CN107657264A - One kind carries out soil profile kind identification method based on KNN classification - Google Patents
One kind carries out soil profile kind identification method based on KNN classification Download PDFInfo
- Publication number
- CN107657264A CN107657264A CN201710430616.6A CN201710430616A CN107657264A CN 107657264 A CN107657264 A CN 107657264A CN 201710430616 A CN201710430616 A CN 201710430616A CN 107657264 A CN107657264 A CN 107657264A
- Authority
- CN
- China
- Prior art keywords
- soil
- spectrum
- curve
- trough
- interpolation
- 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
- 239000002689 soil Substances 0.000 title claims abstract description 216
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000001228 spectrum Methods 0.000 claims abstract description 100
- 238000012360 testing method Methods 0.000 claims abstract description 31
- 239000000284 extract Substances 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000013480 data collection Methods 0.000 claims abstract description 7
- 238000009795 derivation Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 15
- 230000003595 spectral effect Effects 0.000 claims description 10
- 238000002310 reflectometry Methods 0.000 claims description 9
- 238000004611 spectroscopical analysis Methods 0.000 claims description 8
- 230000003321 amplification Effects 0.000 claims description 2
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000010521 absorption reaction Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 229910052742 iron Inorganic materials 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000234314 Zingiber Species 0.000 description 1
- 235000006886 Zingiber officinale Nutrition 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 235000008397 ginger Nutrition 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- 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
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
-
- 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
- G01N21/47—Scattering, i.e. diffuse reflection
- G01N21/4738—Diffuse reflection, e.g. also for testing fluids, fibrous materials
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
Soil profile kind identification method, including following steps are carried out based on KNN classification the invention discloses one kind：S1, for a curve of spectrum, extract its trough feature；S2, for a curve of spectrum, extract its crest feature；S3, for training set and test set data, extract soil spectrum curve trough feature, deposit data acquisition system c1；S4, for training set and test set data, after difference derivation is handled, extract soil spectrum first derivative curve Wave crest and wave trough feature, deposit data acquisition system c2；S5, merging data set c1, c2, obtain attribute data collection L, after operation is normalized to attribute data collection L, based on KNN sorting techniques, carry out the soil types identification of test set soil spectrum curve；S6, based on maximum accounting principle, determine the soil types of test set soil profile.This method has greater advantages in terms of superelevation ties up classification, meets to use in the case where sample data dimension height, soil types are more.
Description
Technical field
The invention belongs to soil remote sensing field, and in particular to one kind based on nearest neighboring rule (K-Nearest Neighbor,
Abbreviation KNN) carry out soil types knowledge method for distinguishing.
Background technology
The classification of soils is the comprehensive differences according to soil entity character, marks off the soil types of different stage, so as to because
Promote agricultural technology and soil amelioration techniques to ground suiting measures to different conditions.Identification for soil types need to obtain substantial amounts of soil body form,
The diagnostic messages such as physics, chemistry or even biology, which part information (such as form, unit weight) can observe by the naked eye or letter
Single measurement obtains, but most of physics, chemical information traditionally need to obtain by laboratory testing analysis.So
Identification and identification typically cost to soil types is higher, and needs the participation of classification of soils expert.
At present, in order to meet the extensive high precision soil information requirement of fast development, quick obtaining and continuous updating are native
Earth information is one of core research contents of soil resource research field.Conventional method is generally deposited because its cost is high, efficiency is low
Sampling scale is bigger than normal, sampling density is partially sparse, investigation frequency it is relatively low the problems such as, it is difficult to satisfaction soil information is entered Mobile state,
Quickly, the demand that low cost is obtained and updated.
Soil base substance composition change be both soil genesis and development process reflection, be to determine soil diagnosis index and
The foundation of Type division, while be also the principal element for influenceing soil reflective spectrum.In soil investigation with drawing, utilizing spectrum
Analytical technology realize soil types it is quick, accurate, be automatically identified in it is theoretically feasible.Realize that the key of above-mentioned application exists
In, the spectroscopic data of acquisition soil, the soil spectrum storehouse of Erecting and improving under the condition determination of relative standard, and then research and develop effective
Mode identification method with reach soil spectrum identification and classification target.
