CN110516816A - Round-the-clock surface temperature generation method and device based on machine learning - Google Patents
Round-the-clock surface temperature generation method and device based on machine learning Download PDFInfo
- Publication number
- CN110516816A CN110516816A CN201910812200.XA CN201910812200A CN110516816A CN 110516816 A CN110516816 A CN 110516816A CN 201910812200 A CN201910812200 A CN 201910812200A CN 110516816 A CN110516816 A CN 110516816A
- Authority
- CN
- China
- Prior art keywords
- surface temperature
- earth
- data set
- data
- machine learning
- 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.)
- Granted
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000005855 radiation Effects 0.000 claims abstract description 69
- 238000012549 training Methods 0.000 claims abstract description 36
- 238000007637 random forest analysis Methods 0.000 claims abstract description 18
- 239000000284 extract Substances 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 43
- 238000002834 transmittance Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 11
- 230000000007 visual effect Effects 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 9
- 230000002708 enhancing effect Effects 0.000 claims description 9
- 238000006116 polymerization reaction Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 12
- 238000003066 decision tree Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013480 data collection Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- YBJHBAHKTGYVGT-ZKWXMUAHSA-N (+)-Biotin Chemical compound N1C(=O)N[C@@H]2[C@H](CCCCC(=O)O)SC[C@@H]21 YBJHBAHKTGYVGT-ZKWXMUAHSA-N 0.000 description 1
- 241000208340 Araliaceae Species 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000012356 Product development Methods 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- FEPMHVLSLDOMQC-UHFFFAOYSA-N virginiamycin-S1 Natural products CC1OC(=O)C(C=2C=CC=CC=2)NC(=O)C2CC(=O)CCN2C(=O)C(CC=2C=CC=CC=2)N(C)C(=O)C2CCCN2C(=O)C(CC)NC(=O)C1NC(=O)C1=NC=CC=C1O FEPMHVLSLDOMQC-UHFFFAOYSA-N 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention discloses a kind of round-the-clock surface temperature generation method and device based on machine learning, the method extract the MODIS data set through remote-sensing inversion using the tool MRT of MODIS；And combined using stationary weather satellite data, with the DEM terrain data of ALOS satellite, it estimates and obtains earth's surface incident solar radiation；The data set of identical space scale is subjected to spatial clustering, with above-mentioned MODIS data set as machine learning training dataset；Surface temperature relational model is constructed by Random Forest model；Estimation has the earth's surface true temperature of cloud covering pixel；There to be the earth's surface true temperature of cloud covering pixel to combine with the data set of cloudless covering pixel, generates round-the-clock surface temperature.The method of the present invention solves the problems, such as current thermal infrared remote sensing, and vulnerable to Influence of cloud, surface temperature product, there are a large amount of blank missing values regions, and realizing has cloud condition surface temperature to estimate, provides important basis for the generation of round-the-clock surface temperature product.
Description
Technical field
The present invention relates to surface temperature monitoring technology more particularly to a kind of round-the-clock surface temperature based on machine learning are raw
At method and device.
Background technique
Surface temperature (Land surface temperature, LST) is as phase between reflection ground vapour in earth surface system
The important parameter of interaction is the key parameter for influencing the processes such as earth's surface ecology, the hydrology, meteorology, is energy between atmosphere and land
The synthesis result of amount exchange and material transport conversion.Therefore, quantitative and accurately acquisition surface temperature spatial-temporal distribution characteristic is over the ground
Gas system capacity balance and ecosystem research have important research significance and value.Also, on region and Global Scale
Resource environment dynamic monitoring needs comprehensive, complete, continuous round-the-clock surface temperature spatial and temporal distributions information, such as applies in agriculture drought
The directions such as calamity forecast, Soil Water management, crop yield forecast, numerical weather forecast and climate change.
In conventional surface temperature observation, ground station observation is most direct mode.But by surface temperature high-altitude
Between heterogeneous influence, the surface temperature of spot measurement has lower spatial representative, is difficult standard by limited ground observation
Really obtain extensive area surface temperature multidate information.In recent years, with the progress and development of satellite remote sensing space exploration technology,
Remote sensing has become the main means for obtaining region and global seismic temperature, wherein monitoring mode the most general includes that thermal infrared is distant
Feel inverting Surface Temperature Retrieval and passive microwave remote sensing inverting surface temperature.
But due to passive microwave Surface Temperature Retrieval precision is lower, between passive microwave data and Thermal Infrared Data it is biggish
The uncertainty of different scale, itself error of passive microwave surface temperature and NO emissions reduction causes data fusion to generate round-the-clockly
There is very big uncertainty in table temperature product；Also, using existing inversion algorithm the problem is that: the earth's surface temperature of estimation
Angle value be not be satellite pass by the moment real cloud under surface temperature, the surface temperature under the conditions of moment clear sky but satellite passes by
Theoretical value, it is difficult to meet the needs of practical application.
Therefore, how to solve the problems, such as the above method, cloud covered areas is round-the-clock effectively under acquisition truth
Surface temperature, so improve surface temperature product spatial continuity, to promoted surface temperature Remote Sensing Products active service and
Application level is of great significance.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.For this purpose, of the invention
First purpose is to propose a kind of round-the-clock surface temperature generation method based on machine learning.
Second object of the present invention is to propose a kind of round-the-clock surface temperature generating means based on machine learning.
Third object of the present invention is to propose a kind of computer equipment.
Fourth object of the present invention is to propose a kind of computer storage medium.
To achieve the above object, in a first aspect, the round-the-clock earth's surface temperature according to an embodiment of the present invention based on machine learning
Spend generation method, which comprises
Using the data processing tools MRT of MODIS, extracted respectively in the product of the corresponding land MODIS through remote-sensing inversion
Have cloud cover pixel, the cloudless corresponding MODIS data set of covering pixel；The MODIS data set includes that normalization is planted
By the surface temperature of index, enhancing vegetation index, leaf area index, surface albedo data and best inversion accuracy；
It is combined using stationary weather satellite data, with the DEM terrain data of ALOS satellite, estimate and obtains earth's surface incidence
Solar radiation；
According to the space scale of the surface temperature, the earth's surface incident solar radiation of identical space scale is polymerize, is obtained
The data set polymerizeing to space scale；The data set of data set and the cloudless covering pixel that the space scale is polymerize,
As machine learning training dataset；
The Random Forest model training machine learning training dataset is selected, constructs and obtains surface temperature relationship mould
Type；The surface temperature relational model is applied to the data set for having cloud covering pixel, is estimated and obtains described having cloud to cover
The earth's surface true temperature of lid pixel；
The earth's surface true temperature for having cloud covering pixel is combined with the data set of the cloudless covering pixel, is generated
Round-the-clock surface temperature.
