CN114492726A - Forest combustible water content inversion algorithm based on remote sensing data - Google Patents

Forest combustible water content inversion algorithm based on remote sensing data Download PDF

Info

Publication number
CN114492726A
CN114492726A CN202111478086.5A CN202111478086A CN114492726A CN 114492726 A CN114492726 A CN 114492726A CN 202111478086 A CN202111478086 A CN 202111478086A CN 114492726 A CN114492726 A CN 114492726A
Authority
CN
China
Prior art keywords
water content
remote sensing
sensing data
canopy
combustible
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.)
Pending
Application number
CN202111478086.5A
Other languages
Chinese (zh)
Inventor
高德民
郭在军
业巧林
牛海峰
李贾
王海娇
李云涛
闫海平
王丹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Forestry University
Original Assignee
Nanjing Forestry University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Forestry University filed Critical Nanjing Forestry University
Priority to CN202111478086.5A priority Critical patent/CN114492726A/en
Publication of CN114492726A publication Critical patent/CN114492726A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention discloses a forest combustible water content inversion algorithm based on remote sensing data, and belongs to the field of deep learning. The method adopts an MLP model to establish a relation algorithm between the remote sensing spectral reflectivity and the water content of the forest canopy vegetation and the earth surface litter, the remote sensing data is preprocessed, the spectral reflectivity is used for inverting the water content of the canopy vegetation and the earth surface litter, and the fitting degree of the finally established model can reach about 0.8. The algorithm also provides an optimization scheme for poor penetrability of optical remote sensing between the canopy and the ground surface, and provides a theoretical basis for large-scale determination of the canopy and the water content of the ground surface litter by a remote sensing estimation method.

