CN108152235A - The content of beary metal inversion method of external spectrum in a kind of joint soil chamber - Google Patents
The content of beary metal inversion method of external spectrum in a kind of joint soil chamber Download PDFInfo
- Publication number
- CN108152235A CN108152235A CN201810233070.XA CN201810233070A CN108152235A CN 108152235 A CN108152235 A CN 108152235A CN 201810233070 A CN201810233070 A CN 201810233070A CN 108152235 A CN108152235 A CN 108152235A
- Authority
- CN
- China
- Prior art keywords
- soil
- content
- spectrum
- sample
- spectral
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/3103—Atomic absorption analysis
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of content of beary metal inversion methods for combining external spectrum in soil chamber, first, spectra pretreatment based on soil indoor spectral, builds the total factor principal component Gradual regression analysis model of heavy metal content in soil inverting, and extract characteristic wave bands principal component;Then, it is built in soil chamber using the suitable Transform Sets of Kennard Stone algorithms selections, the association transformation model of outdoor spectrum, and spectra pretreatment equally is done to transformed soil chamber external spectrum;Finally, the characteristic wave bands principal component of spectra inversion model extraction in the soil chamber of content of beary metal is merged, the regression model of the content of beary metal EO-1 hyperion inverting of the soil chamber external spectrum after building based on conversion process.This is that the inverting of current soil content of beary metal high-spectrum remote-sensing is confined to sampling soil indoor spectral properties study more, it is difficult to effectively directly apply under conditions of a wide range of heavy metal pollution of soil investigation in field, realizes that field directly utilizes a wide range of efficient heavy metal content in soil inverting of soil chamber external spectrum.
Description
Technical field
The present invention relates to environmental monitoring, it is anti-to carry out content of beary metal for external spectrum in more particularly to a kind of joint soil chamber
The method drilled.
Background technology
When obtaining a wide range of high-density soils heavy metal pollution information using conventional sample mode on the spot, it often is faced with into
Huge, the less efficient problem of this consuming, and it is difficult to ensure that rapidly and efficiently dynamic monitoring is carried out for normalization, can not meet state
Family rapidly and efficiently investigates thoroughly heavy metal pollution spatial framework in wide geographic space, finds heavy metal pollution temporal and spatial evolution and shape
Into mechanism, control pollution range expands and transfer, agricultural production of making rational planning for, reduction polluted product damage the great of national health
Demand.
At present, the heavy metal content in soil inversion method based on spectral analysis technique is rapidly progressed, and is had become
The main means of heavy metal pollution of soil investigation.Its cardinal principle is:Utilize spectrum in field spectroradiometer observation soil sample room
Data explore the response relation between heavy metal content in soil and soil all band spectrum (350-2500nm), a structure soil huge sum of money
Belong to content EO-1 hyperion inverse model.Traditionally statistical analysis technique has the advantage that cost is less, efficiency is higher to this method relatively,
But due to soil spectrum difference caused by the environmental conditions difference such as indoor and outdoor air so that indoor spectral is difficult to effectively directly
It, can be in the wild directly using outside soil chamber there is an urgent need for invention one kind applied to the condition of a wide range of heavy metal pollution of soil investigation in field
The method of spectrum efficient inverting heavy metal content in soil on a large scale.
Invention content
In order to effectively solve traditional limitation based on spectra inversion content of beary metal in soil chamber, to realize field soil
Directly using the purpose of soil chamber external spectrum efficient inverting heavy metal content in soil on a large scale during Investigation of Heavy Metals,
Invention provides a kind of content of beary metal inversion method for combining external spectrum in soil chamber.
In order to realize above-mentioned technical purpose, the technical scheme is that,
The content of beary metal inversion method of external spectrum, includes the following steps in a kind of joint soil chamber:
Step 1:Soil sample is acquired, and in room needing the contaminated soil enumeration district for carrying out heavy metal content in soil detection
Then the outdoor spectroscopic data of outer acquisition soil sample send soil sample to progress content of beary metal detection in laboratory, and
The indoor spectral data of soil sample are acquired indoors, are finally based on sampled point heavy metal content in soil measured data and soil chamber
Interior spectroscopic data builds the total factor principal component Gradual regression analysis model of the heavy metal content in soil inverting based on indoor spectral, and
Extract characteristic wave bands principal component;
Step 2:It chooses in suitable conversion sample set structure soil chamber, the association transformation model of outdoor spectrum, to whole
Spectroscopic data does spectrum conversion and spectra pretreatment outside sampling soil room;
Step 3:To the characteristic wave bands principal component extracted in step 1 and conversion in step 2 treated field soil spectrum
Principal component does correlation analysis, extracts significantly correlated principal component as explanatory variable, is contained with the heavy metal-polluted soil measured in step 1
Amount builds the regression model of the content of beary metal inverting based on soil chamber external spectrum as dependent variable.