Second aspect, the round-the-clock surface temperature generating means according to an embodiment of the present invention based on machine learning, comprising:
MODIS data set acquisition module, for the data processing tools MRT using MODIS, on the corresponding land MODIS
Extract respectively in product has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel through remote-sensing inversion；It is described
MODIS data set includes normalized differential vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best anti-
Drill the surface temperature of precision；
Earth's surface incident solar radiation obtains module, for the DEM landform using stationary weather satellite data and ALOS satellite
Data combine, and estimate and obtain earth's surface incident solar radiation；
Machine learning training dataset obtains module, for the space scale according to the surface temperature, by identical sky
Between scale earth's surface incident solar radiation polymerization, obtain space scale polymerization data set；The number that the space scale is polymerize
According to collection and the data set of the cloudless covering pixel, as machine learning training dataset；
Earth's surface true temperature estimation block, for select the Random Forest model training machine learning training dataset,
It constructs and obtains surface temperature relational model；The surface temperature relational model is applied to the data for having cloud covering pixel
Collect, estimate and obtains the earth's surface true temperature for having cloud covering pixel；
Round-the-clock surface temperature generation module, for by it is described have cloud covering pixel earth's surface true temperature with it is described cloudless
The data set of covering pixel combines, and generates round-the-clock surface temperature.
It the third aspect, computer equipment according to an embodiment of the present invention, including memory, processor and is stored in described
On memory and the computer program that can run on the processor, which is characterized in that the processor executes the calculating
The round-the-clock surface temperature generation method based on machine learning as described above is realized when machine program.
Fourth aspect, computer storage medium according to an embodiment of the present invention are stored thereon with computer program, feature
It is, the round-the-clock surface temperature generation method based on machine learning as described above is realized when which is executed by processor.
The round-the-clock surface temperature generation method and device based on machine learning provided according to embodiments of the present invention, passes through
MODIS data set through remote-sensing inversion is extracted using the tool MRT of MODIS；And it is defended using stationary weather satellite data, with ALOS
The DEM terrain data of star combines, and estimates and obtains earth's surface incident solar radiation；The data set of identical space scale is carried out
Spatial clustering, with above-mentioned MODIS data set as machine learning training dataset；Surface temperature is constructed by Random Forest model
Relational model；Estimation has the earth's surface true temperature of cloud covering pixel；There to be the earth's surface true temperature of cloud covering pixel to cover with cloudless
The data set of lid pixel combines, and generates round-the-clock surface temperature.The method of the present invention efficiently solves current thermal infrared remote sensing
There are problems that a large amount of blank missing values region vulnerable to Influence of cloud, surface temperature product, has realized the round-the-clock earth's surface of cloud condition
Temperature estimation provides important basis for the generation of round-the-clock surface temperature product.
Additional aspect and advantage of the invention will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
The structure shown according to these attached drawings obtains other attached drawings.
Fig. 1 is the flow chart of round-the-clock surface temperature generation method of the embodiment of the present invention based on machine learning；
Fig. 2 is original surface temperature product data on the MODIS daytime schematic diagram of MODIS surface temperature product introduction；
Fig. 3 is the surface temperature schematic diagram data reconstructed using embodiment of the present invention method；
Fig. 4 is the structure chart of round-the-clock surface temperature generating means of the embodiment of the present invention based on machine learning；
Fig. 5 is the structural schematic diagram of computer equipment one embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not considered as limiting the invention.
Currently, remote sensing has become the main means for obtaining region and global seismic temperature, wherein the most general monitoring side
Formula includes thermal infrared remote sensing inverting Surface Temperature Retrieval and passive microwave remote sensing inverting surface temperature.But due to passive microwave earth's surface
Temperature retrieval precision is lower, biggish different scale between passive microwave data and Thermal Infrared Data, passive microwave surface temperature
Itself error and the uncertainty of NO emissions reduction cause data fusion generate round-the-clock surface temperature product exist it is very big not really
It is qualitative；Also, using existing inversion algorithm the problem is that: the surface temperature value of estimation be not be that satellite passes by the moment
Surface temperature under real cloud, but satellite passes by under the conditions of moment clear sky surface temperature theoretical value, it is difficult to meet practical application
Demand.
Shown in referring to Fig.1, it is raw that Fig. 1 shows the round-the-clock surface temperature provided in an embodiment of the present invention based on machine learning
At the flow chart of method one embodiment, for ease of description, only the parts related to the embodiment of the present invention are shown.Specifically
, it is somebody's turn to do the round-the-clock surface temperature generation method based on machine learning and is executed by computer.
The present invention specifically includes when it is implemented, being somebody's turn to do the round-the-clock surface temperature generation method based on machine learning:
S101, the data processing tools MRT using MODIS, are extracted respectively in the product of the corresponding land MODIS through remote sensing
Inverting has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel.The MODIS data set includes normalizing
Change the surface temperature of vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best inversion accuracy.
S102, it is combined using stationary weather satellite data, with the DEM terrain data of ALOS satellite, estimates and obtain ground
Table incident solar radiation.
S103, according to the space scale of the surface temperature, the earth's surface incident solar radiation of identical space scale is gathered
It closes, obtains the data set of space scale polymerization.The number of data set and the cloudless covering pixel that the space scale is polymerize
According to collection, as machine learning training dataset.
S104, the Random Forest model training machine learning training dataset is selected, constructs and obtains surface temperature pass
It is model.The surface temperature relational model is applied to the data set for having cloud covering pixel, is estimated and obtains described having
The earth's surface true temperature of cloud covering pixel.
S105, the earth's surface true temperature for having cloud covering pixel and the data set of the cloudless covering pixel are mutually tied
It closes, generates round-the-clock surface temperature.
Round-the-clock surface temperature generation method provided in an embodiment of the present invention based on machine learning is to overcome above-mentioned presence
The problem of, efficiently solving current thermal infrared remote sensing, there are a large amount of blank missing values areas vulnerable to Influence of cloud, surface temperature product
The problem of domain, realizes the round-the-clock surface temperature for effectively obtaining cloud covered areas under truth, and then improves surface temperature product
Spatial continuity, while also by effectively improve surface temperature product Regional Hydrologic, ecology, agricultural and in terms of answer
With level, it is of great significance in the active service and application level for promoting surface temperature Remote Sensing Products.
When it is implemented, the thermal environment situation of earth's surface and vegetation cover condition, surface albedo, Earth Surface Atmosphere force
The factors such as condition are related, and for the relational model of accurate building surface temperature and earth's surface other parameters, selection is suitable and can table
The parameter of sign surface temperature height is key, and wherein the Land Surface Parameters of remote-sensing inversion are important data source.
In step S101, by using the data processing tools MRT of MODIS, in the product of the corresponding land MODIS respectively
Extract has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel through remote-sensing inversion.The MODIS data
Collection includes the ground of normalized differential vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best inversion accuracy
Table temperature.
