CN113034363B - GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method - Google Patents
GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method Download PDFInfo
- Publication number
- CN113034363B CN113034363B CN202110263381.2A CN202110263381A CN113034363B CN 113034363 B CN113034363 B CN 113034363B CN 202110263381 A CN202110263381 A CN 202110263381A CN 113034363 B CN113034363 B CN 113034363B
- Authority
- CN
- China
- Prior art keywords
- scale
- data
- image
- gee
- nitrogen oxide
- 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
- MWUXSHHQAYIFBG-UHFFFAOYSA-N Nitric oxide Chemical compound O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000009467 reduction Effects 0.000 title description 3
- 238000012544 monitoring process Methods 0.000 claims abstract description 9
- 238000013528 artificial neural network Methods 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims abstract description 8
- 230000008859 change Effects 0.000 claims abstract description 7
- 230000015556 catabolic process Effects 0.000 claims description 8
- 238000006731 degradation reaction Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 2
- 210000002569 neuron Anatomy 0.000 claims description 2
- 238000013519 translation Methods 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 claims 1
- 230000006870 function Effects 0.000 abstract description 15
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000000605 extraction Methods 0.000 abstract 1
- 230000008569 process Effects 0.000 description 3
- 238000003915 air pollution Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- MGWGWNFMUOTEHG-UHFFFAOYSA-N 4-(3,5-dimethylphenyl)-1,3-thiazol-2-amine Chemical compound CC1=CC(C)=CC(C=2N=C(N)SC=2)=C1 MGWGWNFMUOTEHG-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002803 fossil fuel Substances 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 230000000813 microbial effect Effects 0.000 description 1
- JCXJVPUVTGWSNB-UHFFFAOYSA-N nitrogen dioxide Inorganic materials O=[N]=O JCXJVPUVTGWSNB-UHFFFAOYSA-N 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000005437 stratosphere Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000005436 troposphere Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4053—Super resolution, i.e. output image resolution higher than sensor resolution
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4046—Scaling the whole image or part thereof using neural networks
Abstract
The application provides a GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid-dropping method, which combines the existing satellite image and a geospatial dataset catalog with a planet scale analysis function and stores the combined functions in a cloud, sets an extraction patch density, uses a pixel to represent a unit, expands a high-resolution frame according to an expansion proportion, distributes the unit to the center of the expanded high-resolution frame, fills a scale space, carries out experience change on the unit to obtain an expanded image, collects nitrogen oxide detection point data, inputs the expanded image and the monitoring point data into a neural network for training, obtains a high-resolution space-time information nitrogen oxide image, and solves the problem that the nitrogen oxide data scale of the existing satellite detection is inaccurate.
Description
Technical Field
The application relates to the technical field of atmospheric remote sensing, in particular to a nitrogen oxide rapid reduction method based on GEE depth space-time experience Kringing region scale.
Background
Nitrogen dioxide NO 2 And nitrogen oxides are commonly referred to collectively as nitrogen oxides NO X . Nitrogen oxides are important trace gases in the earth's atmosphere, present in the troposphere and the stratosphere. Human activity and natural processes may cause nitrogen oxides to enter the atmosphere, such as the combustion of fossil fuels or microbial decomposition processes in the soil, and the generated nitrogen oxides may enter the atmosphere. Conversion of nitrogen oxides to NO under photochemical action 2 Thus NO 2 Is a powerful indicator for measuring the concentration of nitrogen oxides in the atmosphere.
Aiming at nitrogen oxides, the current mainstream detection method is to detect through an optical remote sensing satellite sentinel No. 5 satellite, but the resolution of the current proposed product is 1 degree multiplied by 1 degree, the expected high-scale resolution is far different, the actual scale is reduced, and the data is inaccurate.
Disclosure of Invention
The application provides a nitrogen oxide rapid degradation method based on GEE depth space-time experience Kringing region scale, which aims to solve the problem that the data scale of nitrogen oxide detected by the existing satellite is inaccurate.
A nitrogen oxide rapid degradation method based on GEE depth space-time experience Kringing region scale comprises the following steps:
the GEE platform combines the existing satellite images and the geospatial dataset catalogue with the analysis function of the planet scale and stores the combined images and the geospatial dataset catalogue in a cloud, and the cloud invokes the stored data;
extracting patches from the stored data and setting each pixel of the patches as a unit;
setting a resolution frame, expanding the resolution frame according to an expansion proportion to obtain a high resolution frame, and distributing the units to the center of the high resolution frame;
experience change is carried out on the units to obtain expanded images;
and collecting nitrogen oxide monitoring point data, and inputting the expanded image and the monitoring point data into a neural network for training to obtain a high-resolution space-time information nitrogen oxide image.
