CN115797785B - Method and device for determining farmland irrigation frequency based on microwave remote sensing - Google Patents
Method and device for determining farmland irrigation frequency based on microwave remote sensing Download PDFInfo
- Publication number
- CN115797785B CN115797785B CN202310084951.0A CN202310084951A CN115797785B CN 115797785 B CN115797785 B CN 115797785B CN 202310084951 A CN202310084951 A CN 202310084951A CN 115797785 B CN115797785 B CN 115797785B
- Authority
- CN
- China
- Prior art keywords
- data
- remote sensing
- irrigation
- precipitation
- soil moisture
- 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
- 238000003973 irrigation Methods 0.000 title claims abstract description 146
- 230000002262 irrigation Effects 0.000 title claims abstract description 146
- 238000000034 method Methods 0.000 title claims abstract description 37
- 239000002689 soil Substances 0.000 claims abstract description 115
- 238000001514 detection method Methods 0.000 claims abstract description 50
- 230000003287 optical effect Effects 0.000 claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000010801 machine learning Methods 0.000 claims abstract description 8
- 238000001556 precipitation Methods 0.000 claims description 60
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 30
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000012952 Resampling Methods 0.000 claims description 19
- 238000003860 storage Methods 0.000 claims description 12
- 230000010287 polarization Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 229930002875 chlorophyll Natural products 0.000 claims description 3
- 235000019804 chlorophyll Nutrition 0.000 claims description 3
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 claims description 3
- 238000002156 mixing Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 238000002329 infrared spectrum Methods 0.000 claims 2
- 238000012163 sequencing technique Methods 0.000 claims 1
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000001704 evaporation Methods 0.000 description 21
- 230000008020 evaporation Effects 0.000 description 21
- 230000004907 flux Effects 0.000 description 14
- 230000005855 radiation Effects 0.000 description 10
- 230000008859 change Effects 0.000 description 8
- 230000003746 surface roughness Effects 0.000 description 5
- 230000002354 daily effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000002310 reflectometry Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 208000005156 Dehydration Diseases 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000013505 freshwater Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000003621 irrigation water Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000012466 permeate Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
- 238000009834 vaporization Methods 0.000 description 1
- 230000008016 vaporization Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The embodiment of the invention provides a method and a device for determining farmland irrigation frequency based on microwave remote sensing, and relates to the technical field of farmland irrigation, wherein the method comprises the following steps: acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected; inverting according to the microwave remote sensing data to obtain soil moisture data, and inverting according to the optical remote sensing data to obtain evapotranspiration data; and inputting the soil moisture data, the evapotranspiration data and the moisture related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by training a machine learning component by taking historical soil moisture data, historical evapotranspiration data and historical moisture related data as samples. The scheme is favorable for improving the accuracy and timeliness of determining the farmland irrigation frequency.
Description
Technical Field
The invention relates to the technical field of farmland irrigation, in particular to a method and a device for determining farmland irrigation frequency based on microwave remote sensing.
Background
The accelerated growth of the world population brings great pressure to the global food supply. Enlarging cultivated land can offset the increase in food shortage in many areas, but such enlargement is limited due to adverse and profound effects on biodiversity. Thus, the expansion of irrigation area and improvement of agricultural technology are expected to make a great contribution to meeting the increasing demand for food. Irrigation agriculture accounts for 40% of the global crop yield, and is an important component of world food production, especially in semiarid and arid regions. In addition to maintaining sufficient soil moisture to promote healthy growth and optimal yield, irrigation also protects crops from frost during cold seasons and inhibits weed growth in the grain. In addition, irrigation is an important component of global water circulation and land water balance. Currently, it represents about 70% of the total global fresh water usage and about 90% of the consumer water. Thus, information on the extent, frequency, etc. of irrigation is very important for extensive research including grain safety and economic development, water and energy exchange between land surface and atmosphere, water resource allocation between agriculture and ecosystems, and agriculture-climate interaction and feedback.
The traditional farmland irrigation information detection method comprises the following steps: (1) The manual investigation statistical method is that the staff goes down the basic layer to investigate irrigation information and record by dotting; (2) According to the sensor monitoring method, the soil humidity sensor is arranged in the field, and irrigation information is detected by utilizing the change condition of the soil humidity. The traditional detection method is time-consuming and labor-consuming, can only obtain data of a plurality of points, has limited coverage range and poor representativeness, cannot quickly reflect space irrigation information, and influences the timeliness of farmland irrigation management measures.
Remote sensing is recognized as an effective tool for extracting irrigation information, and many studies are conducted on drawing an irrigation map by using remote sensing, but mainly focusing on drawing an irrigation area, frequency information is not considered. MODIS and Landsat et al data have been used for regional and global irrigation area mapping. The detection of irrigation information using optical telemetry data can be described with the following assumptions: water generated by irrigation permeates into soil, so that the soil is more moist, water can be accumulated on the surface, and the change of the remote sensing humidity index can be used as an index for detecting irrigation.
However, problems with detecting irrigation information by optical remote sensing means are: the farmland is usually that vegetation cover density is great, irrigates moisture and hides under the vegetation generally, and optical sensor hardly accurately catches moisture information, and then has influenced the accuracy that irrigates information detection.