The content of beary metal inversion method of external spectrum, the soil weight described in step 1 in a kind of joint soil chamber
Tenor and its indoor spectral data are obtained by following steps:
Outdoor contaminated soil is ground using grid in the contaminated soil enumeration district for needing to carry out heavy metal content in soil detection
Study carefully target area progress grid to define, and the point of intersection setting soil sampling point acquisition formed between grid is no less than 40 soil-likes
This;Collected soil sample is digested using three acid in laboratory --- atomic absorption spectrophotometry chemical analysis measures
Pedotheque content of beary metal, meanwhile, measure spectroscopic data in soil chamber using field spectroradiometer;
Sampled point soil chamber external spectrum data described in step 1 are obtained by following steps:
While step 1 carries out field soil sample collection, the outdoor of acquisition soil sample is monitored with field spectroradiometer
Spectroscopic data.
The content of beary metal inversion method of external spectrum, the structure base described in step 1 in a kind of joint soil chamber
In the content of beary metal inverting of soil indoor spectral total factor principal component Gradual regression analysis model and extract characteristic wave bands principal component,
Include the following steps:
Step 1):It is converted using spectral resampling method, standard normal, the preprocess method of single order/second-order differential is to soil chamber
Interior spectroscopic data is pre-processed, and builds the total factor principal component successive Regression mould of the content of beary metal based on soil indoor spectral
Type;
Step 2):The characteristic wave bands Principal component of step 1) model is extracted, the soil outdoor optical as content of beary metal
Compose the judgement factor of regression modeling independent variable screening.
The content of beary metal inversion method of external spectrum in a kind of joint soil chamber, in step 1), in soil chamber
The pretreatment that spectroscopic data carries out spectral resampling method includes the following steps:
Using preset length as spacer units, spectrum is carried out to spectroscopic data in the soil chamber of measure using the following formula and is adopted again
Sample:
R=[r (λ1)+···+r(λi)+···+r(λn)]/n
Wherein, r represents reflectance value of the bloom spectral curve after resampling;N is by including in the range of spacer units
Spectral band number, r (λi) represent i-th of spectral band λiReflectivity;
In step 1), the pretreatment that standard normal transformation is carried out to spectroscopic data in soil chamber includes the following steps:
Based on the indoor spectral after resampling, standard normal variable is carried out using the following formula based on weighted average method
It converts to complete normalized:
Wherein, n is by the spectral band number that is included in the range of spacer units;Represent that n band spectrum of i-th of sample is anti-
Penetrate the average value of rate;xi,jRepresent the spectral reflectance values of j-th of wave band of i-th of sample;δ is customized offset;wiTable
Show the standard deviation of i-th of sample;x′i,jRepresent that normalized is completed in j-th of wave band standard normal transformation of i-th of sample
End value;
In step 1), the pretreatment that single order/second-order differential is carried out to spectroscopic data in soil chamber includes the following steps:
The following formula is used to calculate each sample spectroscopic data based on single order/second-order differential to remove interference:
r′(λi)=[r (λi)-r(λi-1)]/2Δλ
r″(λi)=[r (λi)-2r(λi-1)+r(λi-2)]/(2Δλ)2
Wherein, r (λi) it is i-th of spectral band λi, r ' (λi) and r " (λi) it is respectively spectral band λiSingle order/second order it is micro-
Spectrum is divided, Δ λ is spectral band λi-1With λiBetween spacer units.
The content of beary metal inversion method of external spectrum in a kind of joint soil chamber, in step 1), structure is based on room
The total factor principal component Gradual regression analysis model of the heavy metal content in soil of interior spectrum includes the following steps:
Using the pedotheque weight, as input independent variable, measured in step 1 by spectroscopic data in pretreated soil chamber
Tenor is dependent variable, builds total factor principal component Gradual regression analysis model:
Cj=v1jX1+v2jX2+…vijXi…+vmjXm
Y=a0+a1C1+a2C2+…anCn
Wherein, CjRepresent j-th of new component after principal component transform, XiRepresent the spectral reflectivity of i-th of wave band;M is represented
Spectral variables number before principal component transform, vijRepresent the corresponding spy of ith feature value in the correlation matrix of m original variable
Sign vector, Y represent heavy metal content in soil estimated value, anRepresent the corresponding content of beary metal inverting mould of n-th of main variables
Type coefficient.