Description

Forest combustible water content inversion algorithm based on remote sensing data
Technical Field
The invention belongs to the field of deep learning, and particularly relates to a forest combustible water content inversion algorithm based on remote sensing data.
Background
At present, the method for measuring the water content of forest combustible mainly comprises a balanced water content method, a meteorological element method and a remote sensing estimation method. The method for balancing the water content is characterized in that the water content change in a period of time is predicted through a model by comprehensively considering the water content balance, the initial water content of combustible materials, the time and a time-lag factor under an ideal environment. The meteorological element regression method is mainly used for establishing statistical models of various meteorological factors and water content of combustible materials, and mainly comprises a fire hazard scale model method, a comprehensive index method, a Rothermel model, a BEHAVE model and the like. The remote sensing estimation method is generally used along with the development of remote sensing technology, and along with the rapid development of computers and the improvement of satellite technology, the application direction of the remote sensing technology is already expanded to the aspects of detecting soil, vegetation moisture and the like in the 70 s of the 20 th century. The hyperspectral technology appeared in the 90 s of the 20 th century can acquire spectral data of various regions by using optical sensors, and spectral information mainly comes from combustible materials.
Compared with the other two methods, the remote sensing estimation method has the advantages of low cost and large measurement scale. At present, the remote sensing spectrum technology is used for estimating the water content of the live combustible, but in practice, the water content of the dead combustible is lower than that of the live combustible, so that the influence on fire is large, and therefore the remote sensing technology is used for estimating the water content of canopy vegetation and ground surface litter and has important significance in fire prediction. The traditional estimation of the water content of the combustible materials in the region is based on a large amount of artificially measured data, and although the method is high in accuracy, the efficiency is very low, a large amount of manpower and material resources are consumed, and certain damage is caused to the ecology of the region.
Therefore, a new method which is convenient to obtain, high in timeliness and long in detection distance is needed to provide data for inversion of water content of regional combustible materials.
Disclosure of Invention
In view of the above problems in the prior art, an object of the present invention is to provide an inversion algorithm for water content of forest combustible based on remote sensing data.
In order to solve the problems, the technical scheme adopted by the invention is as follows:
a forest combustible water content inversion algorithm based on remote sensing data comprises the following steps:
step 1: extracting remote sensing data, preprocessing the remote sensing data, and cutting and screening sample points;
step 2: proceeding to the sample points marked in the step 1 for sampling on the spot;
and step 3: performing inversion of the water content of the combustible materials in the canopy by adopting an MLP deep learning model;
and4, step 4: and performing surface combustible water content inversion by adopting an MLP deep learning model.
The steps for preprocessing the remote sensing data are as follows:
step 1: sequentially carrying out atmospheric correction on the 10m resolution wave band and the 20m resolution wave band to respectively obtain two groups of L2A-level data of 10m resolution and two groups of 20m resolution wave bands;
step 2: resampling the data obtained in the step 1 to be a 10m resolution wave band by using a nearest neighbor method;
and step 3: and (3) performing band synthesis on the data obtained in the step (2) to generate a true color image.
The field sampling in the step 2 is to collect canopy vegetation and earth surface litter at the sample point, measure and record information such as sample point longitude and latitude, tree species, temperature and humidity, atmospheric pressure and the like, and calculate the moisture content of all samples according to the following formula:
absolute water content
Figure BDA0003394321490000021
Relative water content
Figure BDA0003394321490000022
Wherein, WHIs the wet weight (g) of combustible materials, WDIs combustible dry weight (g).
And the inversion of the water content of the canopy is that red light, green light, near infrared and two short wave infrared in the original data are selected as input ends by using an MLP deep learning model, the inversion of the water content of the combustible substance of the canopy is directly carried out, and machine learning is carried out on the water content of the canopy and the data obtained by actual sampling for multiple times, so that the model is optimized.
The inversion of the water content of the surface combustible is that the original data are processed by using a two-way reflection distribution function to obtain multi-angle remote sensing data, and the inversion is disclosed as follows:
Figure BDA0003394321490000023
where λ is the wavelength, θiIs the angle between the incident direction of the sunlight and the zenith angle, thetarIs the angle between the observation direction and the zenith angle,
Figure BDA0003394321490000024
and
Figure BDA0003394321490000025
the angles of the incident direction and the observation direction in the direction are respectively indicated;
and substituting the obtained data into a 4-scale model in the radiation transmission model, wherein the reflectivity relation of the 4-scale model is as follows:
R＝RTKT+RGKG+RZTKZT+RZGKZG
wherein: rTRepresenting the reflectivity of the canopy illuminated surface;
KTrepresenting the probability of the sensor observing the ground illuminated surface
RGRepresenting the surface illumination surface reflectivity;
KGrepresenting the probability of the sensor observing the ground illuminated surface
RZTRepresenting the reflectivity of the background surface of the canopy;
KZTrepresenting the probability of the sensor observing the background surface of the canopy
RZGRepresenting the reflectivity of the ground background surface;
KZGrepresenting the probability of the sensor observing the ground background surface;
then obtaining surface remote sensing data by using the following formula;
Figure BDA0003394321490000031
in the formula: m is a multiple scattering factor, and A, B and C are relational expressions of M and K components;
and finally, selecting red light, green light, near infrared and two short wave infrared in the obtained surface remote sensing data as input ends in an MLP deep learning model, carrying out inversion on the water content of the surface combustible, and carrying out machine learning with the data obtained by actual sampling for multiple times to optimize the model.