Disclosure of Invention
In view of the above, the embodiment of the invention provides a farmland irrigation frequency determination method based on microwave remote sensing, which aims to solve the technical problem of low accuracy in determining irrigation information in the prior art. The method comprises the following steps:
acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
inverting according to the microwave remote sensing data to obtain soil moisture data, and inverting according to the optical remote sensing data to obtain evapotranspiration data;
and inputting the soil moisture data, the evapotranspiration data and the moisture related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by training a machine learning component by taking historical soil moisture data, historical evapotranspiration data and historical moisture related data as samples.
The embodiment of the invention also provides a farmland irrigation frequency determining device based on microwave remote sensing, which is used for solving the technical problem of low accuracy in determining irrigation information in the prior art. The device comprises:
the data acquisition module is used for acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
the data inversion module is used for inverting the microwave remote sensing data to obtain soil moisture data and inverting the optical remote sensing data to obtain evapotranspiration data;
the irrigation frequency calculation module is used for inputting the soil moisture data, the evapotranspiration data and the water related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by taking historical soil moisture data, historical evapotranspiration data and historical water related data as a sample training machine learning component.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes any farmland irrigation frequency determination method based on microwave remote sensing when executing the computer program so as to solve the technical problem of low accuracy in determining irrigation information in the prior art.
The embodiment of the invention also provides a computer readable storage medium which stores a computer program for executing any farmland irrigation frequency determination method based on microwave remote sensing, so as to solve the technical problem of low accuracy in determining irrigation information in the prior art.
Compared with the prior art, the beneficial effects that above-mentioned at least one technical scheme that this description embodiment adopted can reach include at least: the method comprises the steps that soil moisture data are obtained based on inversion of microwave remote sensing data of an irrigation area to be detected, the evapotranspiration data are obtained based on inversion of optical remote sensing data of the irrigation area to be detected, then the soil moisture data, the evapotranspiration data and water related data of the irrigation area to be detected are input into a detection model, the detection model can output irrigation frequency of the irrigation area to be detected, the detection model is obtained by taking historical soil moisture data, historical evapotranspiration data and historical water related data as sample training machine learning components, the determination of the irrigation frequency of the irrigation area to be detected based on the microwave remote sensing data is achieved, and compared with the scheme that only optical remote sensing data are adopted to detect irrigation information in the prior art, microwave remote sensing has the advantages of being strong in penetrability and sensitive to soil moisture change in all weather, and being capable of capturing the moisture information more accurately than the optical remote sensing data, and the use of the microwave remote sensing data is beneficial to improving accuracy and timeliness of determining the irrigation frequency of farmlands; meanwhile, the humidity index inverted by the soil moisture data instead of the optical remote sensing data is used as an ideal index for detecting irrigation information, and as the field evaporation amount after irrigation is obviously increased, evaporation data is introduced to determine irrigation frequency, and moisture related data is also introduced, so that the soil moisture data, the evaporation data and the moisture related data are combined to determine irrigation frequency, and the accuracy of determining the farmland irrigation frequency is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a farmland irrigation frequency determination method based on microwave remote sensing provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of a 1D-CNN model structure according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of training the above detection model according to an embodiment of the present invention;
FIG. 4 is a block diagram of a computer device according to an embodiment of the present invention;
fig. 5 is a block diagram of a device for determining a frequency of farmland irrigation based on microwave remote sensing according to an embodiment of the present invention.
Detailed Description
Embodiments of the present application are described in detail below with reference to the accompanying drawings.
Other advantages and effects of the present application will become apparent to those skilled in the art from the present disclosure, when the following description of the embodiments is taken in conjunction with the accompanying drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. The present application may be embodied or carried out in other specific embodiments, and the details of the present application may be modified or changed from various points of view and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
In an embodiment of the present invention, a method for determining a frequency of farmland irrigation based on microwave remote sensing is provided, as shown in fig. 1, and the method includes:
step S101: acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
step S102: inverting according to the microwave remote sensing data to obtain soil moisture data, and inverting according to the optical remote sensing data to obtain evapotranspiration data;
step S103: and inputting the soil moisture data, the evapotranspiration data and the moisture related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by training a machine learning component by taking historical soil moisture data, historical evapotranspiration data and historical moisture related data as samples.
As can be seen from the flow shown in fig. 1, in the embodiment of the present invention, the irrigation frequency of the detected irrigation area is determined based on the microwave remote sensing data, and compared with the scheme of detecting irrigation information only by using optical remote sensing data in the prior art, the microwave remote sensing has the advantages of all-day time, all weather, high resolution, strong penetrability and sensitivity to soil moisture change, and can more accurately capture moisture information than the optical remote sensing data, and the use of the microwave remote sensing data is beneficial to improving the accuracy of determining the irrigation frequency of farmlands; meanwhile, the humidity index inverted by the soil moisture data instead of the optical remote sensing data is used as an ideal index for detecting irrigation information, and as the field evaporation amount after irrigation is obviously increased, evaporation data is introduced to determine irrigation frequency, and moisture related data is also introduced, so that the soil moisture data, the evaporation data and the moisture related data are combined to determine irrigation frequency, and the accuracy of determining the farmland irrigation frequency is further improved.