Liu Huanjun etc. (2008,2017) is based on BP neural network method, has obtained Jilin Nongan County black earth, chernozem etc. five
Kind of MAIN SOILS great soil group precise classification (referring to《Classification of soils research based on reflective spectral property》, Liu Huanjun, Zhang Bai,
Deep intelligence, wait spectroscopy and spectrum analysis, 2008,28 (3):624-628.；Utilize multilayer perceptron neural network model combination light
The classification of soils method of spectrum signature parameter, the Chinese patent .CN such as Liu Huanjun, Zhang Xinle 106650819A.2017-05-
10.) the related soil type, in this method is less, has some superiority with BP neural net methods.But work as and face more soil
When type and a large amount of soil samples, the learning time of this method is long, in some instances it may even be possible to do not reach the destination of study, so as in face of
When more soil types and a large amount of soil sample data, general classification performance is unsatisfactory.In addition, it use only in correlation technique
0-20cm topsoil soil samples, for the soil spectrum information of different depth in same profile, fail to make full use of.
The content of the invention
For above-mentioned technical problem, the present invention intends being directed to the soil spectrum information of different depth in soil profile, is extracting
On the basis of training set and test set soil spectrum curvilinear characteristic, based on KNN sorting techniques, test set is accurately and rapidly carried out
The soil types identification of soil profile.
In order to realize above-mentioned technical purpose, the present invention uses following concrete technical scheme：
One kind carries out soil profile kind identification method, including following steps based on KNN classification：
S1, for a curve of spectrum, extract its trough feature；
S2, for a curve of spectrum, extract its crest feature；
S3, for training set and test set data, extract soil spectrum curve trough feature, deposit data acquisition system c1；
S4, for training set and test set data, after difference derivation is handled, using step S1 and step S2 methods, carry
Take soil spectrum first derivative curve Wave crest and wave trough feature, deposit data acquisition system c2；
Data acquisition system c1, c2 that S5, combining step S3 and step S4 are respectively obtained, attribute data collection L is obtained, to attribute number
After operation is normalized according to collection L, based on KNN sorting techniques, the soil types identification of test set soil spectrum curve is carried out；
S6, based on maximum accounting principle, determine the soil types of test set soil profile.
The step S1 is specifically included：
S1.1：Trough extracts, and extracts trough using trough extraction algorithm, and be stored in set G={ gb| b=0 ..., M-1 }
In, wherein, M is trough quantity, gb={ r'a| a=0 ..., N-1 } it is a trough element set, N is the number of the point in the trough
Amount；
S1.2：Trough feature extraction, for a trough gb, perform following operate：
A) the minimum point r' for meeting following conditions is extractedmin, the point is the lowest point point of the trough； r'min≤r'c
(c=0 .., N-1)
Wherein, r'cFor the spectral reflectance values at any point in the trough；
B) according to formula (1) and formula (2), the depth H and width Q of the trough are calculated；
H=(r'0+r'N-1)/2-r'min (1)
Q=N (2)
Wherein, r'0Point value, r' are originated for troughN-1Point value is terminated for trough.
C) circulation performs step a) to step b), until completing the feature extraction of all troughs of the curve of spectrum；
D) by the lowest point point band po sition min of different troughs, the lowest point point value r'min, trough depth H and trough width Q are line by line
It is saved in matrix VM*4In.
The step S2 is specifically included：
S2.1：Crest extracts, and extracts crest using crest extraction algorithm, and be stored in set G'={ g'b'| b'=0 ...,
M'-1 } in, wherein M' is crest quantity, g'b'={ r "a'| a'=0 ..., N'-1 } it is a crest element set, N' is the crest
In point quantity；
S2.2：Crest feature extraction, for a crest g'b', perform following operate：
A) the maximum of points r " for meeting following conditions is extractedmax, the point is the peak maximum of the crest；
r”max≥r”c' (c'=0 .., N'-1)
Wherein, r "c'For the spectral reflectance values at any point in the crest；
B) according to formula (3), (4), calculate the crest depth H ' and width Q'；
H'=(r "0+r”N'-1)/2-r”max (3)
Q'=N'(4)
Wherein, r "0Point value, r " are originated for crestN'-1Point value is terminated for crest.