Present invention selection is based primarily upon the land MODIS product development relevant parameter and mentions by taking the reconstruction of MODIS surface temperature as an example
Take work；Design parameter includes MOD11A1 surface temperature product, MOD13A2 vegetation index product, MCD15A3 leaf area index
Product and MCD43A3 earth surface albedo product, the inversion algorithm of relevant parameter can refer to the specification of each product.
Based on above-mentioned land product, the present invention uses MODIS re-projection tool (MRT), MRT (MODIS
Reprojection Tool) it is a kind of handling implement for MODIS data.It can help user MODIS image again
The map projection of more standard is projected to, and can choose the space subset in image and band subset progress projection transform.
Surface temperature, normalized differential vegetation index (NDVI), enhancing vegetation index are extracted using the data processing tools MRT of MODIS respectively
(EVI), leaf area index (LAI) and surface albedo data.
Further, described to there is cloud covering pixel, the corresponding data set of cloudless covering pixel to select 1 km space ruler
The MODIS data of degree.It is according to the surface temperature of 1 km space scale, the leaf area index of 300 meters of space scales and earth's surface is anti-
It is resampled in the MODIS data set of 1km space scale according to rate data by spatial clustering.That is, according to earth's surface temperature space point
Leaf area index and surface albedo data (500m) are resampled to 1km spatial discrimination by spatial clustering by resolution (1km)
Rate.In addition, controlling file according to the quality of corresponding surface temperature product in the present invention, best inversion accuracy surface temperature is selected
Data are to model data source used the later period.It therefore, include that best inverting is selected from MODIS product in the MODIS data set
The surface temperature of precision.
Further, earth's surface incident solar radiation is the significant energy source that earth's surface heats, to the diurnal periodicity of surface temperature
Change procedure has a very important significance.Therefore, carry out the modeling of surface temperature relational model and need accurate earth's surface incidence too
Positive radiation information.In the present invention, the solar radiation factor indicates that is, the sun is raised up to satellite using accumulative incident solar radiation
It observes moment solar radiation and receives total amount.However, accurately to indicate that earth's surface receives solar radiation situation, the present invention during this
In such a way that stationary weather satellite data mutually cooperate with, using its high time resolution feature, accurately indicate each pixel from the sun
The cloud coverage condition at moonscope moment is risen to, and then effectively estimates effective solar radiation of each pixel.At the same time, it is
Influences of the accurate description different terrain conditions to sun incident radiation, the present invention use high spatial resolution dem data, i.e.,
30 Miho Dockyard EM data of ALOS satellite introduce the orographic factors such as the gradient, slope aspect and estimate offer ground figurate number to earth's surface incident solar radiation
According to support.
Specifically, in a step 102, it is combined using stationary weather satellite data with the DEM terrain data of ALOS satellite,
It estimates and obtains earth's surface incident solar radiation.Wherein, the stationary weather satellite data select the number of 3~5 km space scales
The terrain data of 30m scale is selected according to the DEM terrain data of, the ALOS satellite.
Further, the earth's surface incident solar radiation fRIncluding beam radia Rb, sky radiation RdWith it is neighbouring
Landform radiates Rr。
The beam radia RbIt calculates as follows:
Wherein, EoIt is solar constant.
Dr is same day solar distance correction factor, wherein the calculation formula of dr are as follows:
DOY is day of year in the formula.
τbDirectly to radiate atmospheric transmittance.
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth
Angle, S are terrain slope, A is landform slope aspect.
Further, sky radiation R under complicated landformd, it is to sky radiation R under level terraind,flat, benefit
Acquisition is modified with the visual factor S VF of sky.The sky radiation RdIt calculates as follows:
Rd=Rd,flat×SVF。
Wherein, Rd,flatFor the sky radiation under level terrain, wherein Rd,flatCalculation formula are as follows: Rd,flat=EO
×dr×cos(Zs)×τd, E in the formulaoFor solar constant, dr is same day solar distance correction factor, and wherein the calculating of dr is public
Formula are as follows:
DOY is day of year in the formula, wherein τdTo scatter radiation transmission
Rate, the scattering of sky is a homogeneous scattering under the conditions of ceiling unlimited, is directly radiating and is scattering between radiation in the presence of linear
Relationship, τdCalculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance.
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions above target point
Facet is accumulated and hemisphere area ratio, wherein the calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor in 16 directions in each direction each slope member with rise
The maximum height angle of point slope member, the slope member are the slope surface grid cell with certain slope aspect, the gradient.
Further, the approximate calculation side that Dozier simplifies can be used in the spurious radiation generated for the reflection of slope surface around
Method only considers the average reflection effect of terrain slope, sky viewing factor and surrounding terrain.Therefore, the neighbouring landform spoke
Penetrate RrThe approximate calculation method simplified using Dozier is specifically calculated in the following manner:
Rr=ρ × Ct×(Rb+Rd)。
Wherein, ρ is the average albedo of neighbouring landform.
CtFor topographic structure parameter, CtInclude the anisotropic properties of reflected radiation, due to around slope surface have closely have it is remote,
And the gradient, slope aspect are different, therefore there are anisotropic properties.It and also include slope member, the slope member is that have certain slope aspect, slope
The slope surface grid cell of degree；Because influential on the reflected radiation of a certain slope member is the visible slope member around it, the slope
Geometric effect between first slope member visible with surrounding, it is assumed that underlying surface is lambert's body, then CtCalculation formula are as follows:
Ct=(1+cosS)/2-SVF, it is the visual factor of sky that wherein S, which is the gradient, SVF, and SVF calculates institute in the manner described above
, being consistent property of this paper or more parameter calculation.
Specifically, by ground observation size of data it can be found that in earth's surface incident solar radiation mainly by direct solar radiation,
Scattering radiation is influenced, and neighbouring landform reflected radiation proportion very little.Therefore, in the present invention, neighbouring landform reflects spoke
Penetrating item can put aside.For the difference for finely considering earth's surface incident radiation under different terrain conditions, the present invention uses first
ALOS satellite 30m high spatial resolution dem data is for estimating earth's surface incident solar radiation.
Further, it is based on relation above formula, after rejecting shared item, the earth's surface incident solar radiation fRCan by with
Lower simplified formula calculates:
Wherein, τbDirectly to radiate atmospheric transmittance.
τdTo scatter Radiation Transmittance, wherein τdCalculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance.
ZsFor solar zenith angle.
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth
Angle, S are terrain slope, A is landform slope aspect.
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions above target point
Facet is accumulated and hemisphere area ratio, wherein the calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor in 16 directions in each direction each slope member with
The maximum height angle of starting point slope member, the slope member are the slope surface grid cell with certain slope aspect, the gradient.