In specific implementation, the irrigation area to be detected can be a farmland area with any area and needing irrigation information detection.
In particular, the microwave remote sensing data is remote sensing data which can be used for soil moisture inversion, for example, SAR radar satellite data comprisesThe method can be used for soil moisture inversion.
In specific implementation, the microwave remote sensing data can be preprocessed before the soil moisture data are obtained according to inversion of the microwave remote sensing data. For example, after acquiring a Sentinel-1 Ground Range Detection (GRD) High Resolution (HR) interferometric wide mode (IW) satellite image of an investigation region (i.e., an irrigation region to be detected), preprocessing may be performed using Sentinel-1Toolbox (S1 TBX) software provided by the european union. The preprocessing step may include (1) applying orbit file (2) GRD boundary noise removal (3) thermal noise cancellation (4) radiometric calibration (5) geographic correction to obtain observations and angles of incidence for VV and VH polarizations.
In particular, in order to improve the accuracy of determining the farmland irrigation frequency, in this embodiment, a method for inverting soil moisture data is provided, for example, determining the soil dielectric constant according to the radar backscattering coefficient, the radar incident angle and the polarization mode of the microwave remote sensing data; and converting the dielectric constant of the soil into the water content of the soil volume by adopting a dielectric mixing model.
In practice, the radar backscattering coefficient is mainly affected by factors such as the dielectric constant of soil (mainly depending on soil moisture), the surface roughness and the vegetation coverage. Given that the surface roughness and vegetation status are constant over time, the change in radar backscattering coefficient can be attributed to a change in the dielectric constant of the soil (soil moisture). The soil moisture and its change information can be extracted by repeated observation at multiple times. For 2 time phases T 1 and T2 The ratio of radar backscatter coefficients acquired in these 2 phases can be approximately seen as a function of the soil dielectric constant, radar incidence angle, and polarization, provided that the roughness of the earth's surface remains unchanged during this time interval and that the effect of the earth's surface roughness and vegetation on the radar backscatter coefficients is multiplicative. This model, also known as the Alpha approximation model, can be expressed as:
in the formula Representing radar backscatter coefficients (energy); />Is the radar incident angle; />Is the relative dielectric constant of soil; />Representing the time of radar data acquisition; />Is polarization amplitude, and is a function of radar incidence angle and soil dielectric constant; PP represents the polarization mode, HH or VV. It was found that VV polarization +.>Is more sensitive to the difference of the water content of the soil, thus selecting +.>Inverting farmland soil moisture.
If a small disturbance model is used to describe the scattering of the earth's surface, PP is VV,the polarization amplitude can be expressed as:
after acquiring the SAR images of 2 phases, an observation equation can be obtained according to equation (1):
to ensure the rationality of the hypothesis, 2 images in succession are typically used to construct an observation equation. For SAR images of N-scene continuous observation, the following can be formedThe equations, which make up the equation set, are: />
For equation set (4), there are N soil moisture unknowns and there are innumerable solutions. Solving for a system of equationsNeeds to be matched withAnd the value range of the method is limited, and then a boundary constraint least square algorithm is adopted for solving. For a given radar incidence angle and soil moisture range, +.>The value constraint condition of (2) is expressed as follows:
in the formula Respectively represent polarization amplitude +.>Minimum and maximum values given radar incidence angle and soil moisture range. Solving the equation set by adopting boundary constraint least square to obtain polarization amplitudeFurther, the dielectric constant of soil can be obtained from the formula (2)>Finally, the dielectric constant is converted into the water content of the soil volume by adopting a dielectric mixed model>。
In specific implementation, the spatial resolution of the soil moisture content data obtained by inversion can be determined according to requirements, for example, the spatial resolution of the soil moisture content data obtained by inversion can be 10m, and the time interval can be 6 days.
In a specific implementation, in order to further improve the accuracy of determining the frequency of farmland irrigation, in this embodiment, a method for inverting the evaporation data is proposed, for example, the optical remote sensing data and the meteorological data are input into a SEBAL model, and the SEBAL model inverts to obtain the evaporation data of the irrigation area to be detected based on the principle of surface energy balance.
In specific implementation, the optical remote sensing data may be MODIS (medium resolution imaging spectrometer) data. As the original MODIS data is stored by adopting a hierarchical data format hdf format and a sinusoidal projection SIN, the MRT software provided by NASA is needed to convert the hdf-format MODIS data into a Geotiff format and perform projection conversion, resampling and other operations during use. For example, in inverting the vapor emission data in the SEBAL (land energy balance) model, the MODIS data used may include MOD11A1, MOD13A2, and MCD43B3.