C) circulation performs step a) to step b), until completing the feature extraction of all crests of the curve of spectrum；
D) by the peak maximum band po sition max of different crests, summit point value r "max, crest height H' and wave peak width Q' by
Row be saved in matrix V 'M'*4In.
The step S3 is specifically included：
S3.1:Any soil profile spectroscopic data in training set and test set is read, using equivalance Quadratic Spline Interpolation side
Method, the reflectance value of layer respectively occurs for soil profile, carry out depth interpolation processing, obtain pre-determined distance interval, different depth
The soil spectrum reflectivity T={ t at placei,j| i=0 ..., n-1；J=0 ..., m-1 }, wherein, n represents the soil light after interpolation
Spectral curve quantity, m represent soil spectrum wave band number, ti,jRepresent that i-th curve of spectrum that interpolation is generated is anti-at wave band j
Radiance rate value；
S3.2:Data smoothing, for the soil spectrum reflectivity obtained through depth interpolation processing, carry out 11 point movements
Average smooth processing, obtain smooth rear data acquisition system T'；
S3.3:Envelope place to go is handled, and using envelope removal method, place is normalized to any curve of spectrum in T'
Reason, as a result it is stored in set R={ re| e=0 ..., h-1 } in；Wherein, reRepresent curve of spectrum reflectivity at wave band e+350
Normalization result, h is the curve of spectrum wave band quantity；
S3.4:Trough feature extraction, abovementioned steps S3 is performed to interpolation soil layer data in set R, obtains each interpolation soil layer
Soil spectrum curvilinear characteristic matrix SP*4, wherein, P represents the trough quantity of the interpolation soil layer curve of spectrum；Before extracting matrix S
Two trough features, preserve into set c1；
S3.5:Circulation performs step S3.1 to step S3.4, until completing the data processing of all soil spectrum sections.
The step S4 is specifically included：
S4.1:Data acquisition system T' after obtaining any soil profile spectroscopic data smoothly using abovementioned steps S3.3；
S4.2:According to formula (5), difference and enhanced processing are carried out to data T', are as a result stored in set Y={ yi,j| i=
0,...,n-1；J=0 ..., m-1 } in；
Wherein, D is amplification factor；
S4.3:Envelope removal is handled, and using envelope removal method, any curve of spectrum in set Y is normalized
Processing, as a result it is stored in set R={ re| e=0 ..., h-1 } in；Wherein, reRepresent that the curve of spectrum reflects at wave band e+350
The normalization result of rate, h are the curve of spectrum wave band number；
S4.4:Trough feature extraction, abovementioned steps S4 is performed to interpolation soil layer data in set R, obtains each interpolation soil layer
Soil spectrum curvilinear characteristic matrix UX*4, wherein, X represents the trough quantity of the interpolation soil layer first derivative curve；Choose each interpolation
Soil layer soil spectrum curvilinear characteristic matrix UX*4In first trough the lowest point value and soil of the lowest point wave band as the interpolation soil layer
Spectrum first derivative curve trough feature is preserved into set c2；
S4.5:Crest feature extraction, abovementioned steps S2 is performed to any interpolation soil layer data in set Y, obtains each interpolation
Soil layer soil spectrum curvilinear characteristic matrix WZ*4, wherein, Z represents the crest quantity of the interpolation soil layer first derivative curve；Choose each
Interpolation soil layer soil spectrum curvilinear characteristic matrix WZ*4In first crest summit value and summit wave band as the interpolation soil layer
Soil spectrum first derivative curve crest feature is preserved into set c2；
S4.6:Circulation performs step S4.1 to S4.5, until completing the data processing of all soil spectrum sections.
