石油大学(华东) | SAR wind field and sea wave joint inversion method and system based on data driving |
| CN113408742B (en) * | 2021-06-24 | 2023-06-02 | 桂林理工大学 | High-precision sea surface temperature inversion method based on machine learning |
| CN114265087A (en) * | 2021-12-24 | 2022-04-01 | 青岛镭测创芯科技有限公司 | Sea wave parameter inversion method and related assembly |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB0822468D0 (en) * | 2008-12-10 | 2009-01-14 | Qinetiq Ltd | Method for mitigating the effects of clutter and interference on a radar system |
| CN108596380B (en) * | 2018-04-18 | 2022-11-08 | 中国科学院国家空间科学中心 | Quantitative detection method for sea surface typhoon wind field |
| CN109725316B (en) * | 2018-12-18 | 2020-10-20 | 中国人民解放军国防科技大学 | One-dimensional synthetic aperture microwave radiometer-based sea surface temperature physical inversion method |
| CN109829547B (en) * | 2018-12-18 | 2020-10-09 | 中国人民解放军国防科技大学 | Depth learning-based SST (stimulated Raman Scattering) inversion method for one-dimensional synthetic aperture microwave radiometer |
| CN109655828B (en) * | 2018-12-18 | 2020-06-26 | 中国人民解放军国防科技大学 | SST (stimulated Raman Scattering) depth learning inversion method for multi-frequency one-dimensional synthetic aperture microwave radiometer |
| CN111639746B (en) * | 2020-05-14 | 2023-10-20 | 中国科学院国家空间科学中心 | GNSS-R sea surface wind speed inversion method and system based on CNN neural network |
| CN111638521B (en) * | 2020-06-17 | 2021-09-24 | 中国人民解放军海军潜艇学院 | Sea surface wind speed inversion method of remote sensing data of imaging altimeter |
-
2020
- 2020-09-23 CN CN202011013771.6A patent/CN112180369B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112180369A (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112180369B (en) | Depth learning-based sea surface wind speed inversion method for one-dimensional synthetic aperture radiometer | |
| CN111639747B (en) | GNSS-R sea surface wind speed inversion method and system based on BP neural network | |
| CN109523013B (en) | Air particulate matter pollution degree estimation method based on shallow convolutional neural network | |
| CN111639746B (en) | GNSS-R sea surface wind speed inversion method and system based on CNN neural network | |
| CN111161218A (en) | High-resolution remote sensing image change detection method based on twin convolutional neural network | |
| CN110427818B (en) | Deep learning satellite data cloud detection method supported by hyperspectral data | |
| CN113568055B (en) | Aviation transient electromagnetic data inversion method based on LSTM network | |
| CN113568056B (en) | Aviation transient electromagnetic data inversion method based on convolutional neural network | |
| CN113255972B (en) | Short-term rainfall prediction method based on Attention mechanism | |
| CN109583412A (en) | A kind of training method and its ship detecting method carrying out ship detecting using convolutional neural networks | |
| CN114117886A (en) | Water depth inversion method for multispectral remote sensing | |
| CN109829547B (en) | Depth learning-based SST (stimulated Raman Scattering) inversion method for one-dimensional synthetic aperture microwave radiometer | |
| CN116977311B (en) | Flood disaster area detection method, system, electronic equipment and storage medium | |
| CN115545334A (en) | Land use type prediction method, land use type prediction device, electronic device, and storage medium | |
| CN115859789A (en) | Method for improving inversion accuracy of polar atmosphere temperature profile | |
| CN116879192B (en) | Water bloom prediction method, device, equipment and medium based on satellite remote sensing data | |
| CN116628442B (en) | Groundwater reserve change space downscaling method based on artificial neural network | |
| CN117371303A (en) | Prediction method for effective wave height under sea wave | |
| CN114169646B (en) | Water bloom prediction method, device, electronic equipment and computer readable storage medium | |
| CN117037450A (en) | Environment early warning method, device, equipment and medium of intelligent piling ship | |
| CN115062526B (en) | Three-dimensional ionosphere electron concentration distribution model training method based on deep learning | |
| CN110852415A (en) | Vegetation index prediction method, system and equipment based on neural network algorithm | |
| Turgeon-Pelchat et al. | Deep Learning-Based Classification of Large-Scale Airborne LiDAR Point Cloud | |
| CN110851789B (en) | Island reef shallow sea water depth prediction method based on extreme gradient lifting | |
| CN113657416B (en) | Deep sea sound source ranging method and system based on improved deep neural network |
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 |