VtRepresent whether the moment has cloud, value is 1 under cloudless coverage condition, and having value under cloud coverage condition is 0.The ginseng
Number is based primarily upon stationary weather satellite observation data acquisition.When static meteorological data obtains effective earth's surface temperature data, can recognize
It is otherwise cloud covering for the pixel clear sky.
Further, described directly to radiate atmospheric transmittance τbSpecifically calculate in the following manner:
τb=0.56 (e-0.65M+e-0.095M)。
Wherein, M is air quality ratio, i.e., the air quality that solar radiation direction passes through passes through atmosphere with the direct projection of sun zenith
The ratio of quality, wherein τdCalculation formula are as follows:
P/p in the formula0For height
Function, wherein p/p0Cut-and-try formula are as follows:
p/p0=exp (- z/8434.5), z is height above sea level in the formula.
Wherein,To reflect the solar elevation after correcting by the sun,Calculation formula are as follows:
H in the formulaoFor not corrected solar elevation,For correction coefficient,
Calculation formula are as follows:
Further, after obtaining earth's surface sun incident radiation, in step s 103, the space according to the surface temperature
The earth's surface incident solar radiation of identical space scale polymerize by scale, obtains the data set of space scale polymerization.By the sky
Between scale polymerize data set and it is described it is cloudless covering pixel data set, as machine learning training dataset.
When it is implemented, can be seen that from above-mentioned estimation process, present invention employs the data set of different spaces scale, packets
Include the MODIS data of 1km scale, the stationary weather satellite data of 3-5km scale, 30m scale terrain data.In order to standard
The really change procedure of characterization 1km scale surface temperature, wherein a critically important step is the sky of different spaces scale data
Between matching problem.The present invention mainly concentrates the latitude and longitude information of each pixel by each spatial data, establish coarse resolution pixel with
Spatial correspondence between fine resolution pixel by the way of spatial clustering, will be estimated on 30m scale on this basis
The polymerizations such as solar radiation, the terrain factor of acquisition extremely with MODIS surface temperature same space scale, form the consistent data in space
Collection.At the same time, in conjunction with MODIS surface temperature data quality information, the cloudless pixel of clear sky is extracted, and extracts each Land Surface Parameters shape
At machine learning training dataset；Remaining cloud covering pixel is application data set, for estimating that pixel cloud covers lower earth's surface temperature
Degree evidence.
Further, in step S104, S105, the Random Forest model training machine learning training data is selected
Collect, construct and obtains surface temperature relational model.There is cloud to cover pixel applied to described the surface temperature relational model
Data set is estimated and obtains the earth's surface true temperature for having cloud covering pixel.The earth's surface for having cloud covering pixel is true
Temperature is combined with the data set of the cloudless covering pixel, generates round-the-clock surface temperature.
When it is implemented, the high-precision modeling of surface temperature relational model is to realize to have surface temperature under cloud coverage condition quasi-
The premise really estimated.According to research investigation and early-stage study as a result, the present invention selects random forest to return (Random
Forest Regression) method realize surface temperature relational model establish.
Random forest (RF) is a kind of machine learning model that Breiman is proposed, the essence of this method is decision Tree algorithms,
But the result that multiple decision trees combine is become by single decision tree.Therefore, random forest is made of more decision trees, and forest
In each decision tree between be not associated with, the final output of model is codetermined by each decision tree in forest.It should
Method is by bootstrap resampling technique, the self-service merging of multiple samples that will be extracted from original training sample first, generates
Training sample intersection；Then multiple decision trees and composition RF are generated according to self-service sample set, classification or regression model result are pressed
Decision tree is voted depending on score.In the present invention, the relationship between surface temperature and various parameters and non-linear, and RF model is to more
First synteny is insensitive, can effectively prevent overfitting.
Therefore, the training set based on the cloudless pixel of clear sky constructs surface temperature relational model first, prediction result for
Missing data and non-equilibrium data are more steady, for a variety of observational datas used in the present invention, can handle a large amount of input
Variable simultaneously obtains preferable estimation precision.Therefore, this method does not significantly improve operand while improving precision of prediction,
Compared with the linear regression fit of traditional least square, advantage is become apparent.
Finally, the surface temperature relational model that above-mentioned random forest method training generates is applied to cloud and covers pel data
Collection, and then estimate the earth's surface true temperature of cloud covering pixel, and combine with clear sky pixel, form round-the-clock surface temperature data
Collection, concrete application contrast effect is as shown in Figure 2 and Figure 3, and Fig. 2 shows original surface temperature product data on MODIS daytime,
In since cloud covering causes to study the most of region in area the north and south there is null value.Fig. 3 is reconstructed after using the method for the present invention
The surface temperature information of surface temperature data, most of cloud covering pixel has obtained effective recovery, has reasonable spatial distribution
Pattern.
In conclusion the round-the-clock surface temperature generation method provided in an embodiment of the present invention based on machine learning is to overcome
Above-mentioned problem efficiently solves current thermal infrared remote sensing vulnerable to Influence of cloud, surface temperature product and there are a large amount of skies
The problem of white missing value region, realizes the round-the-clock surface temperature for effectively obtaining cloud covered areas under truth, and then improves earth's surface
The spatial continuity of temperature product, while surface temperature product will be also effectively improved in Regional Hydrologic, ecology, agricultural and meteorology etc.
The application level of aspect is of great significance in the active service and application level for promoting surface temperature Remote Sensing Products.
Referring to shown in Fig. 4, it is raw that Fig. 4 shows the round-the-clock surface temperature provided in an embodiment of the present invention based on machine learning
At the structural schematic diagram of device one embodiment, for ease of description, only the parts related to the embodiment of the present invention are shown.Tool
Body, being somebody's turn to do the round-the-clock surface temperature generating means 10 based on machine learning includes:
MODIS data set acquisition module 11, for the data processing tools MRT using MODIS, in the corresponding land MODIS
Extract respectively in local specialties has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel through remote-sensing inversion.Institute
Stating MODIS data set includes normalized differential vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best
The surface temperature of inversion accuracy.
Earth's surface incident solar radiation obtains module 12, for the DEM using stationary weather satellite data and ALOS satellite
Graphic data combines, and estimates and obtains earth's surface incident solar radiation.
Machine learning training dataset obtains module 13 will be identical for the space scale according to the surface temperature
The earth's surface incident solar radiation of space scale polymerize, and obtains the data set of space scale polymerization.The space scale is polymerize
The data set of data set and the cloudless covering pixel, as machine learning training dataset.
Earth's surface true temperature estimation block 14, for selecting the Random Forest model training machine learning training data
Collect, construct and obtains surface temperature relational model.There is cloud to cover pixel applied to described the surface temperature relational model
Data set is estimated and obtains the earth's surface true temperature for having cloud covering pixel.
