CN108875905B - Direct inversion method for visibility function of atmospheric temperature and humidity profile - Google Patents
Direct inversion method for visibility function of atmospheric temperature and humidity profile Download PDFInfo
- Publication number
- CN108875905B CN108875905B CN201810323166.5A CN201810323166A CN108875905B CN 108875905 B CN108875905 B CN 108875905B CN 201810323166 A CN201810323166 A CN 201810323166A CN 108875905 B CN108875905 B CN 108875905B
- Authority
- CN
- China
- Prior art keywords
- physical parameters
- atmospheric
- sample
- neural network
- visibility function
- 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
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
-
- 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
- G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
-
- 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
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a direct inversion method of a visibility function of an atmospheric temperature and humidity profile, which comprises the steps of inputting the actually observed visibility function into a trained BP neural network to obtain inverted physical parameters of the atmospheric temperature and humidity profile; the training of the BP neural network comprises the following steps: obtaining sample atmospheric physical parameters by using a numerical weather mode, and performing forward modeling on the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain a visibility function corresponding to the sample atmospheric physical parameters; and training the BP neural network by using the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters to obtain the trained BP neural network. The invention changes the two-step inversion method of visibility-brightness temperature-atmospheric physical parameters required by the prior remote sensing atmospheric physical parameters of the synthetic aperture microwave radiometer, provides a direct inversion method of the visibility-atmospheric physical parameters, simplifies the data processing flow and reduces the inversion error, thereby obtaining more accurate atmospheric physical parameters.
Description
Technical Field
The invention belongs to the technical field of atmospheric microwave remote sensing and detection, and particularly relates to a direct inversion method of a visibility function of an atmospheric temperature and humidity profile.
Background
Atmospheric parameters such as temperature and humidity profile, water vapor density and pressure are important parameters in the fields of climate monitoring and weather forecast. Besides, the atmospheric parameters also have important roles in the fields of artificial influence on weather, flood control command and decision and the like.
In the current climate and weather research, satellite-borne remote sensing observation is an important means for acquiring atmospheric parameter information because of the advantages of high economic benefit, wide detection area coverage, high horizontal resolution, continuous observation, global intensive observation and the like. The satellite-borne microwave radiometer has unique advantages in atmospheric remote sensing because microwaves can penetrate through clouds, haze, dust and the like in the atmosphere and are not influenced by solar radiation. The satellite-borne synthetic aperture microwave radiometer uses a large number of sparse small-aperture antenna arrays to synthesize a large observation aperture, so that the size and the weight of the antenna are greatly reduced, and the contradiction between the aperture of the large antenna and the high resolution is effectively relieved. And the synthetic aperture microwave radiometer does not need scanning for imaging, and the imaging time is theoretically equal to the integration time, so that the contradiction between sensitivity, resolution and imaging time can be relieved to a considerable extent, and the application requirement of high-speed imaging is met.
However, the synthetic aperture microwave radiometer obtains the visibility function sampled in the spatial frequency domain by interferometry. And information such as atmospheric temperature, humidity, cloud water parameters and the like is extracted from brightness temperature information of ground-air system radiation detected by the satellite-borne microwave radiometer. Therefore, the synthetic aperture microwave radiometer obtains the atmospheric parameter information, generally, the fourier transform relationship between the brightness temperature distribution and the visibility function is utilized, the observed brightness temperature is obtained through inverse fourier transform or other mathematical operations, and then the information such as the atmospheric temperature, the humidity, the cloud water parameters and the like is extracted from the brightness temperature information through an inversion method. Therefore, the existing synthetic aperture radiometer needs a two-step inversion process of visibility, brightness temperature and atmospheric physical parameters when acquiring atmospheric parameters, the ill-conditioned equation needs to be solved for two inversions, the input data error is amplified by the approximate solution obtained each time, and additional inversion errors are introduced.