In specific implementation, the SEBAL (Surface EnergyBalance Algorithms for Land) model is a model for inverting regional evapotranspiration based on the earth surface energy balance principle, estimates the instantaneous evapotranspiration amount at the moment of image acquisition by an energy balance remainder method, and calculates the total evapotranspiration amount all the day by converting the evapotranspiration value into the daily evapotranspiration amount by combining the instantaneous evapotranspiration amount and the evapotranspiration ratio assuming that the evapotranspiration ratio is constant in one day. According to the method, during calculation, only optical remote sensing images and a small amount of meteorological data (air temperature and wind speed) are input into the SEBAL model, so that the surface net radiation quantity (Rn), the soil heat flux (G), the sensible heat flux (H) and the latent heat flux (λET) can be obtained. The calculation formula of the earth surface energy balance principle in the SEBAL model is as follows:
the MODIS product data can directly provide characteristic parameters such as surface temperature, normalized vegetation index and surface reflectivity for the calculation of the SEBAL model.
The following is a specific calculation method of each surface parameter in the SEBAL model.
(1) Net radiation quantity Rn
Rn is the sum of various radiations obtained from the earth's surface, and can be calculated by the earth's surface albedo, earth's surface emissivity, etc., and the calculation formula is as follows:
for incident solar short-wave radiation, < >>For long wave radiation upstream +.>For down-going long-wave radiation, in +.>;/>Is the reflectivity of the earth surface>Is the emissivity of the earth's surface.
upstream long wave radiationAnd down long wave radiation +.>Can be obtained from the steven-boltzmann law, and the calculation formula is as follows:
wherein Is sun constant, takes the value ∈ ->;/>Is the zenith angle of the sun; />Is the atmospheric emissivity; />Is the emissivity of the earth surface; />Is a Stefin-Boltzmann constant;/>Is the air temperature; />Is the surface temperature; />Is the distance between the earth and the day in astronomical units; j is the serial number of the image acquisition date arranged in the solar calendar. />
Is the unidirectional transmittance of the atmosphereIts value is about +.>Herein will->Converting the sample to a research area surface for calculation, wherein the calculation formula is as follows:
z is the elevation of the research area, and the unit is m, and can be obtained from the remote sensing image.
Reflectivity of earth's surfaceIs an amount characterizing the reflection of solar radiation from the earth's surface, which varies with time-space variations. The earth surface albedo can be obtained by correcting the albedo outside the atmosphere through atmospheric radiation. For MODIS data, the earth surface albedo is the earth surface reflectivity of visible light and near infrared bands, the conversion coefficient is defined as the proportion of each narrow band incident luminous flux to the whole wide band incident flux, and the calculation formula of the earth surface albedo is obtained:
Emissivity ofIs a dimensionless quantity, and the value is between 0 and 1. Emissivity of an object is an expression of the emissivity of the object, herein using normalizationThe vegetation index NDVI represents emissivity, and the calculation formula is as follows:
(2) Soil heat flux G
Soil heat flux refers to the portion of energy stored in the soil and vegetation by conduction through the surface temperatureThe statistical relationship of (2) is obtained by the following formula:
(3) Heat-sensitive flux H
The heat flux is a function of the atmospheric stability, wind speed and surface roughness, and refers to the fraction of energy dissipated into the atmosphere by conduction and convection, as follows:
is air density in ∈>;/>The specific heat of air is fixed pressure and is valued;/>Is the temperature difference between the near ground and the ground at a certain height; />For aerodynamic resistance in>。
Aerodynamic drag rah requires introduction of wind speed,Parameters such as ground elevation and the like are obtained by calculating air density, friction wind speed and surface roughness. The calculation formula is as follows:
wherein, in practical applicationIs the friction wind speed at neutral stability; k is Von Karman constant, and takes on a value of 0.41; />Is +.>Wind speed at (I/O)>Taking 200m or 100m; />Is a weather station +.>The measured wind speed is expressed as +.>。
dT is calculated by using a Moning-Obuhuff theory, and a linear relation between the surface temperature and dT is established by selecting pixels (namely 'cold spots') with vigorous vegetation growth, low temperature and sufficient water supply and pixels (namely 'hot spots') without vegetation coverage, drying and high temperature on a remote sensing image, namely dT=a, and dT is obtained by cyclic recursion calculation, wherein 'cold spots' can be areas or water bodies completely covered by vegetation, and the evapotranspiration amount on 'hot spots' is approximately 0 and can be ignored.
(4) Solar evapotranspiration ET24
The Rn, G and H obtained by the calculation are brought into an energy balance principle formula, the instantaneous latent heat flux lambda ET is calculated, lambda in the latent heat flux is vaporization latent heat and is a constant, ET in the latent heat flux is required evaporation capacity, the instantaneous latent heat flux is converted into daily-scale evaporation capacity through evaporation ratio, and the soil heat flux G can be ignored when the daily evaporation capacity is calculated, so the daily evaporation capacity ET24 is calculated according to the following formula:
In specific implementation, the water related data of the irrigation area to be detected may include data reflecting vegetation water stress and/or data affecting irrigation water, for example, may include greenness index and precipitation amount.
In particular, the greenness indexIs the ratio of the near infrared to the green spectrum, which is more sensitive to chlorophyll content than other vegetation indices and is therefore considered a good indicator for monitoring vegetation water stress. Adding spatial precipitation data can exclude interference of precipitation on spatial irrigation information mapping.