Compared with the prior art, the method is based on the MLP deep learning model, inversion is carried out on the water content by adopting remote sensing data, only a small amount of actually measured data is adopted for testing the accuracy of the model, remote sensing data is adopted, large-range remote detection can be carried out, the acquisition is convenient and fast, the timeliness is high, and a novel method is provided for inverting the water content of the combustible materials in the region
Drawings
FIG. 1 is a schematic flow chart of the algorithm of the present invention;
FIG. 2 is a true color remote sensing image;
FIG. 3 is a sample area spline point distribution;
FIG. 4 is a diagram of an MLP deep learning model architecture;
FIG. 5 is a diagram comparing the water content model training of the combustible substance in the canopy with the actual error;
FIG. 6 is a comparison graph of predicted values and true values of the canopy MLP model:
FIG. 7 is a graph comparing the training of the model of the moisture content of the combustible substances on the ground surface with the actual error;
FIG. 8 is a graph comparing predicted values and true values of a surface MLP model;
FIG. 9 is a diagram showing inversion effects of water content of combustible substances in the canopy;
FIG. 10 is a graph showing the inversion effect of water content of surface combustibles.
Detailed Description
The invention is further described with reference to specific examples.
Example 1
The inversion structure of the algorithm is shown in fig. 1.
Firstly, remote sensing data of a target area is obtained through a Sentinel second satellite, the target area of the selected area is a chongli area of Zhangzhou city in Hebei province, remote sensing data of a research area is generated on a track R075 by a satellite Sentinel-2B at 8, 25, 3 and 8 months in 2020 within 5 minutes and 49 seconds, and as shown in figure 1, the data level is L1C data which is subjected to geometric correction, radiometric calibration and calculation of atmospheric expression reflectivity. The preprocessing operations of the data include atmospheric correction, resampling and cropping. After respectively carrying out atmospheric correction on a 10m resolution wave Band and a 20m resolution wave Band by a Sen2Cor plug-in unit Of an SNAP (sensor Application platform), four groups Of L2A level data (two groups Of 10m resolution and 20m resolution) can be obtained, then resampling is carried out on the data to obtain 10m resolution, a resampling method is a nearest neighbor method, then a true color image (figure 1) is generated by wave Band synthesis (R: G: B: Band4: Band3: Band2) and then is cut, finally a remote sensing image in a research area range is obtained, and a plurality Of sample points (ROI) are randomly selected in a vegetation coverage area to carry out field sampling.
The water content of the combustible materials of the vegetation canopy and the earth surface litter needs to be obtained and researched, 200 sample points are uniformly and widely distributed and screened in a remote sensing image cut in a research area and then marked on a vector boundary diagram of a Chongli area, as shown in FIG. 2, and the size of each sample point is set to be 45m multiplied by 45 m. After the sample point is located on the spot, canopy vegetation and litter on the earth surface of the sample point are collected by a direct acquisition method from 7/1/2021 to 9/1/2021 in the front of a target area, and information such as longitude and latitude, tree species, temperature and humidity, atmospheric pressure and the like of the sample point is measured and recorded.
And after all samples are taken back, weighing the samples and recording the samples as the wet weight of the combustible, then putting the samples into a baking oven for continuous constant-temperature drying, measuring the dried weight and recording the weight as the wet weight of the combustible after the weight is constant, and finally calculating the water content of all samples according to a formula 1.
Absolute Moisture Content (AMC)
Figure BDA0003394321490000041
Relative Moisture Content (RMC)
Figure BDA0003394321490000042
Wherein W _ H is wet weight (g) of combustible, and W _ D is dry weight (g) of combustible.
The near infrared band is positioned in a high reflection area of the plant and is also positioned in a strong absorption area of the water body, the short wave infrared band is positioned between the absorption bands of the water body, and the water content of the combustible has obvious correlation with the spectral reflectivities of the two bands. The spectral moisture index method is mainly characterized in that a spectral index is calculated according to spectral reflectivity and is compared with measured data so as to calculate the moisture content of the combustible materials in the canopy, and for the moisture content of the ground surface litter, the remote sensing data is required to be processed according to a radiation transmission model. Firstly, preliminarily obtaining ground remote sensing through satellite remote sensing data, and inverting the water content of the combustible through the spectral reflectivity and the spectral water content index.
Meanwhile, due to the existence of the canopy shielding problem, the canopy shielding problem is solved through a radiation transmission model, and the correlation between the moisture content of the combustible and the spectral reflectivity is analyzed through an MLP-based model.
After the remote sensing data and the data collected on the spot are collected, the data are input into a computer for deep learning of an inversion model, and an MLP deep learning model is adopted in the invention.
The MLP network structure comprises an input layer, a hidden layer and an output layer, is a common model in deep learning, learns the characteristics of input data by constructing a multilayer neural network, has strong adaptivity, and is widely applied to the research of regression prediction at present.
Based on the model, preprocessing such as atmospheric correction and resampling are carried out on the acquired data at the L1C level, the data are processed into remote sensing data at the L2A level, and then a target research area is cut out to serve as a sampling point of model inversion. Then, correlation analysis is carried out on the basis of all wave bands provided by remote sensing data, and 5 characteristic variables of red light (B3), green light (B4), near infrared (B8) and two short wave infrared (B11 and B12) wave bands are selected as multi-independent variable input of the MLP model. And selecting the most appropriate MLP deep learning model according to the spectral characteristics of the water content. Model structure as shown in fig. 4, two fully connected layers each contain 64 nodes, and use relu (rectified Linear unit) as an activation function, and the output layer uses a Linear function as an activation function.