Therefore, in the prior art, the synthetic aperture microwave radiometer needs to perform two inversion processes to obtain the atmospheric physical parameters, so that the error of input data is amplified, and additional inversion errors are introduced.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a direct inversion method of a visibility function of an atmospheric temperature and humidity profile, so that the technical problems that the error of input data is amplified and additional inversion errors are introduced due to the fact that a synthetic aperture microwave radiometer needs to perform two-step inversion processes to obtain atmospheric physical parameters in the prior art are solved.
In order to achieve the above object, the present invention provides a method for directly inverting a visibility function of an atmospheric temperature and humidity profile, comprising:
inputting the actually observed visibility function into the trained BP neural network to obtain inverted atmospheric temperature and humidity profile physical parameters;
the training of the BP neural network comprises:
(1) obtaining sample atmospheric physical parameters by using a numerical weather mode, and performing forward modeling on the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain a visibility function corresponding to the sample atmospheric physical parameters;
(2) and training the BP neural network by using the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters to obtain the trained BP neural network.
Further, the step (1) comprises:
(1-1) obtaining sample atmospheric physical parameters in a field of view by using a numerical weather mode, and performing forward modeling on the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain a forward-modeling brightness temperature image in the field of view;
and (1-2) carrying out Fourier transform on the forward evolution bright temperature image in the visual field to obtain a sample visibility function corresponding to the sample atmospheric physical parameters on the spatial frequency domain.
Further, the step (2) comprises:
(2-1) setting the training times of the BP neural network by taking the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters as a sample set;
and (2-2) training the BP neural network by using the sample set, and correcting each layer of parameters of the BP neural network according to back propagation every time of training until the training times are reached or the training is continued to generate a negative optimization effect, so as to obtain the trained BP neural network.
Further, a quantitative conjugate gradient method is adopted when the BP neural network is trained by utilizing the sample set.
Further, the real part and the imaginary part of the visibility function in the sample set are respectively used as two variables in a set of input, so that the BP neural network is trained by using complex numbers.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) the method realizes the direct inversion method of the visibility-atmospheric physical parameters, simplifies the data processing flow and reduces the inversion error, thereby obtaining more accurate atmospheric physical parameters. Therefore, the technical problems that in the prior art, the synthetic aperture microwave radiometer needs to obtain atmospheric physical parameters through two inversion processes, so that the error of input data is amplified, and extra inversion errors are introduced are solved.
(2) The invention trains the BP neural network by taking the visibility function as input data, and the visibility function is complex. Usually, the training BP neural network can only accept real numbers as input data, and aiming at a special application scene of inversion of a visibility function-atmospheric physical parameter, the real parts and the imaginary parts of the visibility function are respectively used as two variables in a group of inputs, so that the training of the BP neural network by using complex input data is realized.
(3) The common training algorithm of the bright temperature-atmospheric physical parameter inversion BP neural network is a learning rate-variable momentum gradient descent algorithm, an RPROP (elastic BP) algorithm and a Bayesian regularization algorithm. In the invention, a quantized Conjugate Gradient (SCG) method is adopted when the sample set is used for training the BP neural network, so that the algorithm calculation amount in the training process is smaller and the training speed is high.
Drawings
FIG. 1 is a flow chart of a method for directly inverting a visibility function of an atmospheric temperature and humidity profile according to an embodiment of the present invention;
FIG. 2 is a graph of the RMSE for the direct inversion of the visibility function versus atmospheric temperature provided by example 1 of the present invention;
FIG. 3 is a graph of the RMSE for a conventional two-step inversion provided in example 1 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
As shown in fig. 1, a method for directly inverting a visibility function of an atmospheric temperature and humidity profile includes:
inputting the actually observed visibility function into the trained BP neural network to obtain inverted atmospheric temperature and humidity profile physical parameters;
the training of the BP neural network comprises:
(1) obtaining sample atmospheric physical parameters in a field of view by using a numerical weather mode, and forward modeling the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain forward-modeled brightness temperature images in the field of view; and carrying out Fourier transform on the forward bright temperature image in the visual field to obtain a sample visibility function corresponding to the atmospheric physical parameters of the sample in the spatial frequency domain.