The content of beary metal inversion method of external spectrum in a kind of joint soil chamber, the step 2 include following
Step:
Step is 1.:Deleting indoor and outdoor soil spectrum steam influences wave band, obtains effective soil spectrum data;
Step is 2.:Pass through the Euclidean distance between calculating sampling soil indoor spectral, choosing using Kennard-Stone algorithms
Select Transform Sets sample:
Wherein, rikWith rjkThe spectral reflectance values of k-th of wave band of representative sample i and sample j respectively;dijRepresent sampling
Euclidean distance between soil sample i and sample j, p are the wavelength number of sample spectra;
Step is 3.:Using direct correcting algorithm construction step 2) choose Transform Sets sample indoor and outdoor spectrum conversion
Model:
Wherein, E is residual matrix, XQFor spectral value in analysis sample room, XqTo analyze sample room's external spectrum value,For
Spectral value after centralization processing in analysis sample room,For spectral value after centralization processing outside analysis sample room;
Step is 4.:Spectrum outside pedotheque room is carried out light by external spectrum transformation model in the soil chamber 3. built using step
Spectrum conversion, obtains transformed soil chamber external spectrum data;
Step is 5.:Using spectral resampling method-standard normal transformation-single order/second-order differential spectra preprocess method
To step, 4. transformed soil chamber external spectrum carries out spectra data prediction.
The content of beary metal inversion method of external spectrum, step 3 include the following steps in a kind of joint soil chamber:
Step 1:The soil chamber external spectrum principal component after conversion process is extracted using principal component analytical method;
Step 2:The spectrum number of principal components evidence and heavy metal of step 1 extraction are calculated using Pearson correlation analysis method
The related coefficient of the principal component judgement factor that spectra inversion model is selected into content soil chamber;
Step 3:Two related coefficient result of calculation of analytical procedure chooses the spectrum master with judgement factor related coefficient maximum
Compositional data as content of beary metal inverse model input independent variable, using the pedotheque content of beary metal that step 1 measures as because
Variable builds linear regression model (LRM).
The technical effects of the invention are that based in soil chamber, in outdoor spectrum association consistency, take geographical ring into account
Border element influences the variation of soil spectrum reflectance value, the present invention is based in soil chamber, the association modulus of conversion of outdoor spectrum
Type builds the content of beary metal inverting for realizing external spectrum in joint soil chamber.Specifically, the present invention proposes, based in soil chamber
The spectra pretreatment of spectrum, builds the total factor principal component Gradual regression analysis model of heavy metal content in soil inverting, and extract
Characteristic wave bands principal component;It is built in soil chamber using the suitable Transform Sets of Kennard-Stone algorithms selections, the pass of outdoor spectrum
Join transformation model, and spectra pretreatment equally is done to transformed soil chamber external spectrum;Finally, content of beary metal is merged
The characteristic wave bands principal component of spectra inversion model extraction in soil chamber, based on the soil chamber external spectrum data after conversion process, structure
Build content of beary metal EO-1 hyperion inverse model.This is that the inverting of current soil content of beary metal high-spectrum remote-sensing is confined to sampling soil more
Earth indoor spectral properties study, it is difficult to it effectively directly applies under conditions of a wide range of heavy metal pollution of soil investigation in field, it is real
Existing field directly utilizes a wide range of efficient heavy metal content in soil inverting of soil chamber external spectrum.
Description of the drawings
Fig. 1 is the indoor external spectrum joint conversion key technology schematic diagram of the present invention;
In Fig. 2, (a) is the content of beary metal total factor principal component analysis model result example based on soil indoor spectral
Figure, (b) are the characteristic wave bands Principle component extraction result exemplary plot based on model result.
In Fig. 3, (a) is the heavy metal content in soil inverse model result exemplary plot of the indoor external spectrum of joint, (b) be based on
The soil spectrum response characteristic wave band recognition result exemplary plot of the content of beary metal of model result.
Specific embodiment
Here is to a preferred embodiment of the invention, with reference to the detailed description of attached drawing progress.
1st, it builds the total factor principal component Gradual regression analysis model of the content of beary metal inverting based on soil indoor spectral and carries
Take characteristic wave bands principal component, the step of heavy metal content in soil indoor spectral inverting characteristic wave bands Principle component extraction that the present invention uses
Suddenly include:
Step 1):It is sampled in soil investigation area using grid acquisition mode, with 1m × 1m resolution ratio to contaminated soil
Research target area carries out grid and defines i.e. gridding, and the point of intersection setting sampled point formed between grid acquires soil sample;
Using ASD field spectroradiometers acquisition soil sampling point outdoor spectroscopic data (350-2500nm);Then by the soil sample of acquisition
Bring into laboratory, above-mentioned sampled point pedotheque acquired using nine point sampling methods, digested using three acid --- Atomic absorption (stone
Black stove) spectrophotometry laboratory chemical analysis measure pedotheque content of beary metal, equally measured with ASD field spectroradiometers
Soil sample indoor spectral data (350-2500nm);Same soil sample has thus been obtained in outdoor-monitoring and indoor monitoring
The spectroscopic data obtained under the conditions of varying environment, that is, the two groups of spectrum used are converted below.