Based on an MLP deep learning model, inverting the water content of the combustible substances in the canopy, selecting 5 wave band reflectivities counted by B3, B4, B8, B11 and B12 as input, selecting 70% of data (140 data counted by B) in samples as training samples, and performing iterative training 1000 times by using Mean Squared Error (MSE) as a loss function. A comparison graph of training errors and actual errors in the training process is shown in FIG. 5, and the errors of both the training errors and the actual errors can be controlled within 1, so that the model training effect is good. After training is finished, 30% of data (60 in total) in the samples are selected as test samples, prediction is carried out by using a model and is compared with an actual value, a line drawing graph 6 is drawn, the actual fitting degree (R ^2) is calculated, and the calculation result is 0.843.
Inverting the moisture content of the combustible of the ground surface withered besides the combustible of the open canopy, and predicting the moisture content of the ground surface withered by adopting a two-way reflection distribution function, wherein the two-way reflection distribution function is defined as the illumination intensity of the radiation reflected along the reflection direction (namely the observation direction)
Figure BDA0003394321490000051
With the intensity of radiation at the surface of the observation target
Figure BDA0003394321490000052
The function formula of the ratio is as follows:
Figure BDA0003394321490000053
wherein λ is wavelength (nm), θiIs the angle between the incident direction of the sunlight and the zenith angle, thetarIs the angle between the observation direction and the zenith angle,
Figure BDA0003394321490000054
and
Figure BDA0003394321490000055
respectively, refer to the angle of the incident direction and the observation direction in azimuth.
The reflectivity relation of the radiation transmission model 4-scale model is as follows:
R＝RTKT+RGKG+RZTKZT+RZGKZG(4)
wherein: rTRepresenting the reflectivity of the canopy illuminated surface;
KTrepresenting the probability of the sensor observing the ground illuminated surface
RGRepresenting the surface illumination surface reflectivity;
KGrepresenting the probability of the sensor observing the ground illuminated surface
RZTRepresenting the reflectivity of the background surface of the canopy;
KZTrepresenting the probability of the sensor observing the background surface of the canopy
RZGRepresenting the reflectivity of the ground background surface;
KZGrepresenting the probability of the sensor observing the ground background surface.
Spectral reflectivity R of canopy with observation angles of alpha and betaαAnd RβThe relationship is as follows:
Figure BDA0003394321490000061
in the formula: m is a multiple scattering factor, and A, B and C are relations between M and K components.
The problem of canopy sheltering exists in utilizing remote sensing data to carry out the moisture content inversion to earth's surface combustible substance, and the spectral reflectance in vegetation region is decided by factors such as blade, soil jointly, and it is not a plane rigid body, and radiation can pass the vegetation canopy and pass through the multiple scattering effect again, escapes from the upper strata of vegetation at last, is received by the remote sensing. The remote sensing data is obtained as a two-dimensional planar model, while the vegetation area is a three-dimensional model. Therefore, in order to obtain the surface reflectivity, the BRDF is used for processing the original remote sensing data to obtain the multi-angle remote sensing data for the remote sensing data through ENVI, then the multi-angle remote sensing data is substituted into a 4-scale model in a radiation transmission model, a multiple scattering factor M and an observation probability K are obtained based on the actually measured data and a formula 4, and then the surface remote sensing data is obtained according to a formula 5.
Based on remote sensing data and surface litter water content inversion of an MLP deep learning model, 5 wave band reflectivities counted by B3, B4, B8, B11 and B12 are selected as input, 70% of data (140 data counted in total) in a sample are selected as training samples, and the mean square error is used as a loss function for 1000 times of iterative training. A comparison graph of the training error and the actual error in the training process is shown in FIG. 7, the training error in the training process is continuously reduced, iteration is not performed after the actual error tends to be stable, and the overfitting phenomenon is prevented. The final value of the actual average absolute error was 7.69. After training is completed, 30% of data (60 in total) in the samples are selected as test samples, prediction is carried out by using a model and is compared with an actual value, a line graph 8 is drawn, and an actual fitting degree (R) is calculated2) The calculation result was 0.448.
The time of the remote sensing data image selected by the embodiment is consistent with the time period from 8 months and 25 days in 2020 to 7 months and 1 days in 2021 to 9 months and 1 days in 2021, and the vegetation condition has no obvious change.
FIG. 9 is a gray scale diagram and a distribution diagram of water content of a Chongli area inverted by taking spectral reflectance as an input variable according to an MLP model, wherein the main tree species of the west of the Chongli area are shrubs, the main tree species of the east of the Chongli area are trees, the water absorption processes of canopy blades of the shrubs and the trees are similar, the interception of part of the canopy of the shrubs is higher than that of the canopy of the trees, and therefore the water content of canopy vegetation of the Chongli area is distributed in a mode that the west of the Chongli area is higher than that of the east of the trees.
FIG. 10 is a gray scale diagram and a distribution diagram of water content of a Chongli area inverted by taking spectral reflectance as an input variable according to an MLP model, wherein the main tree species in the southeast of the Chongli area are trees, the Chongli area is in midsummer during field investigation, the withered and fallen objects in the trees in summer have obvious surface runoff interception and the effect of inhibiting soil water evaporation, and the water content is higher at the moment, so that the water content of the withered and fallen objects in the area is higher than that in the west.
The average water content of the canopy vegetation in the whole Chongli area is 35 percent, and the average water content of the ground surface litter is 52 percent. The inversion accuracy of the water content of the combustible canopy is high, the fitting degree is 0.843, and although the canopy shielding problem exists in the earth surface litter, the fitting degree is 0.448 by the inversion model with high accuracy obtained after the remote sensing data is processed by the radiation transmission model.