In particular, in order to further improve accuracy of determining the frequency of farmland irrigation, in this embodiment, it is proposed that before determining the frequency of farmland irrigation, the data of soil moisture data, evapotranspiration data, greenness index and precipitation are matched on a spatial scale, for example, resampling is performed on the soil moisture data, the evapotranspiration data, the greenness index and the precipitation according to a preset spatial resolution, so that the spatial resolutions of the soil moisture data, the evapotranspiration data, the greenness index and the precipitation after resampling are all preset spatial resolutions, and then the resampled soil moisture data, the evapotranspiration data, the greenness index and the precipitation are input into the detection model.
In practice, due to irrigation frequency (village scale), evapotranspiration (1 km), soil moisture (10 m), precipitationAnd the greenness index GI (10 m) has different spatial resolution and data structure, and resampling is needed to match the data sets on spatial scale in order to improve the accuracy of irrigation frequency, so that soil moisture data, evapotranspiration data, greenness index and precipitation can be resampled according to the preset spatial resolution, and the specific size of the preset spatial resolution can be determined according to the requirement, for example, the preset spatial resolutionThe rate may be 10m resolution, and data coarser than 10m resolution is resampled to 10m resolution.
In practice, each resampled precipitation pixel may be covered by a gridIn order to accurately determine the precipitation of each resampled precipitation pixel, either completely or by several grid sections, in order to improve the accuracy of the precipitation data, in this embodiment, it is proposed to calculate the precipitation amount of each resampled pixel by the following formula:
wherein ,precipitation of the pixel after resampling, +.>Is->A grid of said resampled pixels comprising an area percentage of said resampled pixels, ">Is->And the precipitation amount of the grids forming the pixel after resampling is n, and the total number of the grids forming the pixel after resampling is n.
In practice, after resampling the irrigation frequency (village scale) to 10m, a portion of 10m pixels still appear at village boundaries and are covered by several different village grids, and the irrigation frequency of a portion of pixels can be calculated by using the area percentage weighting formula (24) and then rounded.
In particular, in order to further improve the accuracy of determining the frequency of farmland irrigation, in this embodiment, it is proposed to interpolate the data such as the soil moisture data, the evapotranspiration data, the greenness index, and the precipitation amount into equidistant time-series data sets, for example, linearly interpolate the resampled greenness index and precipitation amount so that the linearly interpolated greenness index and precipitation amount match the time-series of the soil moisture data, the evapotranspiration data does not need to be interpolated every day, and then sort the soil moisture data, the evapotranspiration data, the linearly interpolated greenness index, and the linearly interpolated precipitation amount by time-series, to obtain the time-series-ordered and spatially-resolved uniform soil moisture data, the evapotranspiration data, the greenness index, and the precipitation amount corresponding to each pixel, and further input the time-series-ordered and spatially-resolved soil moisture data, the evapotranspiration index, and the detection model corresponding to each pixel.
In practice, the above time series is established at the pixel level. Because the time interval of the SAR inversion soil moisture data is 6 days, linear interpolation is carried out on the greenness index GI and the rainfall Pre so as to match the SAR inversion soil moisture data, the evapotranspiration is all daily, and interpolation processing is not needed. To remove noise in time series, use is made ofThe filter performs smoothing treatment on the time sequence, and finally performs normalization treatment on soil moisture SM, evapotranspiration quantity ET, greenness index GI and precipitation quantity Pre.
In particular, the detection model can be trainedThe network is obtained, as shown in fig. 2, three columns of data corresponding to the characteristics are respectively the soil volume water content, the evapotranspiration and the water related data which are sequenced on the time sequence, and the data in the thick line frame are represented as a data matrix of 3X3 input into the convolution kernelThe mapping relation of the detection model to the irrigation frequency (IrrigationFrequency, IF) is as follows:
to avoid the occurrence of decimal numbers in the irrigation frequency during the detection process, the output of the detection model needs to be converted into the form of a supervised classification problem, rather than a regression problem. Then, a relation model of irrigation frequency IF and soil moisture SM, evapotranspiration quantity ET, greenness index GI and precipitation quantity Pre is established.
In the implementation, as shown in fig. 3, during the process of training the detection model, the microwave remote sensing data of the irrigation area to be detected in the past period of time is collected, and the microwave remote sensing data is used for historical useData as an example byObtaining the water content of the historical soil volume through data inversion, obtaining the historical evaporation capacity through SEBAL model inversion by adopting MODIS, resampling the historical precipitation capacity (rainfall), the historical greenness index, the historical soil volume water content and the historical evaporation capacity according to the preset spatial resolution, realizing space size matching, sorting the resampled historical precipitation capacity (rainfall), the historical greenness index, the historical soil volume water content and the historical evaporation capacity according to the historical time sequence, realizing time sequence matching, obtaining a data set corresponding to each pixel time sequence, finally taking irrigation range data, irrigation frequency, the historical precipitation capacity (rainfall), the historical greenness index, the historical soil volume water content and the historical evaporation capacity of an irrigation area to be detected as samples, taking different characteristic combinations of pixels corresponding to the time sequence as a training sample set, and inputting the training sample set to the pixel time sequence>The model is trained by using a space cross validation method, and is circulated for 10 times each time90% of the samples were extracted for training, 10% for verification, and finally the average error was calculated.