Optionally, the unit is subjected to an empirical variation formula of:
γ(h)＝Nugget+b|h| a
γ(h)＝Nugget+b|h|
γ(h)＝Nugget+b|h| a *ln(|h|)
obtaining a mean error function:
optionally, combining the existing satellite image and geospatial dataset catalog with the planet-scale analysis function and storing in the cloud step at the GEE platform, further comprising: and the GEE platform performs code translation and data operation on the data stored in the platform.
Optionally, the filling scale space comprises:
cross-dividing the units subjected to the experience change to obtain overlapping samples;
and carrying out semi-variation function estimation on the overlapped samples.
Optionally, the training of the neural network includes:
the convolution layer obtains the regional information through local calculation and transmits the regional information to the next layer of neurons.
Optionally, the step of extracting the patch is:
extracting an image from low resolution to high resolution under the same longitude;
and setting the patch as a high latitude vector to generate a group of characteristic diagrams.
Optionally, the step of determining the scale of extracting the unit is:
selecting a group of image data, and setting the number of images in the data as N;
the scale of extracting the unit is N x 1.
According to the technical scheme, the method for quickly reducing the nitrogen oxides based on the GEE depth space-time experience Kringing region scale is provided, the existing satellite image and the geospatial dataset catalog are combined with the analysis function of the planet scale through the GEE platform and stored in the cloud, the density of the extracted patches is set, one pixel is used for representing one unit, a high-resolution frame is expanded according to the expansion proportion, the unit is distributed to the center of the expanded high-resolution frame, the scale space is filled, the unit is subjected to empirical change, the expanded image is obtained, nitrogen oxide detection point data are collected, the expanded image and the monitoring point data are input into a neural network for training, the high-resolution space-time information nitrogen oxide image is obtained, and the problem that the nitrogen oxide data scale of the existing satellite detection is inaccurate is solved.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flow chart of a method for rapid degradation of nitrogen oxides based on GEE depth spatiotemporal experience Kringing region scale.
Detailed Description
Reference will now be made in detail to the embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The embodiments described in the examples below do not represent all embodiments consistent with the present application. Merely as examples of systems and methods consistent with some aspects of the present application as detailed in the claims.
In the technical scheme provided by the application, the method for quickly reducing the nitrogen oxide based on the GEE depth space-time experience Kringing region scale is provided, so that the problem of inaccurate data scale of nitrogen oxide detected by the existing satellite is solved. Referring to fig. 1, the method for rapidly reducing nitrogen oxides comprises the following steps:
s1: the GEE platform combines the existing satellite images and the geospatial dataset catalogue with the analysis function of the planet scale and stores the combined images and the geospatial dataset catalogue in a cloud, and cloud data are called through the cloud.
GEE (generic name Google Earth Engine) is a tool for processing satellite image data in batches, belonging to Google Earth series. Compared with the traditional image processing tools such as ENVI, the GEE can rapidly process huge images in batches. The vegetation indexes such as NDVI can be rapidly calculated through GEE, the crop related yield can be predicted, the drought growth variation can be monitored, the global forest variation can be monitored, and the like.
Kriging (Kriging) is to assign different weights to each sample position according to different spatial positions of the samples and different correlation degrees among the samples, and perform a sliding weighted average to estimate the average grade of the central block.
And calling the cloud-stored data by using a GEE platform and calling a code entry platform transferred by the following logic, wherein the GEE combines the existing satellite image and geospatial dataset catalog with the planet-scale analysis function and stores the combined data and the planet-scale analysis function in the cloud. The storage mode can flexibly call the data, and the cloud data updating speed greatly exceeds that of a single storage device.
S2: extracting a patch from the stored data, and setting each pixel of the patch as a unit.
This operation extracts (necessarily at the same latitude and longitude) the image Y (in years) from the low resolution and represents each patch as a high-dimensional vector. These vectors include a set of feature maps, the number of which corresponds to the dimensions of the vector, using densely extracted patches, with a set of age data, with each pixel as a unit. For example: the data of the age is one image per month, namely 12 images in total; then the scale of the data we extract as a unit for each pixel is then 12 x 1.