Claims (5)

1. A forest combustible water content inversion algorithm based on remote sensing data is characterized by comprising the following steps:
step 1: extracting remote sensing data, preprocessing the remote sensing data, and cutting and screening sample points;
step 2: proceeding to the sample points marked in the step 1 for sampling on the spot;
and step 3: performing inversion of the water content of the combustible materials in the canopy by adopting an MLP deep learning model;
and4, step 4: and performing surface combustible water content inversion by adopting an MLP deep learning model.
2. The forest combustible water content inversion algorithm based on remote sensing data as claimed in claim 1, wherein the step of preprocessing the remote sensing data is as follows:
step 1: sequentially carrying out atmospheric correction on the 10m resolution wave band and the 20m resolution wave band to respectively obtain two groups of L2A-level data of 10m resolution and two groups of 20m resolution wave bands;
step 2: resampling the data obtained in the step 1 to be a 10m resolution wave band by using a nearest neighbor method;
and step 3: and (3) performing band synthesis on the data obtained in the step (2) to generate a true color image.
3. The forest combustible water content inversion algorithm based on remote sensing data as claimed in claim 1, wherein the field sampling in step 2 is to collect canopy vegetation and litter on the earth surface at a sample point, measure and record information such as longitude and latitude, tree species, temperature and humidity and atmospheric pressure of the sample point, and calculate water content of all samples according to the following formula:
absolute water content
Figure FDA0003394321480000011
Relative water content
Figure FDA0003394321480000012
Wherein, WHIs the wet weight (g) of combustible materials, WDIs combustible dry weight (g).
4. The forest combustible water content inversion algorithm based on remote sensing data as claimed in claim 1, wherein the canopy water content inversion is that red light, green light, near infrared and two short wave infrared in original data are selected as input ends by using an MLP deep learning model, inversion of canopy combustible water content is directly carried out, machine learning is carried out for multiple times with data obtained by actual sampling, and the model is optimized.
5. The forest combustible water content inversion algorithm based on remote sensing data of claim 1, wherein the inversion of the surface combustible water content is that the original data is processed by using a two-way reflection distribution function to obtain multi-angle remote sensing data, and the expression is as follows:
Figure FDA0003394321480000013
where λ is the wavelength, θiIs the angle between the incident direction of the sunlight and the zenith angle, thetarIs the angle between the observation direction and the zenith angle,
Figure FDA0003394321480000021
and
Figure FDA0003394321480000022
the angles of the incident direction and the observation direction in the direction are respectively indicated;
and substituting the obtained data into a 4-scale model in the radiation transmission model, wherein the reflectivity relation of the 4-scale model is as follows:
R＝RTKT+RGKG+RZTKZT+RZGKZG
wherein: rTRepresenting the reflectivity of the canopy illuminated surface;
KTrepresenting the probability of the sensor observing the ground illuminated surface
RGRepresenting the surface illumination surface reflectivity;
KGrepresenting the probability of the sensor observing the ground illuminated surface
RZTRepresenting the reflectivity of the background surface of the canopy;
KZTrepresenting the probability of the sensor observing the background surface of the canopy
RZGRepresenting the reflectivity of the ground background surface;
KZGrepresenting the probability of the sensor observing the ground background surface;
then obtaining surface remote sensing data by using the following formula;
Figure FDA0003394321480000023
in the formula: m is a multiple scattering factor, and A, B and C are relational expressions of M and K components;
and finally, selecting red light, green light, near infrared and two short wave infrared in the obtained surface remote sensing data as input ends in an MLP deep learning model, carrying out inversion on the water content of the surface combustible, and carrying out machine learning with the data obtained by actual sampling for multiple times to optimize the model.
CN202111478086.5A 2021-12-06 2021-12-06 Forest combustible water content inversion algorithm based on remote sensing data Pending CN114492726A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111478086.5A CN114492726A (en) 2021-12-06 2021-12-06 Forest combustible water content inversion algorithm based on remote sensing data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111478086.5A CN114492726A (en) 2021-12-06 2021-12-06 Forest combustible water content inversion algorithm based on remote sensing data