Step 2):Spectroscopic data carries out spectral resampling method in the soil chamber measured using 10nm as spacer units to step 1);
Specific calculation formula is as follows:
R=[r (λ1)+…+r(λi)+…+r(λn)]/n
In formula, r represents reflectance value of the bloom spectral curve after resampling;N in the range of 10nm spacer units by wrapping
The spectral band number contained, r (λi) represent i-th of spectral band λiReflectivity;
Step 3):Based on the curve of spectrum after resampling, standard normal is specifically carried out to it using weighted average method
Change of variable, to complete the normalized of the collecting sample curve of spectrum.Its specific calculation formula is as follows:
Wherein, n is the quantity of spectral band;Represent the average value of n band spectrum reflectivity of i-th of sample;xi,jTable
Show the spectral reflectance values of j-th of wave band of i-th of sample;δ is customized offset;wiRepresent the standard of i-th of sample
Difference;x′i,jRepresent that the end value of normalized is completed in j-th of wave band standard normal transformation of i-th of sample;
Step 4):Each sample spectroscopic data is calculated using single order/second-order differential processing, it is linear to remove partial linear or class
Spectral background value to the interference of heavy metal spectrum (such as:Noise spectrum), reach enhanced spectrum feature difference, extraction spectral signature is inhaled
The purpose of take-up.Specific formula for calculation is as follows:
r′(λi)=[r (λi)-r(λi-1)]/2Δλ
r″(λi)=[r (λi)-2r(λi-1)+r(λi-2)]/(2Δλ)2
Wherein, r (λi) it is i-th of spectral band λi, r ' (λi) and r " (λi) it is respectively spectral band λiSingle order/second order it is micro-
Spectrum is divided, Δ λ is spectral band λi-1With λiBetween spacer units 10nm;
Step 5):Using spectroscopic data in above-mentioned pretreated soil chamber as input independent variable, the soil of measure in step 1)
Earth sample content of beary metal is dependent variable, builds total factor principal component Gradual regression analysis model.Specific formula for calculation is as follows:
Cj=v1jX1+v2jX2+…vijXi…+vmjXm
Y=a0+a1C1+a2C2+…anCn
In formula, CjRepresent j-th of new component after principal component transform, XiRepresent the spectral reflectivity of i-th of wave band;M is represented
Spectral variables number before principal component transform, vijRepresent the corresponding spy of ith feature value in the correlation matrix of m original variable
Sign vector, Y represent heavy metal content in soil estimated value, anRepresent the corresponding content of beary metal inverting mould of n-th of main variables
Type coefficient;
Step 6):The model independent variable that will be established in step 5) with total factor principal component stepwise regression method, that is, be based on
The characteristic wave bands Principal component extraction of the content of beary metal inverse model of soil indoor spectral, the soil chamber as content of beary metal
The judgement factor of external spectrum regression modeling independent variable screening;
2nd, the association transformation model structure of external spectrum and the conversion process of soil chamber external spectrum, the present invention use in soil chamber
Soil chamber in, outdoor spectrum transformation model structure and to the conversion of soil chamber external spectrum and spectra pre-treatment step packet
It includes:
Step 1):Removing indoor and outdoor soil spectrum steam influences wave band (1360-1490nm and 1810-1960nm), obtains
Obtain effective soil spectrum data;
Step 2):Pass through the Euclidean distance between calculating sampling soil indoor spectral, choosing using Kennard-Stone algorithms
Select Transform Sets sample.Euclidean distance specific formula for calculation is as follows:
Wherein, rikWith rjkThe spectral reflectance values of k-th of wave band of representative sample i and sample j respectively;dijRepresent sampling
Euclidean distance between soil sample i and sample j, p are the wavelength number of sample spectra;
Kennard-Stone algorithms implement step:
(1) first, two samples of distance maximum between sample two-by-two are selected as first and second Transform Sets sample
Product;
(2) then, remaining sample and the distance between sampling sheet are calculated respectively;
(3) for each remaining sample, the shortest distance between sampling product is chosen, and then selects these
Sample in the shortest distance corresponding to relatively longest distance, as third Transform Sets sample;
(4) step (3) is repeated until the number of selected Transform Sets sample is equal to pre-determined number;
Step 3):Using direct correcting algorithm construction step 2) choose Transform Sets sample indoor and outdoor spectrum conversion
Model, concrete principle are as follows:
(1) difference of the same conversion indoor and outdoor spectroscopic data of pedotheque represents the soil under two kinds of measuring environments
The difference of spectral signature, there are following conversion relational expressions between the two:
XQ=XqA+E
Wherein, XQFor spectral value in analysis sample room, XqFor analysis sample room external spectrum value, A is converted for indoor and outdoor spectrum
Matrix, E are residual matrix;
(2) when spectral value X in sample room will be analyzedIWith outdoor spectral value XiAfter carrying out centralization processing respectively, according to upper
The transfer equation stated has following relational expression:
Wherein,Spectral value after being handled for centralization in analysis sample room,For centralization processing outside analysis sample room
Spectral value afterwards, then the calculated value of spectrum transition matrix A be:
(3) residual matrix can be obtained by substituting the above to:
Interior, the outdoor spectrum transformation model structure of soil chamber is completed as a result,;