In specific implementation, in the process of training the detection model, a historical related data set in a first preset duration can be used as a sample for training, in the process of determining the farmland irrigation frequency of an irrigation area to be detected by using the detection model, a related data set in a second preset duration can be used for detection, the first preset duration and the second preset duration can be the same or different, and specific durations of the first preset duration and the second preset duration can be determined according to requirements, for example, a plurality of months, a half year, a year and the like.
In this embodiment, a computer device is provided, as shown in fig. 4, including a memory 401, a processor 402, and a computer program stored in the memory and capable of running on the processor, where the processor implements any of the above-mentioned farmland irrigation frequency determination methods based on microwave remote sensing when executing the computer program.
In particular, the computer device may be a computer terminal, a server or similar computing means.
In this embodiment, a computer-readable storage medium is provided, in which a computer program for executing any of the above-described method for determining a frequency of farmland irrigation based on microwave remote sensing is stored.
In particular, computer-readable storage media, including both permanent and non-permanent, removable and non-removable media, may be used to implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable storage media, as defined herein, does not include transitory computer-readable media (transshipment) such as modulated data signals and carrier waves.
Based on the same inventive concept, the embodiment of the invention also provides a farmland irrigation frequency determining device based on microwave remote sensing, as described in the following embodiment. Because the principle of solving the problem of the farmland irrigation frequency determining device based on the microwave remote sensing is similar to that of the farmland irrigation frequency determining method based on the microwave remote sensing, the implementation of the farmland irrigation frequency determining device based on the microwave remote sensing can be referred to the implementation of the farmland irrigation frequency determining method based on the microwave remote sensing, and repeated parts are omitted. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements the intended function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
Fig. 5 is a block diagram of a device for determining a frequency of farmland irrigation based on microwave remote sensing according to an embodiment of the present invention, as shown in fig. 5, the device includes:
the data acquisition module 501 is used for acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
the data inversion module 502 is configured to invert the microwave remote sensing data to obtain soil moisture data, and invert the microwave remote sensing data to obtain evapotranspiration data;
the irrigation frequency calculation module 503 is configured to input the soil moisture data, the evapotranspiration data, and the related data of the water in the irrigation area to be detected into a detection model, and the detection model outputs the irrigation frequency of the irrigation area to be detected, where the detection model is obtained by training a machine learning component with historical soil moisture data, historical evapotranspiration data, and historical related data of the water as samples.
In one embodiment, a data inversion module includes:
the soil moisture data inversion unit is used for determining the dielectric constant of the soil according to the radar backscattering coefficient, the radar incidence angle and the polarization mode of the microwave remote sensing data; and converting the dielectric constant of the soil into the water content of the soil volume by adopting a dielectric mixing model.
In one embodiment, the data inversion module further comprises:
the evapotranspiration data inversion unit is used for inputting the optical remote sensing data and the meteorological data into an SEBAL model, and the SEBAL model is used for inverting the evapotranspiration data of the irrigation area to be detected based on the earth surface energy balance principle.
In one embodiment, the irrigation frequency calculation module is configured to resample the soil moisture data, the evapotranspiration data, the greenness index and the precipitation according to a preset spatial resolution, and the resampled soil moisture data, the evapotranspiration data, the greenness index and the precipitation are input into the detection model.
In one embodiment, the apparatus further comprises:
a precipitation determination module for calculating the precipitation of each pixel after resampling by the following formula:
wherein ,precipitation of the pixel after resampling, +.>Is->A grid of said resampled pixels comprising an area percentage of said resampled pixels, ">Is->And the precipitation amount of the grids forming the pixel after resampling is n, and the total number of the grids forming the pixel after resampling is n.
In one embodiment, the irrigation frequency calculation module is further configured to perform linear interpolation on the resampled greenness index and the precipitation, match the linear interpolation greenness index and the precipitation with the time sequence of the soil moisture data, sort the soil moisture data, the evapotranspiration data, the linear interpolation greenness index and the linear interpolation precipitation according to the time sequence, and input the soil moisture data, the evapotranspiration data, the greenness index and the precipitation corresponding to each pixel, which are sorted according to the time sequence and have consistent spatial resolution, into the detection model.
The embodiment of the invention realizes the following technical effects: the method has the advantages that the irrigation frequency of the detected irrigation area is determined based on the microwave remote sensing data, and compared with the scheme that only optical remote sensing data is adopted to detect irrigation information in the prior art, the microwave remote sensing has the advantages of being all-day, all-weather, high in resolution, strong in penetrability and sensitive to soil moisture change, moisture information can be captured more accurately than the optical remote sensing data, and the accuracy of determining the irrigation frequency of farmlands can be improved due to the use of the microwave remote sensing data; meanwhile, the humidity index inverted by the soil moisture data instead of the optical remote sensing data is used as an ideal index for detecting irrigation information, and as the field evaporation amount after irrigation is obviously increased, evaporation data is introduced to determine irrigation frequency, and moisture related data is also introduced, so that the soil moisture data, the evaporation data and the moisture related data are combined to determine irrigation frequency, and the accuracy of determining the farmland irrigation frequency is further improved.