Round-the-clock surface temperature generation module 15, for by it is described have cloud covering pixel earth's surface true temperature and the nothing
The data set of cloud covering pixel combines, and generates round-the-clock surface temperature.
Further, described to there is cloud covering pixel, the corresponding data set of cloudless covering pixel to select in described device
The MODIS data of 1 km space scale, the stationary weather satellite data select the data, described of 3~5 km space scales
The DEM terrain data of ALOS satellite selects the terrain data of 30m scale.
According to the surface temperature of 1 km space scale, by the leaf area index and surface albedo number of 300 meters of space scales
According to being resampled to by spatial clustering in the MODIS data set of 1km space scale.
Further, in described device, the earth's surface incident solar radiation fRIncluding beam radia Rb, sky scattering
Radiate RdR is radiated with neighbouring landformr。
The beam radia RbIt calculates as follows:
Wherein, EoIt is solar constant.
Dr is same day solar distance correction factor, wherein the calculation formula of dr are as follows:
DOY is day of year in the formula.
τbDirectly to radiate atmospheric transmittance.
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth
Angle, S are terrain slope, A is landform slope aspect.
Further, in described device, the sky radiation RdIt calculates as follows:
Rd=Rd,flat×SVF。
Wherein, Rd,flatFor the sky radiation under level terrain, wherein Rd,flatCalculation formula are as follows: Rd,flat=EO
×dr×cos(Zs)×τd, E in the formulaoFor solar constant, dr is same day solar distance correction factor, and wherein the calculating of dr is public
Formula are as follows:
DOY is day of year, τ in the formuladTo scatter Radiation Transmittance,
Middle τdCalculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance.
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions above target point
Facet is accumulated and hemisphere area ratio, wherein the calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor in 16 directions in each direction each slope member with rise
The maximum height angle of point slope member, the slope member are the slope surface grid cell with certain slope aspect, the gradient.
Further, in described device, the neighbouring landform radiates RrThe approximate calculation method simplified using Dozier, tool
Body calculates in the following manner:
Rr=ρ × Ct×(Rb+Rd)。
Wherein, ρ is the average albedo of neighbouring landform.
CtFor topographic structure parameter, wherein CtCalculation formula are as follows:
Ct=(1+cosS)/2-SVF, it is the visual factor of sky that wherein S, which is the gradient, SVF,.
Further, in described device, the earth's surface incident solar radiation fRCalculating can be simplified as follows:
Wherein, τbDirectly to radiate atmospheric transmittance.
τdTo scatter Radiation Transmittance, wherein τdCalculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance.
ZsFor solar zenith angle.
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth
Angle, S are terrain slope, A is landform slope aspect.
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions above target point
Facet is accumulated and hemisphere area ratio, wherein the calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor in 16 directions in each direction each slope member with
The maximum height angle of starting point slope member, the slope member are the slope surface grid cell with certain slope aspect, the gradient.
VtRepresent whether the moment has cloud, value is 1 under cloudless coverage condition, and having value under cloud coverage condition is 0.
Further, described directly to radiate atmospheric transmittance τ in described devicebSpecifically calculate in the following manner:
τb=0.56 (e-0.65M+e-0.095M)。
Wherein, M is air quality ratio, i.e., the air quality that solar radiation direction passes through passes through atmosphere with the direct projection of sun zenith
The ratio of quality, wherein τdCalculation formula are as follows:
P/p in the formula0For height
Function, wherein p/p0Cut-and-try formula are as follows:
p/p0=exp (- z/8434.5), z is height above sea level in the formula.
Wherein,To reflect the solar elevation after correcting by the sun,Calculation formula are as follows:
H in the formulaoFor not corrected solar elevation,For correction coefficient,
Calculation formula are as follows:
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
For device or system class embodiment, since it is basically similar to the method embodiment, so be described relatively simple, it is related
Place illustrates referring to the part of embodiment of the method.
Referring to Figure 5, Fig. 5 shows the structural schematic diagram of computer equipment embodiment provided in an embodiment of the present invention,
For ease of description, only the parts related to the embodiment of the present invention are shown.Specifically, the computer equipment 500 includes storage
Device 502, processor 501 and it is stored in the computer program that can be run in the memory 502 and on the processor 501
5021, the processor 501 is realized when executing the computer program such as the step of above-described embodiment the method, such as Fig. 1
Shown in S101 to S105 the step of.Alternatively, the processor 501 realizes above-described embodiment institute when executing the computer program
State the function of each module/unit in device, such as the function of module 11 to 15 shown in Fig. 2.
Illustratively, the computer program can be divided into one or more module/units, one or more
A module/unit is stored in the memory 502, and is executed by the processor 501, to complete the present invention.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program in the computer equipment 500 is described.For example, the computer program can be divided
It is cut into MODIS data set acquisition module 11, earth's surface incident solar radiation obtains module 12, machine learning training dataset obtains mould
Block 13, earth's surface true temperature estimation block 14, round-the-clock surface temperature generation module 15.
MODIS data set acquisition module 11, for the data processing tools MRT using MODIS, in the corresponding land MODIS
Extract respectively in local specialties has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel through remote-sensing inversion.Institute
Stating MODIS data set includes normalized differential vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best
The surface temperature of inversion accuracy.
Earth's surface incident solar radiation obtains module 12, for the DEM using stationary weather satellite data and ALOS satellite
Graphic data combines, and estimates and obtains earth's surface incident solar radiation.
Machine learning training dataset obtains module 13 will be identical for the space scale according to the surface temperature
The earth's surface incident solar radiation of space scale polymerize, and obtains the data set of space scale polymerization.The space scale is polymerize
The data set of data set and the cloudless covering pixel, as machine learning training dataset.
Earth's surface true temperature estimation block 14, for selecting the Random Forest model training machine learning training data
Collect, construct and obtains surface temperature relational model.There is cloud to cover pixel applied to described the surface temperature relational model
Data set is estimated and obtains the earth's surface true temperature for having cloud covering pixel.
Round-the-clock surface temperature generation module 15, for by it is described have cloud covering pixel earth's surface true temperature and the nothing
The data set of cloud covering pixel combines, and generates round-the-clock surface temperature.
The computer equipment 500 may include, but be not limited only to processor 501, memory 502.Those skilled in the art
It is appreciated that figure is only the example of computer equipment 500, the restriction to computer equipment 500 is not constituted, may include ratio
More or fewer components are illustrated, perhaps combine certain components or different components, such as the computer equipment 500 is also
It may include input-output equipment, network access equipment, bus etc..
Alleged processor 501 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors 501, digital signal processor 501 (Digital Signal Processor, DSP), dedicated integrated electricity
Road (Application Specific Integrated Circuit, ASIC), field programmable gate array
(FieldProgrammable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor are patrolled
Collect device, discrete predetermined hardware component etc..General processor 501 can be microprocessor 501 or the processor 501 can also be with
It is any conventional processor 501 etc..