Step 4):Spectrum outside pedotheque room is carried out by light using external spectrum transformation model in the soil chamber of step 3) structure
Spectrum conversion, obtains transformed soil chamber external spectrum data;
Step 5):Using spectral resampling method-standard normal transformation-single order/second-order differential spectra preprocess method
To step 4), transformed soil chamber external spectrum carries out spectra data prediction;
3rd, combine the content of beary metal inverting regression model structure of external spectrum in soil chamber, the fusion huge sum of money that the present invention uses
Belong to the characteristic wave bands principal component of spectra inversion model extraction in the soil chamber of content, outside the soil chamber after building based on conversion process
The step of regression model of the content of beary metal EO-1 hyperion inverting of spectrum, includes:
Step 1):The soil chamber external spectrum principal component after conversion process is extracted using principal component analytical method.It is specific to calculate
Formula is as follows:
Dj=q1jX1+q2jX2+…qijXi…+qmjXm
In formula, i-th of wave band reflectivity XiIt represents;Ingredient D after transformationjIt represents, coefficient qmjIt is that m variable is related
The corresponding feature vector of j-th of characteristic value of coefficient square;
Step 2):The spectrum number of principal components evidence and heavy metal of step 1) extraction are calculated using Pearson correlation analysis method
The related coefficient of the principal component judgement factor that soil spectrum inverse model is selected into content room.Its calculation formula is:
In formula, DjRepresent the vector that j-th of principal component of n sample is formed, CpRepresent what is built based on indoor soil spectrum
The matrix that p-th of main variables of content of beary metal inverse model extraction are formed, σ represent standard deviation, and related coefficient t was examined
0.05 horizontal significant wave band is significantly correlated principal component in journey;
Step 3):Analytical procedure 2) related coefficient result of calculation, choose the spectrum master with judgement factor related coefficient maximum
Compositional data inputs independent variable using pedotheque content of beary metal as dependent variable as content of beary metal inverse model, and structure is linear
Regression model.Specific formula for calculation is as follows:
Y=b0+b1D1+b2D2+…bnDn
In formula, Y represents heavy metal content in soil estimated value, DjRepresent j-th of new component after principal component transform, bnIt represents
The corresponding content of beary metal inverse model coefficient of n-th of main variables;
So far, the content of beary metal inverse model structure for combining external spectrum in soil chamber is completed.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, it is impossible to assert
The specific embodiment of the present invention is only limitted to this, for those of ordinary skill in the art to which the present invention belongs, is not taking off
Under the premise of from present inventive concept, several simple deduction or replace can also be made, should all be considered as belonging to the present invention by institute
Claims of submission determine scope of patent protection.
Claims (7)
1. a kind of content of beary metal inversion method for combining external spectrum in soil chamber, which is characterized in that include the following steps:
Step 1:Soil sample is acquired, and adopt in outdoor needing the contaminated soil enumeration district for carrying out heavy metal content in soil detection
Collect the outdoor spectroscopic data of soil sample, then send soil sample to progress content of beary metal detection in laboratory, and in room
The indoor spectral data of interior acquisition soil sample, are finally based on sampled point heavy metal content in soil measured data and soil room light
Modal data, builds the total factor principal component Gradual regression analysis model of the heavy metal content in soil inverting based on indoor spectral, and extracts
Characteristic wave bands principal component;
Step 2:It chooses in suitable conversion sample set structure soil chamber, the association transformation model of outdoor spectrum, to whole samplings
Soil chamber external spectrum data do spectrum conversion and spectra pretreatment;
Step 3:It is main to the characteristic wave bands principal component extracted in step 1 and conversion in step 2 treated field soil spectrum into
Divide and do correlation analysis, extract significantly correlated principal component as explanatory variable, made with the heavy metal content in soil measured in step 1
For dependent variable, the regression model of the content of beary metal inverting based on soil chamber external spectrum is built.
2. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 1, feature exist
In heavy metal content in soil and its indoor spectral data described in step 1 are obtained by following steps:
Target is studied to outdoor contaminated soil using grid needing the contaminated soil enumeration district for carrying out heavy metal content in soil detection
Area carries out grid and defines, and the point of intersection setting soil sampling point acquisition formed between grid is no less than 40 soil samples;
Collected soil sample is digested using three acid in laboratory --- atomic absorption spectrophotometry chemical analysis measures soil
Earth sample content of beary metal, meanwhile, measure spectroscopic data in soil chamber using field spectroradiometer;
Sampled point soil chamber external spectrum data described in step 1 are obtained by following steps:
While step 1 carries out field soil sample collection, the outdoor spectrum of acquisition soil sample is monitored with field spectroradiometer
Data.
3. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 1, feature exist
In the total factor principal component successive Regression mould of content of beary metal inverting of the structure based on soil indoor spectral described in step 1
Type simultaneously extracts characteristic wave bands principal component, includes the following steps:
Step 1):It is converted using spectral resampling method, standard normal, the preprocess method of single order/second-order differential is to soil room light
Modal data is pre-processed, and builds the total factor principal component Gradual regression analysis model of the content of beary metal based on soil indoor spectral;
Step 2):The characteristic wave bands Principal component of step 1) model is extracted, the soil chamber external spectrum as content of beary metal returns
Return the judgement factor of modeling independent variable screening.
4. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 3, feature exist
In in step 1), the pretreatment that spectral resampling method is carried out to spectroscopic data in soil chamber includes the following steps:
Using preset length as spacer units, spectral resampling method is carried out to spectroscopic data in the soil chamber of measure using the following formula:
R=[r (λ1)+…+r(λi)+…+r(λn)]/n
Wherein, r represents reflectance value of the bloom spectral curve after resampling;N is by the spectrum that is included in the range of spacer units
Wave band number, r (λi) represent i-th of spectral band λiReflectivity;
In step 1), the pretreatment that standard normal transformation is carried out to spectroscopic data in soil chamber includes the following steps:
Based on the indoor spectral after resampling, standard normal variable transformation is carried out using the following formula based on weighted average method
To complete normalized:
Wherein, n is by the spectral band number that is included in the range of spacer units;Represent n band spectrum reflectivity of i-th of sample
Average value;xi,jRepresent the spectral reflectance values of j-th of wave band of i-th of sample;δ is customized offset;wiRepresent the
The standard deviation of i sample;x′i,jRepresent that the result of normalized is completed in j-th of wave band standard normal transformation of i-th of sample
Value;
In step 1), the pretreatment that single order/second-order differential is carried out to spectroscopic data in soil chamber includes the following steps:
The following formula is used to calculate each sample spectroscopic data based on single order/second-order differential to remove interference:
r′(λi)=[r (λi)-r(λi-1)]/2Δλ
r″(λi)=[r (λi)-2r(λi-1)+r(λi-2)]/(2Δλ)2
Wherein, r (λi) it is i-th of spectral band λi, r ' (λi) and r " (λi) it is respectively spectral band λiSingle order/second-order differential light
Spectrum, Δ λ are spectral band λi-1With λiBetween spacer units.
5. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 3, feature exist
In, in step 1), build the heavy metal content in soil based on indoor spectral total factor principal component Gradual regression analysis model include with
Lower step:
Using the pedotheque heavy metal, as input independent variable, measured in step 1 by spectroscopic data in pretreated soil chamber
Content is dependent variable, builds total factor principal component Gradual regression analysis model:
Cj=v1jX1+v2jX2+…vijXi…+vmjXm
Y=a0+a1C1+a2C2+…anCn
Wherein, CjRepresent j-th of new component after principal component transform, XiRepresent the spectral reflectivity of i-th of wave band;M represent it is main into
Divide the spectral variables number before transformation, vijRepresent in the correlation matrix of m original variable the corresponding feature of ith feature value to
Amount, Y represent heavy metal content in soil estimated value, anRepresent the corresponding content of beary metal inverse model system of n-th of main variables
Number.
6. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 1, feature exist
In the step 2 includes the following steps:
Step is 1.:Deleting indoor and outdoor soil spectrum steam influences wave band, obtains effective soil spectrum data;
Step is 2.:Using Kennard-Stone algorithms by the Euclidean distance between calculating sampling soil indoor spectral, selection turns
Change collection sample:
Wherein, rikWith rjkThe spectral reflectance values of k-th of wave band of representative sample i and sample j respectively;dijRepresent sampling soil
Euclidean distance between sample i and sample j, p are the wavelength number of sample spectra;
Step is 3.:Using direct correcting algorithm construction step 2) choose Transform Sets sample indoor and outdoor spectrum transformation model:
Wherein, E is residual matrix, XQFor spectral value in analysis sample room, XqTo analyze sample room's external spectrum value,For analysis
Spectral value after centralization processing in sample room,For spectral value after centralization processing outside analysis sample room;
Step is 4.:Spectrum outside pedotheque room is carried out spectrum turn by external spectrum transformation model in the soil chamber 3. built using step
It changes, obtains transformed soil chamber external spectrum data;
Step is 5.:Using spectral resampling method-standard normal transformation-single order/second-order differential spectra preprocess method to step
Suddenly 4. transformed soil chamber external spectrum carries out spectra data prediction.