It will be apparent to those skilled in the art that the modules or steps of the embodiments of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a storage device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than what is shown or described, or they may be separately fabricated into individual integrated circuit modules, or a plurality of modules or steps in them may be fabricated into a single integrated circuit module. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, and various modifications and variations can be made to the embodiments of the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A farmland irrigation frequency determining method based on microwave remote sensing is characterized by comprising the following steps:
acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
inverting according to the microwave remote sensing data to obtain soil moisture data, and inverting according to the optical remote sensing data to obtain evapotranspiration data;
inputting the soil moisture data, the evapotranspiration data and the moisture related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by taking historical soil moisture data, historical evapotranspiration data and historical moisture related data as sample training machine learning components, the moisture related data of the irrigation area to be detected comprises a greenness index and a precipitation amount, the greenness index is the ratio of near infrared spectrum to the greenness spectrum, and the greenness index influences the chlorophyll content of vegetation;
inputting the soil moisture data, the evapotranspiration data and the moisture related data of the irrigation area to be detected into a detection model, wherein the detection model comprises the following steps:
resampling the soil moisture data, the evapotranspiration data, the greenness index and the precipitation according to a preset spatial resolution, and inputting the resampled soil moisture data, the evapotranspiration data, the greenness index and the precipitation into the detection model;
further comprises:
the precipitation amount of each pixel after resampling is calculated by the following formula:
wherein ,precipitation of the pixel after resampling, +.>Is->A grid of said resampled pixels comprising an area percentage of said resampled pixels, ">Is->Precipitation amount of grids forming the resampled pixel, wherein n is total number of grids forming the resampled pixel;
inputting the resampled soil moisture data, the evapotranspiration data, the greenness index and the precipitation into the detection model, wherein the method comprises the following steps of:
and carrying out linear interpolation on the resampled greenness index and the precipitation, wherein the greenness index and the precipitation after linear interpolation are matched with the soil moisture data time sequence so as to interpolate the soil moisture data, the evapotranspiration data, the greenness index and the precipitation into data sets with equidistant time sequences, then sequencing the soil moisture data, the evapotranspiration data, the linearly interpolated greenness index and the linearly interpolated precipitation according to the time sequences, and inputting the soil moisture data, the evapotranspiration data, the greenness index and the precipitation which are corresponding to each pixel and sequenced according to the time sequences and have consistent spatial resolution into the detection model.
2. The method for determining the frequency of farmland irrigation based on microwave remote sensing according to claim 1, wherein the inversion of the microwave remote sensing data to obtain soil moisture data comprises the steps of:
determining the dielectric constant of soil according to the radar backscattering coefficient, the radar incident angle and the polarization mode of the microwave remote sensing data;
and converting the dielectric constant of the soil into the water content of the soil volume by adopting a dielectric mixing model.
3. The method for determining a frequency of farmland irrigation based on microwave remote sensing according to claim 1, wherein obtaining evapotranspiration data based on inversion of the optical remote sensing data comprises:
inputting the optical remote sensing data and the meteorological data into an SEBAL model, and inverting the SEBAL model based on a ground surface energy balance principle to obtain the evapotranspiration data of the irrigation area to be detected.
4. A device for determining a frequency of farmland irrigation based on microwave remote sensing, comprising:
the data acquisition module is used for acquiring microwave remote sensing data and optical remote sensing data of an irrigation area to be detected;
the data inversion module is used for inverting the microwave remote sensing data to obtain soil moisture data and inverting the optical remote sensing data to obtain evapotranspiration data;
the irrigation frequency calculation module is used for inputting the soil moisture data, the evapotranspiration data and the water related data of the irrigation area to be detected into a detection model, and outputting the irrigation frequency of the irrigation area to be detected by the detection model, wherein the detection model is obtained by taking historical soil moisture data, historical evapotranspiration data and historical water related data as sample training machine learning components, the water related data of the irrigation area to be detected comprises a greenness index and a precipitation amount, the greenness index is the ratio of near infrared spectrum to greenness spectrum, and the greenness index influences the chlorophyll content of vegetation;
the irrigation frequency calculation module is used for resampling the soil moisture data, the evapotranspiration data, the greenness index and the precipitation according to a preset spatial resolution, and inputting the resampled soil moisture data, the evapotranspiration data, the greenness index and the precipitation into the detection model;
the apparatus further comprises:
a precipitation determination module for calculating the precipitation of each pixel after resampling by the following formula:
wherein ,precipitation of the pixel after resampling, +.>Is->A grid of said resampled pixels comprising an area percentage of said resampled pixels, ">Is->Precipitation amount of grids forming the resampled pixel, wherein n is total number of grids forming the resampled pixel;
the irrigation frequency calculation module is further configured to perform linear interpolation on the resampled greenness index and the precipitation, perform linear interpolation on the greenness index and the precipitation after linear interpolation, match the time series of the soil moisture data with the time series of the soil moisture data, interpolate the soil moisture data, the evapotranspiration data, the greenness index and the precipitation into a data set with equidistant time series, sort the soil moisture data, the evapotranspiration data, the linearly interpolated greenness index and the linearly interpolated precipitation according to the time series, and input the time series-ordered and spatial resolution-consistent soil moisture data, the evapotranspiration data, the greenness index and the precipitation corresponding to each pixel into the detection model.
5. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method for determining the frequency of farmland irrigation based on microwave remote sensing according to any of claims 1to 3 when executing the computer program.
6. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method for determining a frequency of farmland irrigation based on microwave remote sensing as claimed in any one of claims 1to 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310084951.0A CN115797785B (en) | 2023-02-09 | 2023-02-09 | Method and device for determining farmland irrigation frequency based on microwave remote sensing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310084951.0A CN115797785B (en) | 2023-02-09 | 2023-02-09 | Method and device for determining farmland irrigation frequency based on microwave remote sensing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115797785A CN115797785A (en) | 2023-03-14 |
CN115797785B true CN115797785B (en) | 2023-05-05 |
Family
ID=85430602
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310084951.0A Active CN115797785B (en) | 2023-02-09 | 2023-02-09 | Method and device for determining farmland irrigation frequency based on microwave remote sensing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115797785B (en) |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110610054B (en) * | 2019-09-23 | 2021-03-23 | 北京师范大学 | Method and system for constructing cuboid inversion model of soil humidity |
CN111079256B (en) * | 2019-11-18 | 2023-08-29 | 珠江水利委员会珠江水利科学研究院 | Irrigation water effective utilization coefficient measuring and calculating method based on remote sensing |
BR112022018681A2 (en) * | 2020-03-17 | 2022-12-06 | Heartland Ag Tech Inc | SYSTEMS AND METHODS FOR PREDICTIVE MAINTENANCE OF IRRIGATION SYSTEMS |
CN113470175B (en) * | 2021-06-11 | 2023-02-28 | 北京大学 | Irrigation area mapping method based on optical trapezoid model |
CN115530054A (en) * | 2022-10-12 | 2022-12-30 | 河北省科学院应用数学研究所 | Irrigation control method and device, electronic equipment and storage medium |
-
2023
- 2023-02-09 CN CN202310084951.0A patent/CN115797785B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN115797785A (en) | 2023-03-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zeng et al. | A review of vegetation phenological metrics extraction using time-series, multispectral satellite data | |
Yang et al. | Daily Landsat-scale evapotranspiration estimation over a forested landscape in North Carolina, USA, using multi-satellite data fusion | |
Huang et al. | Jointly assimilating MODIS LAI and ET products into the SWAP model for winter wheat yield estimation | |
Peng et al. | Spatial downscaling of satellite soil moisture data using a vegetation temperature condition index | |
Zhao et al. | Normalization of the temporal effect on the MODIS land surface temperature product using random forest regression | |
Hu et al. | Improved estimates of monthly land surface temperature from MODIS using a diurnal temperature cycle (DTC) model | |
Maselli et al. | Environmental monitoring and crop forecasting in the Sahel through the use of NOAA NDVI data. A case study: Niger 1986–89 | |
Bala et al. | Correlation between potato yield and MODIS‐derived vegetation indices | |
CN108169161A (en) | A kind of corn planting regional soil humidity appraisal procedure based on modified MODIS indexes | |
Imukova et al. | Determining the spatial and temporal dynamics of the green vegetation fraction of croplands using high-resolution RapidEye satellite images | |
WO2018107245A1 (en) | Detection of environmental conditions | |
Patel et al. | Satellite-derived vegetation temperature condition index to infer root zone soil moisture in semi-arid province of Rajasthan, India | |
Yeom et al. | Mapping rice area and yield in northeastern Asia by incorporating a crop model with dense vegetation index profiles from a geostationary satellite | |
Lee et al. | A comparative study on generating simulated Landsat NDVI images using data fusion and regression method—the case of the Korean Peninsula | |
Nourani et al. | Hydrological model parameterization using NDVI values to account for the effects of land cover change on the rainfall–runoff response | |
Verma et al. | Modelling of sugarcane yield using LISS-IV data based on ground LAI and yield observations | |
Zhao et al. | Capability of existing drought indices in reflecting agricultural drought in China | |
Wang et al. | Research on cropping intensity mapping of the Huai River Basin (China) based on multi-source remote sensing data fusion | |
Shamloo et al. | Agricultural drought survey using MODIS-based image indices at the regional scale: case study of the Urmia Lake Basin, Iran | |
Yang et al. | Capturing the spatiotemporal variations in the gross primary productivity in coastal wetlands by integrating eddy covariance, Landsat, and MODIS satellite data: A case study in the Yangtze Estuary, China | |
Yamamoto et al. | Uncertainty quantification in land surface temperature retrieved from Himawari-8/AHI data by operational algorithms | |
Xu et al. | Exploring the potential of Gaofen-1/6 for crop monitoring: Generating daily decametric-resolution leaf area index time series | |
Mekonnen | Satellite remote sensing for soil moisture estimation: Gumara catchment, Ethiopia | |
Bocco et al. | Estimating soybean ground cover from satellite images using neural-networks models | |
CN115797785B (en) | Method and device for determining farmland irrigation frequency based on microwave remote sensing |
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 |