Publications (1)

Publication Number Publication Date
CN114492726A true CN114492726A (en) 2022-05-13

Family

ID=81492058

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111478086.5A Pending CN114492726A (en) 2021-12-06 2021-12-06 Forest combustible water content inversion algorithm based on remote sensing data

Country Status (1)

Country Link
CN (1) CN114492726A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117949342A (en) * 2024-03-27 2024-04-30 四川省林业和草原调查规划院(四川省林业和草原生态环境监测中心) On-line measuring device for moisture content of under-forest withered matters
CN118428413A (en) * 2024-07-02 2024-08-02 南京信息工程大学 Deep learning model for estimating surface water content and application
CN118501103A (en) * 2024-05-30 2024-08-16 奥谱天成(成都)信息科技有限公司 Adaptive plant leaf moisture content detection method and detection model building method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117949342A (en) * 2024-03-27 2024-04-30 四川省林业和草原调查规划院(四川省林业和草原生态环境监测中心) On-line measuring device for moisture content of under-forest withered matters
CN118501103A (en) * 2024-05-30 2024-08-16 奥谱天成(成都)信息科技有限公司 Adaptive plant leaf moisture content detection method and detection model building method
CN118501103B (en) * 2024-05-30 2024-09-20 奥谱天成(成都)信息科技有限公司 Adaptive plant leaf moisture content detection method and detection model building method
CN118428413A (en) * 2024-07-02 2024-08-02 南京信息工程大学 Deep learning model for estimating surface water content and application

Similar Documents

Publication Publication Date Title
CN114492726A (en) Forest combustible water content inversion algorithm based on remote sensing data
Le Maire et al. Calibration and validation of hyperspectral indices for the estimation of broadleaved forest leaf chlorophyll content, leaf mass per area, leaf area index and leaf canopy biomass
CN107796764B (en) Method for constructing wheat leaf area index estimation model based on three-band vegetation index
CN109977548B (en) Leaf area index remote sensing inversion method for different ranges
Miller et al. The ARM Mobile Facility and its first international deployment: Measuring radiative flux divergence in West Africa
Hou et al. An algorithm for hyperspectral remote sensing of aerosols: 2. Information content analysis for aerosol parameters and principal components of surface spectra
CN112213287A (en) Coastal beach salinity inversion method based on remote sensing satellite image
CN111044516B (en) Remote sensing estimation method for chlorophyll content of rice
Rautiainen et al. Coupling forest canopy and understory reflectance in the Arctic latitudes of Finland
CN111855590A (en) Remote sensing inversion model and method for rice leaf starch accumulation
Bayat et al. Retrieval of land surface properties from an annual time series of Landsat TOA radiances during a drought episode using coupled radiative transfer models
CN115526098A (en) Remote sensing calculation method for leaf area index of surface vegetation in mining area and electronic equipment
Hang et al. Estimation of chlorophyll-a concentration in Lake Taihu from Gaofen-1 wide-field-of-view data through a machine learning trained algorithm
Alexandrov et al. Remote sensing of atmospheric aerosols and trace gases by means of multifilter rotating shadowband radiometer. Part II: Climatological applications
Karakoc et al. Ratio-based vegetation indices for biomass estimation depending on grassland characteristics
CN111595806A (en) Method for monitoring soil carbon component by using mid-infrared diffuse reflection spectrum
CN114755189B (en) Feature-optimized self-attention-based hyperspectral satellite LAI inversion method
CN115775357A (en) Hyperspectrum and machine learning based vegetation equivalent water thickness estimation method
CN115876721A (en) Crop classification method, system, medium, computer equipment and terminal
Wang et al. Comparing broad-band and red edge-based spectral vegetation indices to estimate nitrogen concentration of crops using casi data
Shi et al. Water properties in the La Plata River Estuary from VIIRS observations
Jin et al. Rmote sensing of leaf water content for winter wheat using grey relational analysis (GRA), stepwise regression method (SRM) and partial least squares (PLS)
CN114397276B (en) Regional soil humidity monitoring method based on equivalent precipitation estimation method
Xu et al. Nondestructive detection of yellow peach quality parameters based on 3D-CNN and hyperspectral images
CN118429818B (en) Water chlorophyll a concentration inversion method based on Sentinel-2 satellite image reconstruction

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