7. a kind of content of beary metal inversion method for combining external spectrum in soil chamber according to claim 1, feature exist
In step 3 includes the following steps:
Step 1:The soil chamber external spectrum principal component after conversion process is extracted using principal component analytical method;
Step 2:The spectrum number of principal components evidence and content of beary metal of step 1 extraction are calculated using Pearson correlation analysis method
The related coefficient of the principal component judgement factor that spectra inversion model is selected into soil chamber;
Step 3:Two related coefficient result of calculation of analytical procedure chooses the spectrum principal component with judgement factor related coefficient maximum
Data as content of beary metal inverse model input independent variable, the pedotheque content of beary metal measured using step 1 as dependent variable,
Build linear regression model (LRM).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810233070.XA CN108152235B (en) | 2018-03-21 | 2018-03-21 | Heavy metal content inversion method combining soil indoor and outdoor spectra |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810233070.XA CN108152235B (en) | 2018-03-21 | 2018-03-21 | Heavy metal content inversion method combining soil indoor and outdoor spectra |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108152235A true CN108152235A (en) | 2018-06-12 |
CN108152235B CN108152235B (en) | 2020-09-22 |
Family
ID=62456647
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810233070.XA Active CN108152235B (en) | 2018-03-21 | 2018-03-21 | Heavy metal content inversion method combining soil indoor and outdoor spectra |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108152235B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110991064A (en) * | 2019-12-11 | 2020-04-10 | 广州城建职业学院 | Soil heavy metal content inversion model generation method and system, storage medium and inversion method |
CN111323352A (en) * | 2020-04-09 | 2020-06-23 | 中南大学 | Regional PM2.5 remote sensing inversion model fusing fine particulate matter concentration data |
CN111537468A (en) * | 2020-06-02 | 2020-08-14 | 江南大学 | NIR (near infrared) detection method for total sugar in wheat starch milk in citric acid production |
CN111579500A (en) * | 2020-05-20 | 2020-08-25 | 湖南城市学院 | Heavy metal content support vector machine regression method combining wave bands and ratios of indoor and outdoor spectrums |
CN112649383A (en) * | 2020-12-31 | 2021-04-13 | 核工业北京地质研究院 | Ornithogalum tenuifolium leaf uranium content inversion prediction method based on spectral characteristics |
CN112683813A (en) * | 2020-12-31 | 2021-04-20 | 核工业北京地质研究院 | Hyperspectral extraction method for soil pollution information of hard rock type uranium mine |
CN114529838A (en) * | 2022-04-24 | 2022-05-24 | 江西农业大学 | Soil nitrogen content inversion model construction method and system based on convolutional neural network |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101915744B (en) * | 2010-07-05 | 2012-11-07 | 北京航空航天大学 | Near infrared spectrum nondestructive testing method and device for material component content |
CN103712923A (en) * | 2013-12-23 | 2014-04-09 | 浙江大学 | Method for eliminating moisture influence factor in field in-situ soil measurement spectrums |
CN106918565A (en) * | 2017-03-02 | 2017-07-04 | 中南大学 | Heavy metal-polluted soil Cd contents Inverse modeling and its spectral response characteristics wave band recognition methods based on indoor standard specimen bloom spectrum signature |
CN106932557A (en) * | 2017-04-12 | 2017-07-07 | 山东省科学院海洋仪器仪表研究所 | A kind of soil nutrient Model transfer method between different regions recommended based on many algorithms |
CN106951720A (en) * | 2017-04-12 | 2017-07-14 | 山东省科学院海洋仪器仪表研究所 | Soil nutrient model transfer method based on canonical correlation analysis and linear interpolation |
CN106990056A (en) * | 2017-04-20 | 2017-07-28 | 武汉大学 | A kind of total soil nitrogen spectrum appraising model calibration samples collection construction method |
-
2018
- 2018-03-21 CN CN201810233070.XA patent/CN108152235B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101915744B (en) * | 2010-07-05 | 2012-11-07 | 北京航空航天大学 | Near infrared spectrum nondestructive testing method and device for material component content |
CN103712923A (en) * | 2013-12-23 | 2014-04-09 | 浙江大学 | Method for eliminating moisture influence factor in field in-situ soil measurement spectrums |
CN106918565A (en) * | 2017-03-02 | 2017-07-04 | 中南大学 | Heavy metal-polluted soil Cd contents Inverse modeling and its spectral response characteristics wave band recognition methods based on indoor standard specimen bloom spectrum signature |
CN106932557A (en) * | 2017-04-12 | 2017-07-07 | 山东省科学院海洋仪器仪表研究所 | A kind of soil nutrient Model transfer method between different regions recommended based on many algorithms |