S3: setting a resolution frame, expanding the resolution frame according to the expansion proportion to obtain a high resolution frame, and distributing the units to the center of the high resolution frame.
The patches are extended, each group of patches in years, containing a total of 12 months of data (taking the month average product as an example), by extending pixels of a specified proportion, i.e. odd coefficients of 9, 25, 49, etc., outwards, and by extending a high resolution frame, the image scale is extended on the size of the original input image, with the proportion of the extension being Sc (this parameter is typically chosen to be 9, 25, 49, 81, etc.). Each pixel of the original incoming low resolution image is then assigned to the center of the expanded frame. The filling of the expansion unit in the expanded image is completed by utilizing adjacent yellow elements by means of the algorithm of the next step.
S4: and performing empirical variation on the units to obtain the expanded image.
The operation combines the patch groups of the current unit and the peripheral unit to carry out nonlinear mapping, and the mapping considers the space factor and greatly restores the real and objective world. In the last step, the originally entered cells are allocated into the expanded frame. In this step, filling of the spatial scale is done first, we first perform empirical transformation on the data, and then cross-divide the large amount of data into overlapping large amounts of small samples (all at 100 points). Then, performing a large number of times of semi-variation function estimation on each small sample, wherein the estimated times are usually set to be 100 times; because the number of times of estimating the same sample is overlarge, a large number of curves estimated by different functions can be generated, therefore, a curve formula with the largest number of times of curve overlapping is selected as the estimation of the current sample, and the iteration is repeated, and finally, the image after global expansion is acquired.
Not only the correlation of the position of the point to be estimated with the known data position is taken into account, but also the spatial correlation of the variables.
S5: and collecting nitrogen oxide monitoring point data, and inputting the expanded image and the monitoring point data into a neural network for training to obtain a high-resolution space-time information nitrogen oxide image.
The operation carries out deep learning iteration on the real-time site data corresponding to the patch and the ground, and the final high-resolution nitrogen oxide data is predicted by using the trained model, so that the step successfully integrates time sequence information, and the expanded high-resolution nitrogen oxide image is more similar to real data.
Further, in step 4, the following formula is involved:
γ(h)＝Nugget+b|h| a (1)
γ(h)＝Nugget+b|h| (2)
γ(h)＝Nugget+b|h| a *ln(|h|) (3)
further defined is a mean error function MSE as this loss function:
where F (Y_i; θ) is the reconstructed high resolution image and X is the truth image on this method. The convolution parameter is θ= { W 1 ，W 2 ，W 3 ，B 1 ，B 2 ，B 3 }。
Compared with the prior art, the invention has the advantages and beneficial effects that: the deep space-time experience kriging method based on the GEE platform provided by the invention not only solves the problems of insufficient local storage and calculation force, but also has better precision and stronger stability of the high-resolution image product generated by the method, and more difficultly, the method completely considers space-time information, the time information is integrated, the reliability of the data after expansion is greatly improved, the problem of insufficient resolution of nitrogen oxides on a small scale is also greatly reduced, and more accurate near-real-time data is provided for researching air pollution.
According to the technical scheme, the method for quickly reducing the nitrogen oxides based on the GEE depth space-time experience Kringing region scale is provided, the existing satellite image and the geospatial dataset catalog are combined with the analysis function of the planet scale and stored in the cloud end through the GEE platform, the patch is extracted, one pixel is used for representing one unit, a high-resolution frame is expanded according to the expansion proportion, the unit is distributed to the center of the high-resolution frame, the scale space is filled, the unit is subjected to experience change, the expanded image is obtained, nitrogen oxide detection point data are collected, the expanded image and the monitoring point data are input into a neural network for training, the high-resolution space-time information nitrogen oxide image is obtained, and the problem that the nitrogen oxide data scale of the existing satellite detection is inaccurate is solved.
Compared with the prior art, the application point and beneficial effect: the deep space-time experience kriging method based on the GEE platform provided by the invention not only solves the problems of insufficient local storage and calculation force, but also has better precision and stronger stability of the high-resolution image product generated by the method, and more difficultly, the method completely considers space-time information, the time information is integrated, the reliability of the data after expansion is greatly improved, the problem of insufficient resolution of nitrogen oxides on a small scale is also greatly reduced, and more accurate near-real-time data is provided for researching air pollution.
The foregoing detailed description of the embodiments is merely illustrative of the general principles of the present application and should not be taken in any way as limiting the scope of the invention. Any other embodiments developed in accordance with the present application without inventive effort are within the scope of the present application for those skilled in the art.