(2) And taking the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters as a sample set, and respectively taking the real part and the imaginary part of the visibility function in the sample set as two variables in a group of inputs, thereby realizing the complex number training of the BP neural network. Setting the training times of the BP neural network; and training the BP neural network by using the sample set by adopting a quantitative conjugate gradient method, and correcting each layer of parameters of the BP neural network according to back propagation every time the BP neural network is trained until the training times are reached or the training is continued to generate a negative optimization effect, thereby obtaining the trained BP neural network.
Example 1
The direct inversion method of the visibility function of the atmospheric temperature and humidity profile, provided by the invention, aims at the real-time application of microwave remote sensing temperature and humidity profile reconstruction, can effectively simplify the data processing flow, reduce the calculation complexity and reduce the inversion error, and is a novel inversion method of the atmospheric temperature and humidity profile. Example 1 takes the atmospheric temperature and humidity profile of the sea area of 11/12 th in 2017, 150-.
In the first step, a 25-element one-dimensional sparse synthetic aperture array is selected to simulate a satellite-borne one-dimensional synthetic aperture microwave radiometer, and the array can generate 137 baselines in the spatial frequency domain.
And secondly, forward generating 8 channels of light temperature images TB with the frequency of 50.3-57.29GHz by NWP based on a microwave radiation transmission RT mode, and inputting the 8 channels of light temperature images TB into a forward observation model of the synthetic aperture radiometer to simulate and calculate a visibility function V of the actually observed 8 frequencies. The final total of 2700 sets of samples results in 2700 of 2700 × 8 × 137 visibility functions V. Taking the real and imaginary parts of the visibility function separately, each set of input samples is 8 × 137 × 2 in size.
And thirdly, obtaining a corresponding atmospheric temperature profile in a WRF (Weather Research and Weather modeling) mode, wherein each 100 pixel points are a group of samples, each sample has 40 layers corresponding to different areas on the space and different heights, and 2700 × 40 × 100 atmospheric temperature values are generated in total.
And fourthly, setting the number of hidden layers of the BP neural network to be trained to be 1 and the number of hidden layer nodes to be 25. The supervised training algorithm adopts a quantized Conjugate Gradient (SCG) method, and is based on the optimization algorithm, the calculation amount is small, the maximum training frequency is set to be 3000 times, and the maximum failure frequency is 15 times. A 2700 sets of atmospheric temperature samples and visibility function samples were obtained, 90% as training set and 10% as test set.
And fifthly, inputting 8 × 137 visibility function samples of each visibility function into the BP neural network, and obtaining 40 × 100 atmospheric temperature values corresponding to the corresponding temperature samples. And each training time, correcting parameters of each layer of the network according to the back propagation until a preset training time is reached or the training is continued to generate a negative optimization effect.
And sixthly, verifying the effectiveness of the trained network by using the test group samples, calculating the inversion mean square error of 300 groups of test samples, and averaging the atmospheric temperature inversion mean square errors of each layer of height to obtain an atmospheric temperature profile inversion error RMSE (RMSE), as shown in figure 2.
And seventhly, selecting the atmospheric parameters in the same time and the same area, solving the atmospheric temperature profile by using the traditional two-step inversion process, and solving the inversion error RMSE after the atmospheric temperature profile is averaged, as shown in FIG. 3.
Eighth, the RMSE obtained by the direct inversion of the present invention is compared with the RMSE obtained by the conventional two-step inversion, as shown in fig. 2 and 3.
From the above results, it can be seen that: the inversion error obtained by adopting the new visibility-atmospheric temperature direct inversion method is smaller than that generated by the traditional two-step inversion process of visibility-bright temperature-atmospheric physical parameters in most atmospheric layers. The method not only effectively simplifies the data processing process, reduces the calculated amount, but also greatly improves the inversion precision of the atmospheric temperature profile.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (5)
1. A method for directly inverting a visibility function of an atmospheric temperature and humidity profile is characterized by comprising the following steps:
inputting the actually observed visibility function into the trained BP neural network to obtain inverted atmospheric temperature and humidity profile physical parameters;
the training of the BP neural network comprises:
(1) obtaining sample atmospheric physical parameters by using a numerical weather mode, and performing forward modeling on the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain a visibility function corresponding to the sample atmospheric physical parameters;
(2) and training the BP neural network by using the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters to obtain the trained BP neural network.