CN106951720A (en) * | 2017-04-12 | 2017-07-14 | 山东省科学院海洋仪器仪表研究所 | Soil nutrient model transfer method based on canonical correlation analysis and linear interpolation |
CN106990056A (en) * | 2017-04-20 | 2017-07-28 | 武汉大学 | A kind of total soil nitrogen spectrum appraising model calibration samples collection construction method |
Non-Patent Citations (2)
Title |
---|
姜晓璐等: "广东省东南部菜地水田砷含量空间分布", 《农业工程学报》 * |
涂宇龙等: "矿区土壤Cu含量高光谱反演建模", 《光谱学与光谱分析》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110991064A (en) * | 2019-12-11 | 2020-04-10 | 广州城建职业学院 | Soil heavy metal content inversion model generation method and system, storage medium and inversion method |
CN111323352A (en) * | 2020-04-09 | 2020-06-23 | 中南大学 | Regional PM2.5 remote sensing inversion model fusing fine particulate matter concentration data |
CN111323352B (en) * | 2020-04-09 | 2021-08-24 | 中南大学 | Regional PM2.5 remote sensing inversion model fusing fine particulate matter concentration data |
CN111579500A (en) * | 2020-05-20 | 2020-08-25 | 湖南城市学院 | Heavy metal content support vector machine regression method combining wave bands and ratios of indoor and outdoor spectrums |
CN111537468A (en) * | 2020-06-02 | 2020-08-14 | 江南大学 | NIR (near infrared) detection method for total sugar in wheat starch milk in citric acid production |
CN112649383A (en) * | 2020-12-31 | 2021-04-13 | 核工业北京地质研究院 | Ornithogalum tenuifolium leaf uranium content inversion prediction method based on spectral characteristics |
CN112683813A (en) * | 2020-12-31 | 2021-04-20 | 核工业北京地质研究院 | Hyperspectral extraction method for soil pollution information of hard rock type uranium mine |
CN114529838A (en) * | 2022-04-24 | 2022-05-24 | 江西农业大学 | Soil nitrogen content inversion model construction method and system based on convolutional neural network |
Also Published As
Publication number | Publication date |
---|---|
CN108152235B (en) | 2020-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108152235A (en) | The content of beary metal inversion method of external spectrum in a kind of joint soil chamber | |
CN107179291B (en) | Method for predicting content of heavy metal elements in soil based on terahertz spectrum and depth automatic encoder | |
CN106918565B (en) | Heavy metal-polluted soil Cd content Inverse modeling and its spectral response characteristics wave band recognition methods based on indoor standard specimen bloom spectrum signature | |
CN110174359B (en) | Aviation hyperspectral image soil heavy metal concentration assessment method based on Gaussian process regression | |
CN107421894A (en) | Based on unmanned plane EO-1 hyperion inverting heavy metal in soil pollution monitoring method | |
CN110807562A (en) | Regional bridge risk prediction method and system | |
WO2020207070A1 (en) | Method and system for evaluating shenzhen sea water quality | |
CN113866102B (en) | Soil health investigation monitoring method based on spectrum | |
CN111178169B (en) | Urban surface covering fine classification method and device based on remote sensing image | |
CN103792246A (en) | X-ray fluorescence spectrum quantitative analysis method based on LM-BP neural network | |
Yang et al. | The impact analysis of water body landscape pattern on urban heat island: a case study of Wuhan City | |
CN109635249B (en) | Water body turbidity inversion model establishing method, water body turbidity inversion model detecting method and water body turbidity inversion model detecting device | |
CN114460013B (en) | Coastal wetland vegetation overground biomass GAN model self-learning remote sensing inversion method | |
CN112561235A (en) | Ecological vulnerability evaluation method suitable for expressway road area | |
CN107844870A (en) | Heavy metal content in soil Forecasting Methodology based on Elman neural network models | |
CN103712923A (en) | Method for eliminating moisture influence factor in field in-situ soil measurement spectrums | |
CN115759524B (en) | Soil productivity grade identification method based on remote sensing image vegetation index | |
CN114813651A (en) | Remote sensing water quality inversion method combining difference learning rate and spectrum geometric characteristics | |
CN113466143A (en) | Soil nutrient inversion method, device, equipment and medium | |
CN106126879B (en) | A kind of soil near-infrared spectrum analysis prediction technique based on rarefaction representation technology | |
CN101446828A (en) | Nonlinear process quality prediction method | |
Chen et al. | Hyperspectral detection of sugar content for sugar-sweetened apples based on sample grouping and SPA feature selecting methods | |
CN114357891A (en) | Hyperspectral remote sensing quantitative inversion method for soil cadmium element content | |
Li et al. | Identifying spatial patterns of synchronization between NDVI and climatic determinants using joint recurrence plots | |
CN113887493A (en) | Black and odorous water body remote sensing image identification method based on ID3 algorithm |
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 |