The memory 502 can be the internal storage unit of the computer equipment 500, such as computer equipment 500
Hard disk or memory.The memory 502 is also possible to the External memory equipment of the computer equipment 500, such as the meter
Calculate the plug-in type hard disk being equipped on machine equipment 500, intelligent memory card (Smart Media Card, SMC), secure digital
(Secure Digital, SD) card, flash card (Flash Card) etc..Further, the memory 502 can also both include
The internal storage unit of the computer equipment 500 also includes External memory equipment.The memory 502 by store it is described based on
Other programs and data needed for calculation machine program 5021 and the computer equipment 500.The memory 502 can be also used for
Temporarily store the data that has exported or will export.
The embodiment of the invention also provides a kind of computer readable storage medium, computer-readable recording medium storage has meter
Calculation machine program is realized the step in the method as described in above-described embodiment, such as is schemed when computer program is executed by processor 501
Step S101 to S105 shown in 1.Alternatively, the computer program realizes institute in above-described embodiment when being executed by processor 501
State the function of each module/unit in device, such as the function of module 11 to 15 shown in Fig. 2.
The computer program can be stored in a computer readable storage medium, and the computer program is by processor
501 when executing, it can be achieved that the step of above-mentioned each embodiment of the method.Wherein, the computer program includes computer program generation
Code, the computer program code can be source code form, object identification code form, executable file or certain intermediate forms
Deng.The computer-readable medium may include: any entity or device, record that can carry the computer program code
Medium, USB flash disk, mobile hard disk, magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), with
Machine access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..
It should be noted that the computer-readable medium include content can according in jurisdiction legislation and specially
The requirement of benefit practice carries out increase and decrease appropriate, such as in certain jurisdictions, computer-readable according to legislation and patent practice
Medium do not include be electric carrier signal and telecommunication signal.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
The steps in the embodiment of the present invention can be sequentially adjusted, merged and deleted according to actual needs.
Module or unit in system of the embodiment of the present invention can be combined, divided and deleted according to actual needs.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronics predetermined hardware or computer software and electronics predetermined hardware.These
Function is executed actually with predetermined hardware or software mode, specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this realization
It should not be considered as beyond the scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed device/computer equipment 500 and method, it can
To realize by another way.For example, 500 embodiment of device/computer equipment described above is only schematical,
For example, the division of the module or unit, only a kind of logical function partition, can there is other division side in actual implementation
Formula, such as multiple units or components can be combined or can be integrated into another system, or some features can be ignored, or not
It executes.Another point, shown or discussed mutual coupling or direct-coupling or communication connection can be to be connect by some
Mouthful, the INDIRECT COUPLING or communication connection of device or unit can be electrical property, mechanical or other forms.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of round-the-clock surface temperature generation method based on machine learning, which is characterized in that the described method includes:
Using the data processing tools MRT of MODIS, having through remote-sensing inversion is extracted respectively in the product of the corresponding land MODIS
Cloud covers pixel, the cloudless corresponding MODIS data set of covering pixel；The MODIS data set includes that normalization vegetation refers to
Number, the surface temperature for enhancing vegetation index, leaf area index, surface albedo data and best inversion accuracy；
It is combined using stationary weather satellite data, with the DEM terrain data of ALOS satellite, estimate and obtains the earth's surface incidence sun
Radiation；
According to the space scale of the surface temperature, the earth's surface incident solar radiation of identical space scale is polymerize, sky is obtained
Between scale polymerize data set；The data set of data set and the cloudless covering pixel that the space scale is polymerize, as
Machine learning training dataset；
The Random Forest model training machine learning training dataset is selected, constructs and obtains surface temperature relational model；It will
The surface temperature relational model is applied to the data set for having cloud covering pixel, estimates and obtains described having cloud to cover pixel
Earth's surface true temperature；
The earth's surface true temperature for having cloud covering pixel is combined with the data set of the cloudless covering pixel, generates whole day
Wait surface temperature.
2. the round-the-clock surface temperature generation method according to claim 1 based on machine learning, which is characterized in that described
There are cloud covering pixel, the corresponding data set of cloudless covering pixel to select the MODIS data, described quiet of 1 km space scale
Only weather satellite data selects the DEM terrain data of the data of 3~5 km space scales, the ALOS satellite to select 30m scale
Terrain data；
According to the surface temperature of 1 km space scale, the leaf area index of 500 meters of space scales and surface albedo data are led to
Spatial clustering is crossed to be resampled in the MODIS data set of 1km space scale.
3. the round-the-clock surface temperature generation method according to claim 1 based on machine learning, which is characterized in that described
Earth's surface incident solar radiation fRIncluding beam radia Rb, sky radiation RdR is radiated with neighbouring landformr；
The beam radia RbIt calculates as follows:
Wherein, EoIt is solar constant；
Dr is same day solar distance correction factor, wherein the calculation formula of dr are as follows:
DOY is day of year in the formula；
τbDirectly to radiate atmospheric transmittance；
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth, S
It is landform slope aspect for terrain slope, A.
4. the round-the-clock surface temperature generation method according to claim 3 based on machine learning, which is characterized in that described
Sky radiation RdIt calculates as follows:
Rd=Rd,flat×SVF；
Wherein, Rd,flatFor the sky radiation under level terrain, wherein Rd,flatCalculation formula are as follows: Rd,flat=EO×dr×
cos(Zs)×τd, E in the formulaoFor solar constant, dr is same day solar distance correction factor, wherein the calculation formula of dr are as follows:
DOY is day of year, τ in the formuladTo scatter Radiation Transmittance, wherein τd
Calculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance；
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions facet above target point
It is long-pending with hemisphere area ratio, the wherein calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor each slope member and starting point slope in each direction in 16 directions
The maximum height angle of member, the slope member are the slope surface grid cell with certain slope aspect, the gradient.
5. the round-the-clock surface temperature generation method according to claim 4 based on machine learning, which is characterized in that described
Neighbouring landform radiates RrThe approximate calculation method simplified using Dozier is specifically calculated in the following manner:
Rr=ρ × Ct×(Rb+Rd)；
Wherein, ρ is the average albedo of neighbouring landform；
CtFor topographic structure parameter, wherein CtCalculation formula are as follows:
Ct=(1+cosS)/2-SVF, it is the visual factor of sky that wherein S, which is the gradient, SVF,.