Claims (5)
1. The nitrogen oxide rapid degradation method based on GEE depth space-time experience Kringing region scale is characterized by comprising the following steps:
the GEE platform combines the existing satellite images and the geospatial dataset catalogue with the analysis function of the planet scale and stores the combined images and the geospatial dataset catalogue in a cloud, and cloud data is called through the cloud;
extracting a patch from the cloud data, and setting each pixel of the patch as a unit;
setting a resolution frame, expanding the resolution frame according to an expansion ratio to obtain a high resolution frame, and distributing the units to the center of the high resolution frame;
experience change is carried out on the units to obtain expanded images;
collecting nitrogen oxide monitoring point data, inputting the expanded image and the monitoring point data into a neural network for training to obtain a high-resolution time-space information nitrogen oxide image;
experience changes are performed on the units, and the expanded image is obtained by the steps of:
firstly, filling the space scale, and performing experience change on the data; cross-partitioning the data into overlapping first samples, the first samples being 100 data points; performing a semi-variational function estimation on each first sample, wherein the number of estimation times is set to 100 times; selecting a curve formula with the largest curve overlapping times as the estimation of the current sample, and repeatedly iterating to obtain an image after global expansion;
wherein the empirical variation formula for the unit is:
γ(h)＝Nugget+b|h| a ，
γ(h)＝Nugget+b|h|，
γ(h)＝Nugget+b|h| a *ln(|h|)，
obtaining a mean error function:
wherein F (Y) i The method comprises the steps of carrying out a first treatment on the surface of the θ) is the reconstructed high resolution image, X is the truth image of the method, and the convolution parameters are θ= { W 1 ，W 2 ，W 3 ，B 1 ，B 2 ，B 3 }。
2. The GEE depth spatiotemporal experience Kringing region scale-based nox rapid degradation method of claim 1, wherein the GEE platform combines existing satellite images and geospatial dataset catalogues with a planet scale analysis function and stores in cloud step, further comprising: and the GEE platform performs code translation and data operation on the data stored in the platform.
3. The GEE depth spatiotemporal experience Kringing region scale based nox rapid degradation method of claim 1, wherein the neural network training comprises:
the convolution layer obtains the regional information through local calculation and transmits the regional information to the next layer of neurons.
4. The get depth spatiotemporal experience Kringing region scale based nox rapid degradation method of claim 1, wherein the step of extracting patches is:
extracting an image from low resolution to high resolution under the same longitude;
and setting the patch as a high latitude vector to generate a group of characteristic diagrams.
5. The method for rapid degradation of nitrogen oxides based on GEE depth spatiotemporal experience Kringing region scale according to claim 1, wherein the step of determining the scale of extracting the units is:
selecting a group of image data, and setting the number of images in the data as N;
the scale of extracting the unit is N x 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110263381.2A CN113034363B (en) | 2021-03-11 | 2021-03-11 | GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110263381.2A CN113034363B (en) | 2021-03-11 | 2021-03-11 | GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113034363A CN113034363A (en) | 2021-06-25 |
| CN113034363B true CN113034363B (en) | 2024-04-09 |
Family
ID=76469572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110263381.2A Active CN113034363B (en) | 2021-03-11 | 2021-03-11 | GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113034363B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106033540A (en) * | 2016-05-27 | 2016-10-19 | 北京大学第医院 | Automatic analyzing method and system for vaginal microecological morphology |
| CN110059538A (en) * | 2019-02-27 | 2019-07-26 | 成都数之联科技有限公司 | A kind of identifying water boy method based on the intensive neural network of depth |
| CN110068655A (en) * | 2019-04-24 | 2019-07-30 | 中国科学院城市环境研究所 | A kind of air-ground integrated atmospheric monitoring system in day |
| CN110751094A (en) * | 2019-10-21 | 2020-02-04 | 北京师范大学 | Crop yield estimation technology based on GEE comprehensive remote sensing image and deep learning method |
| CN111723525A (en) * | 2020-06-23 | 2020-09-29 | 南通大学 | PM2.5 inversion method based on multi-source data and neural network model |
| CN111861884A (en) * | 2020-07-15 | 2020-10-30 | 南京信息工程大学 | Satellite cloud image super-resolution reconstruction method based on deep learning |