2. The direct inversion method of the visibility function of the atmospheric temperature and humidity profile as claimed in claim 1, wherein the step (1) comprises:
(1-1) obtaining sample atmospheric physical parameters in a field of view by using a numerical weather mode, and performing forward modeling on the sample atmospheric physical parameters by using a synthetic aperture microwave radiometer model to obtain a forward-modeling brightness temperature image in the field of view;
and (1-2) carrying out Fourier transform on the forward evolution bright temperature image in the visual field to obtain a sample visibility function corresponding to the sample atmospheric physical parameters on the spatial frequency domain.
3. The direct inversion method of the visibility function of the atmospheric temperature and humidity profile according to claim 1 or 2, wherein the step (2) comprises:
(2-1) setting the training times of the BP neural network by taking the sample atmospheric physical parameters and the visibility function corresponding to the sample atmospheric physical parameters as a sample set;
and (2-2) training the BP neural network by using the sample set, and correcting each layer of parameters of the BP neural network according to back propagation every time of training until the training times are reached or the training is continued to generate a negative optimization effect, so as to obtain the trained BP neural network.
4. The method for directly inverting the visibility function of the atmospheric temperature and humidity profile as claimed in claim 3, wherein a quantized conjugate gradient method is adopted when the BP neural network is trained by using the sample set.
5. The method of claim 3, wherein the real part and the imaginary part of the visibility function in the sample set are respectively used as two variables in a set of input, thereby realizing complex training of the BP neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810323166.5A CN108875905B (en) | 2018-04-09 | 2018-04-09 | Direct inversion method for visibility function of atmospheric temperature and humidity profile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810323166.5A CN108875905B (en) | 2018-04-09 | 2018-04-09 | Direct inversion method for visibility function of atmospheric temperature and humidity profile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108875905A CN108875905A (en) | 2018-11-23 |
| CN108875905B true CN108875905B (en) | 2021-04-20 |
Family
ID=64326284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810323166.5A Active CN108875905B (en) | 2018-04-09 | 2018-04-09 | Direct inversion method for visibility function of atmospheric temperature and humidity profile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108875905B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109725317B (en) * | 2018-12-18 | 2021-06-01 | 中国人民解放军国防科技大学 | Sea surface bright temperature imaging simulation method based on one-dimensional synthetic aperture microwave radiometer |
| 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 |
| CN110633650A (en) * | 2019-08-22 | 2019-12-31 | 首都师范大学 | Convolutional neural network face recognition method and device based on privacy protection |
| CN110632599A (en) * | 2019-09-03 | 2019-12-31 | 华中科技大学 | Atmospheric temperature profile direct inversion method and system |
| CN111126591B (en) * | 2019-10-11 | 2023-04-18 | 重庆大学 | Magnetotelluric deep neural network inversion method based on space constraint technology |
| CN110826693B (en) * | 2019-10-29 | 2022-10-14 | 华中科技大学 | Three-dimensional atmospheric temperature profile inversion method and system based on DenseNet convolutional neural network |
| CN111651934B (en) * | 2020-05-25 | 2024-03-22 | 华中科技大学 | Ice cloud profile inversion method |
| CN112197865B (en) * | 2020-09-02 | 2021-08-20 | 华中科技大学 | Estimation method and system for observation brightness temperature data error of satellite-borne microwave radiometer |
| CN115270068B (en) * | 2022-09-27 | 2023-01-06 | 山东省科学院海洋仪器仪表研究所 | Method for quickly estimating atmospheric moisture delay of offshore inclined path based on buoy platform |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504256A (en) * | 2014-12-12 | 2015-04-08 | 中国电子科技集团公司第二十二研究所 | Estimation algorithm for accurately inverting boundary layer temperature profile |