6. the round-the-clock surface temperature generation method according to claim 3 based on machine learning, which is characterized in that described
Earth's surface incident solar radiation fRCalculating can be simplified as follows:
Wherein, τbDirectly to radiate atmospheric transmittance；
τdTo scatter Radiation Transmittance, wherein τdCalculation formula are as follows:
τd=0.271-0.294 × τb, τ in the formulabDirectly to radiate atmospheric transmittance；
ZsFor solar zenith angle；
θ is the angle of sun straight rays and earth's surface slope surface normal, wherein the calculation formula of cos θ are as follows:
Cos θ=cosZscosS+cosZssinScos(As- A), Z in the formulasFor solar zenith angle, AsFor solar azimuth, S
It is landform slope aspect for terrain slope, A；
SVF is the visual factor of sky, is defined as 2 space π of hemisphere being divided into n equal portions, hemisphere visible portions facet above target point
It is long-pending with hemisphere area ratio, the wherein calculation formula of SVF are as follows:
N takes 16, h in the formulaiFor each slope member and starting point in each direction in 16 directions
The maximum height angle of slope member, the slope member are the slope surface grid cell with certain slope aspect, the gradient；
VtRepresent whether the moment has cloud, value is 1 under cloudless coverage condition, and having value under cloud coverage condition is 0.
7. the round-the-clock surface temperature generation method according to claim 5 or 6 based on machine learning, which is characterized in that
It is described directly to radiate atmospheric transmittance τbSpecifically calculate in the following manner:
τb=0.56 (e-0.65M+e-0.095M)；
Wherein, M is air quality ratio, i.e., the air quality that solar radiation direction passes through passes through air quality with the direct projection of sun zenith
Ratio, wherein τdCalculation formula are as follows:
P/p in the formula0For the letter of height
It counts, wherein p/p0Cut-and-try formula are as follows:
p/p0=exp (- z/8434.5), z is height above sea level in the formula；
Wherein,To reflect the solar elevation after correcting by the sun,Calculation formula are as follows:
H in the formulaoFor not corrected solar elevation,For correction coefficient, meter
Calculate formula are as follows:
8. a kind of round-the-clock surface temperature generating means based on machine learning, which is characterized in that described device includes:
MODIS data set acquisition module, for the data processing tools MRT using MODIS, in the corresponding land MODIS product
Middle extract respectively has cloud to cover pixel, the cloudless corresponding MODIS data set of covering pixel through remote-sensing inversion；It is described
MODIS data set includes normalized differential vegetation index, enhancing vegetation index, leaf area index, surface albedo data and best anti-
Drill the surface temperature of precision；
Earth's surface incident solar radiation obtains module, for the DEM terrain data using stationary weather satellite data and ALOS satellite
It combines, estimate and obtains earth's surface incident solar radiation；
Machine learning training dataset obtains module, for the space scale according to the surface temperature, by identical space ruler
The earth's surface incident solar radiation of degree polymerize, and obtains the data set of space scale polymerization；Data set that the space scale is polymerize,
And the data set of the cloudless covering pixel, as machine learning training dataset；
Earth's surface true temperature estimation block, for selecting the Random Forest model training machine learning training dataset, building
And obtain surface temperature relational model；By the surface temperature relational model be applied to the data set for having cloud covering pixel,
It estimates and obtains the earth's surface true temperature for having cloud covering pixel；
Round-the-clock surface temperature generation module, for by it is described have cloud covering pixel earth's surface true temperature and the cloudless covering
The data set of pixel combines, and generates round-the-clock surface temperature.
9. a kind of computer equipment, including memory, processor and it is stored on the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
Round-the-clock surface temperature generation method described in 7 any one based on machine learning.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
The Shi Shixian round-the-clock surface temperature generation method as claimed in any one of claims 1 to 7 based on machine learning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910812200.XA CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910812200.XA CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110516816A true CN110516816A (en) | 2019-11-29 |
| CN110516816B CN110516816B (en) | 2021-05-28 |
Family
ID=68629222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910812200.XA Active CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516816B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652404A (en) * | 2020-02-06 | 2020-09-11 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| CN112487346A (en) * | 2020-10-26 | 2021-03-12 | 中国农业科学院农业资源与农业区划研究所 | Mountain land surface temperature remote sensing retrieval method |
| CN112560570A (en) * | 2020-09-29 | 2021-03-26 | 中国科学院大学 | High-resolution earth surface temperature estimation method based on cooperative downscaling and data fusion |
| CN112818605A (en) * | 2021-02-07 | 2021-05-18 | 武汉大学 | Method and system for rapidly estimating earth surface albedo |
| CN113158570A (en) * | 2021-04-26 | 2021-07-23 | 电子科技大学 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
| CN113255148A (en) * | 2021-06-04 | 2021-08-13 | 中国科学院地理科学与资源研究所 | Method for estimating all-weather air temperature and space-time distribution thereof based on MODIS product data |
| CN113408742A (en) * | 2021-06-24 | 2021-09-17 | 桂林理工大学 | High-precision sea surface temperature inversion method based on machine learning |
| CN113743000A (en) * | 2021-08-13 | 2021-12-03 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN114169215A (en) * | 2021-08-06 | 2022-03-11 | 中国矿业大学(北京) | Surface temperature inversion method coupling remote sensing and regional meteorological model |
| CN114781148A (en) * | 2022-04-14 | 2022-07-22 | 河北地质大学 | Surface temperature inversion method and system for thermal infrared remote sensing cloud coverage pixel |
| CN117390969A (en) * | 2023-12-05 | 2024-01-12 | 易智瑞信息技术有限公司 | Method for generating surface temperature, electronic equipment and storage medium |
| CN117554300A (en) * | 2024-01-10 | 2024-02-13 | 中国科学院、水利部成都山地灾害与环境研究所 | Remote sensing space downscaling method for mountain land surface albedo site observation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293522A (en) * | 2013-05-08 | 2013-09-11 | 中国科学院光电研究院 | Intermediate infrared two-channel remote sensing data surface temperature inversion method and device |
| CN107065036A (en) * | 2017-04-19 | 2017-08-18 | 中国农业科学院农业资源与农业区划研究所 | A kind of method that joint remote sensing and meteorological data obtain round-the-clock evapotranspiration |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN108388956A (en) * | 2018-01-18 | 2018-08-10 | 华北电力大学 | Consider the photovoltaic power prediction technique of attenuation |
| CN109635309A (en) * | 2018-10-17 | 2019-04-16 | 广州地理研究所 | A kind of surface temperature space NO emissions reduction method |