| CN112257551A (en) * | 2020-10-19 | 2021-01-22 | 北京市劳动保护科学研究所 | Method and system for identifying nitrogen oxide pollution source and determining emission |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180039885A1 (en) * | 2016-08-04 | 2018-02-08 | International Business Machines Corporation | Satellite-based location identification of methane-emitting sites |
-
2021
- 2021-03-11 CN CN202110263381.2A patent/CN113034363B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106033540A (en) * | 2016-05-27 | 2016-10-19 | 北京大学第医院 | Automatic analyzing method and system for vaginal microecological morphology |
| CN110059538A (en) * | 2019-02-27 | 2019-07-26 | 成都数之联科技有限公司 | A kind of identifying water boy method based on the intensive neural network of depth |
| CN110068655A (en) * | 2019-04-24 | 2019-07-30 | 中国科学院城市环境研究所 | A kind of air-ground integrated atmospheric monitoring system in day |
| CN110751094A (en) * | 2019-10-21 | 2020-02-04 | 北京师范大学 | Crop yield estimation technology based on GEE comprehensive remote sensing image and deep learning method |
| CN111723525A (en) * | 2020-06-23 | 2020-09-29 | 南通大学 | PM2.5 inversion method based on multi-source data and neural network model |
| CN111861884A (en) * | 2020-07-15 | 2020-10-30 | 南京信息工程大学 | Satellite cloud image super-resolution reconstruction method based on deep learning |
| CN112257551A (en) * | 2020-10-19 | 2021-01-22 | 北京市劳动保护科学研究所 | Method and system for identifying nitrogen oxide pollution source and determining emission |
Non-Patent Citations (1)
| Title |
|---|
| 基于火焰自由基和软测量方法的燃烧污染物预测;孙愉佳;《中国优秀硕士学位论文全文数据库信息科技辑》;I138-1071 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113034363A (en) | 2021-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11333796B2 (en) | Spatial autocorrelation machine learning-based downscaling method and system of satellite precipitation data | |
| CN112836610B (en) | Land use change and carbon reserve quantitative estimation method based on remote sensing data | |
| CN110298211B (en) | River network extraction method based on deep learning and high-resolution remote sensing image | |
| CN112749627A (en) | Method and device for dynamically monitoring tobacco based on multi-source remote sensing image | |
| US11836605B2 (en) | Meteorological big data fusion method based on deep learning | |
| CN112949414B (en) | Intelligent surface water body drawing method for wide-vision-field high-resolution six-satellite image | |
| CN109300128B (en) | Transfer learning image processing method based on convolution neural network hidden structure | |
| CN115561181B (en) | Water quality inversion method based on unmanned aerial vehicle multispectral data | |
| CN113935249B (en) | Upper-layer ocean thermal structure inversion method based on compression and excitation network | |
| CN111539453A (en) | Global ionized layer electron total content prediction method based on deep cycle neural network | |
| Chen et al. | ARU-net: Reduction of atmospheric phase screen in SAR interferometry using attention-based deep residual U-net | |
| Hou et al. | On the value of available MODIS and Landsat8 OLI image pairs for MODIS fractional snow cover mapping based on an artificial neural network | |
| He et al. | Deep learning approaches to spatial downscaling of GRACE Terrestrial Water Storage Products using EALCO Model over Canada | |
| CN113688172B (en) | Landslide susceptibility evaluation model training method, landslide susceptibility evaluation device and medium | |
| Li et al. | Real-time and seamless monitoring of ground-level pm2. 5 using satellite remote sensing | |
| CN113034363B (en) | GEE depth space-time experience Kringing region scale-based nitrogen oxide rapid reduction method | |
| CN116091936A (en) | Agricultural condition parameter inversion method for fusing point-land block-area scale data | |
| CN111325384B (en) | NDVI prediction method combining statistical characteristics and convolutional neural network model | |
| CN117589646B (en) | Method, device, equipment and medium for monitoring concentration of atmospheric fine particulate matters | |
| Jia et al. | Siamese Meets Diffusion Network: SMDNet for Enhanced Change Detection in High-Resolution RS Imagery | |
| CN117036987B (en) | Remote sensing image space-time fusion method and system based on wavelet domain cross pairing | |
| Hadiyanto et al. | Generating himawari-8 time series data for meteorological application | |
| CN117272182B (en) | Daily air temperature prediction method, device, medium and equipment | |
| Wei et al. | A Soil Moisture Downscaling Residual Dense Network Considering Spatiotemporal Relationship | |
| Yu et al. | UFNet: Joint U-Net and fully connected neural network to bias correct precipitation predictions from climate models |
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 |