| CN105652284A (en) * | 2016-01-05 | 2016-06-08 | 中国科学院遥感与数字地球研究所 | Determination method and device of long-wave downward radiation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9339785B2 (en) * | 2013-12-18 | 2016-05-17 | Battelle Memorial Institute | Methods and systems for acoustically-assisted hydroprocessing at low pressure |
-
2018
- 2018-04-09 CN CN201810323166.5A patent/CN108875905B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504256A (en) * | 2014-12-12 | 2015-04-08 | 中国电子科技集团公司第二十二研究所 | Estimation algorithm for accurately inverting boundary layer temperature profile |
| CN105652284A (en) * | 2016-01-05 | 2016-06-08 | 中国科学院遥感与数字地球研究所 | Determination method and device of long-wave downward radiation |
Non-Patent Citations (2)
| Title |
|---|
| 《Microwave Remote Sensing of The Temperature and Distribution of Sulfur Compounds in the Lower Atmosphere of Venus》;Jon M.Jenkins 等;《Icarus》;20020831;第158卷(第2期);全文 * |
| 《云对地基微波辐射计反演湿度廓线的影响》;车云飞 等;《应用气象学报》;20150331;第26卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108875905A (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108875905B (en) | Direct inversion method for visibility function of atmospheric temperature and humidity profile | |
| Bosch et al. | Daily solar irradiation estimation over a mountainous area using artificial neural networks | |
| CN109543721B (en) | Solar irradiance ultra-short-term forecasting method under sunny cloudy weather condition | |
| CN108009989A (en) | A kind of microwave remote sensing image super-resolution rebuilding method based on SRCNN | |
| CN109472393B (en) | Spatial downscaling precipitation data detection method and device and electronic equipment | |
| CN109557540B (en) | Total variation regularization correlation imaging method based on non-negative constraint of target scattering coefficient | |
| CN110221360A (en) | A kind of power circuit thunderstorm method for early warning and system | |
| CN111210483B (en) | Simulated satellite cloud picture generation method based on generation of countermeasure network and numerical mode product | |
| CN111610524B (en) | Ice cloud profile inversion method and system based on one-dimensional variational algorithm | |
| Zuo et al. | Ten-minute prediction of solar irradiance based on cloud detection and a long short-term memory (LSTM) model | |
| CN110632599A (en) | Atmospheric temperature profile direct inversion method and system | |
| CN111781598B (en) | DSN-based high-resolution two-dimensional ISAR imaging method | |
| CN113009490A (en) | Radar three-dimensional wind field inversion method based on high-resolution mode dynamic constraint | |
| CN116148796A (en) | Strong convection weather proximity forecasting method based on radar image extrapolation | |
| CN109785242A (en) | A kind of solution mixing method based on the high spectrum image by wave band generalized bilinear model | |
| Sun et al. | Convective initiation nowcasting over China from Fengyun-4A measurements based on TV-L 1 optical flow and BP_Adaboost neural network algorithms | |
| CN105118035A (en) | Self-adaptive optical spot signal extraction method based on sparse representation | |
| CN110969609B (en) | Microwave remote sensing bright temperature image cloud detection method | |
| CN115952743B (en) | Multi-source precipitation data collaborative downscaling method and system coupled with random forest and HASM | |
| CN112668615B (en) | Satellite cloud picture prediction method based on depth cross-scale extrapolation fusion | |
| Ling et al. | Automatic detection of sunspots and extractionof their feature parameters | |
| Mao et al. | Deep-learning-based flying animals migration prediction with weather radar network | |
| CN117148360B (en) | Lightning approach prediction method and device, electronic equipment and computer storage medium | |
| CN117437230B (en) | Photovoltaic power station power prediction method and system based on image restoration | |
| Mahfouz et al. | Conditional generation of cloud fields |
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 |