| CN109885959A (en) * | 2019-03-05 | 2019-06-14 | 中国科学院地理科学与资源研究所 | A kind of surface temperature robust NO emissions reduction method |
| CN110175375A (en) * | 2019-05-13 | 2019-08-27 | 中国科学院遥感与数字地球研究所 | A kind of earth's surface Calculation method for solar radiation based on deep learning |
-
2019
- 2019-08-30 CN CN201910812200.XA patent/CN110516816B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293522A (en) * | 2013-05-08 | 2013-09-11 | 中国科学院光电研究院 | Intermediate infrared two-channel remote sensing data surface temperature inversion method and device |
| CN107065036A (en) * | 2017-04-19 | 2017-08-18 | 中国农业科学院农业资源与农业区划研究所 | A kind of method that joint remote sensing and meteorological data obtain round-the-clock evapotranspiration |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN108388956A (en) * | 2018-01-18 | 2018-08-10 | 华北电力大学 | Consider the photovoltaic power prediction technique of attenuation |
| CN109635309A (en) * | 2018-10-17 | 2019-04-16 | 广州地理研究所 | A kind of surface temperature space NO emissions reduction method |
| CN109885959A (en) * | 2019-03-05 | 2019-06-14 | 中国科学院地理科学与资源研究所 | A kind of surface temperature robust NO emissions reduction method |
| CN110175375A (en) * | 2019-05-13 | 2019-08-27 | 中国科学院遥感与数字地球研究所 | A kind of earth's surface Calculation method for solar radiation based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| PAN HUANG，WEI ZHAO: "The Preliminary Investigation on the Uncertainties Associated with surface solar radiation estimation in mountainous areas", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| 白琳等: "基于随机森林算法的近地表气温遥感反演研究", 《地球信息科学》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652404B (en) * | 2020-02-06 | 2023-04-28 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111652404A (en) * | 2020-02-06 | 2020-09-11 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| CN112560570A (en) * | 2020-09-29 | 2021-03-26 | 中国科学院大学 | High-resolution earth surface temperature estimation method based on cooperative downscaling and data fusion |
| CN112560570B (en) * | 2020-09-29 | 2023-07-25 | 中国科学院大学 | High-resolution earth surface temperature estimation method for collaborative downscaling and data fusion |
| CN112487346A (en) * | 2020-10-26 | 2021-03-12 | 中国农业科学院农业资源与农业区划研究所 | Mountain land surface temperature remote sensing retrieval method |
| CN112487346B (en) * | 2020-10-26 | 2021-07-23 | 中国农业科学院农业资源与农业区划研究所 | Mountain land surface temperature remote sensing retrieval method |
| CN112818605A (en) * | 2021-02-07 | 2021-05-18 | 武汉大学 | Method and system for rapidly estimating earth surface albedo |
| CN113158570A (en) * | 2021-04-26 | 2021-07-23 | 电子科技大学 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
| CN113255148A (en) * | 2021-06-04 | 2021-08-13 | 中国科学院地理科学与资源研究所 | Method for estimating all-weather air temperature and space-time distribution thereof based on MODIS product data |
| CN113408742A (en) * | 2021-06-24 | 2021-09-17 | 桂林理工大学 | High-precision sea surface temperature inversion method based on machine learning |
| CN114169215A (en) * | 2021-08-06 | 2022-03-11 | 中国矿业大学(北京) | Surface temperature inversion method coupling remote sensing and regional meteorological model |
| CN113743000B (en) * | 2021-08-13 | 2023-04-18 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN113743000A (en) * | 2021-08-13 | 2021-12-03 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN114781148A (en) * | 2022-04-14 | 2022-07-22 | 河北地质大学 | Surface temperature inversion method and system for thermal infrared remote sensing cloud coverage pixel |
| CN117390969A (en) * | 2023-12-05 | 2024-01-12 | 易智瑞信息技术有限公司 | Method for generating surface temperature, electronic equipment and storage medium |
| CN117390969B (en) * | 2023-12-05 | 2024-03-12 | 易智瑞信息技术有限公司 | Method for generating surface temperature, electronic equipment and storage medium |
| CN117554300A (en) * | 2024-01-10 | 2024-02-13 | 中国科学院、水利部成都山地灾害与环境研究所 | Remote sensing space downscaling method for mountain land surface albedo site observation |
| CN117554300B (en) * | 2024-01-10 | 2024-03-19 | 中国科学院、水利部成都山地灾害与环境研究所 | Remote sensing space downscaling method for mountain land surface albedo site observation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110516816B (en) | 2021-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516816A (en) | Round-the-clock surface temperature generation method and device based on machine learning | |
| CN106226260A (en) | A kind of combination microwave and the Soil Moisture Inversion method of infrared remote sensing image | |
| Zhang et al. | Estimating leaf area index from MODIS and surface meteorological data using a dynamic Bayesian network | |
| CN103003818A (en) | System and method for geomatic modeling of a diverse resource base across broad landscapes | |
| CN103090868B (en) | A kind of preparation method utilizing the typical feature reference diagram of Classification of Multispectral Images | |
| CN110659450B (en) | Ground surface temperature angle normalization method based on component temperatures | |
| CN104063718A (en) | Method for selecting remote sensing data and classification algorithms in crop identification and area estimation | |
| CN102175209A (en) | Effective sampling method for crop cultivated area measurement under support of historical remote sensing product data | |
| Coops et al. | An environmental domain classification of Canada using earth observation data for biodiversity assessment | |
| Hufkens et al. | Impacts and uncertainties of upscaling of remote-sensing data validation for a semi-arid woodland | |
| US20210223433A1 (en) | Determination of location-specific weather information for agronomic decision support | |
| CN102354328A (en) | Leaf area index (LAI) product inversion method and system for global earth surface | |
| Jiao et al. | Evaluation of four sky view factor algorithms using digital surface and elevation model data | |
| Handayani et al. | Estimation of built-up and green volume using geospatial techniques: A case study of Surabaya, Indonesia | |
| Hu et al. | Influence of emissivity angular variation on land surface temperature retrieved using the generalized split-window algorithm | |
| CN108647401A (en) | A kind of basin nitrogen and phosphorus pollution appraisal procedure based on space remote sensing technology | |
| CN110458618A (en) | Agricultural land value calculation method, device, server and storage medium | |
| Liang et al. | A synergic method of Sentinel-1 and Sentinel-2 images for retrieving soil moisture content in agricultural regions | |
| CN101876700A (en) | Radiation intensity-based method for simulating radiation transfer of complex terrain area | |
| CN102004848A (en) | Method and system for improving calculation accuracy of photosynthetically active radiation (PAR) | |
| CN105403201B (en) | A kind of remote sensing images atmospheric path radiation acquisition methods based on pixel analysis | |
| CN111199092A (en) | Solar radiation remote sensing estimation method and system and data processing device | |
| CN109885808A (en) | A kind of near surface meteorological element calculation method | |
| Nigam et al. | Modeling temporal growth profile of vegetation index from Indian geostationary satellite for assessing in-season progress of crop area | |
| He et al. | Retrieval of Rugged Mountainous Areas Land Surface Temperature from High-spatial-resolution Thermal Infrared Remote Sensing Data |
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 |