The step S5 is specifically included：
S5.1：Data acquisition system c1, c2 that combining step S3 and step S4 are respectively obtained, obtain attribute data collection L={ li,j
| i=0 ..., l1-1；J=0 ..., l2-1 }, wherein, l1 represents whole soil profile interpolation soil layer quantity, and l2 represents each
Soil layer feature quantity；
S5.2：According to formula (6), to all characteristic element l in data set Li,jOperation is normalized, is normalized
Characteristic set L' afterwards；
l'i,j=(li,j-lmin,j)/(lmax,j-lmin,j) (6)
Wherein, lmin,jFor the minimum value of all j-th of features of interpolation soil layer, lmax,jIt is special for all interpolation soil layer jth
The maximum of sign；
S5.3：Arrange parameter k, KNN algorithms are called, obtain the classification knot of each interpolation soil layer curve of spectrum of test set sample
Fruit.
The step S6 is specifically included：
S6.1：Calculate the proportion λ of each soil types in each interpolation soil layer curve of spectrum classification results in a soil profile
(A)；Find λ (A) maximum MAXλ(A), then it is assumed that the section is class A；
S6.2：Circulation performs step S6.1, until the soil types for completing all soil profiles in test set differentiates.
It is of the invention compared with existing soil identification technology, have the advantages that：
The present invention is proposed after a kind of comprehensive utilization envelope removal processing after soil spectrum curvilinear characteristic and derivation processing
Soil spectrum curvilinear characteristic, carry out soil profile type with KNN sorting techniques and quickly know method for distinguishing, this method mainly has
Following characteristics：
1) spectral information of different depth soil sample is taken full advantage of.
2) fully utilize soil spectrum curvilinear characteristic and derivation processing after soil spectrum curvilinear characteristic.
3) the KNN methods used, than BP neural network method, there are greater advantages in terms of superelevation ties up classification, meet
Used in the case that sample data dimension is high, soil types is more.
Brief description of the drawings
Fig. 1 is the flow chart of the inventive method；
Embodiment
Technical scheme is described in further details with reference to the accompanying drawings and examples.
The soil sample that the present embodiment uses picks up from 649 sections.These sections divide according to Soil Taxonomy
For 10 order of soils, 23 subclasses, 56 great soil groups and 120 subclass.Using the spectrophotometer (Agilent of Cary 5000
Technologies the 350-2500nm diffusing reflection spectrums of sample) are measured, spectrum sample is at intervals of 1nm.
The present embodiment selects 553 soil spectrum cross-sectional datas therein as training set, 96 soil spectrum section numbers
According to as test set.For data prediction, Spectra feature extraction, the spectral classification based on KNN, soil profile type identification etc.
Whole process, be further described the present invention.Fig. 1 is the flow chart of the inventive method.
1. soil spectrum curvilinear characteristic extracts.
1-1, any soil profile spectroscopic data in training set and test set is read, interpolation is set at intervals of 1cm, using etc.
Product Quadratic Spline Interpolation method, for the reflectance value of each interpolation soil layer of soil profile, depth interpolation processing is carried out, obtained different
Soil spectrum reflectivity T={ t of the depth using 1cm as intervali,j| i=0 ..., 73432；J=0 ..., 2150 }, it is as follows
Shown in table 1.
Table 1
1-2, the soil spectrum reflectivity t for being obtained through depth interpolation processingi,j, it is smooth to carry out 11 rolling averages
Processing, smooth rear data acquisition system T' is obtained, it is as shown in table 2 below.
Table 2
1-3, trough feature extraction.Envelope removal processing is carried out to any interpolation soil layer data in set T ', before execution
Step S1 is stated, obtains each interpolation soil layer soil spectrum curvilinear characteristic matrix.It is as shown in table 3 below；
Table 3
2. soil spectrum first derivative curvilinear characteristic extracts.
2-1, soil profile spectroscopic data smoothly rear data acquisition system T ' is obtained using abovementioned steps S3.3 (as shown in table 2)；
2-2, according to formula (5), to data T ' carry out difference and enhanced processing, be as a result stored in set Y, such as table 4 below institute
Show；
Table 4
2-3, trough feature extraction, envelope removal processing is carried out to any interpolation soil layer data in set Y, performed foregoing
Step S1, obtain each interpolation soil layer soil spectrum first derivative curve trough eigenmatrix U.Choose each interpolation soil layer soil spectrum
The lowest point value of first trough is led with single order of the lowest point wave band as the interpolation soil layer in first derivative curve trough eigenmatrix U
Number curve trough feature is preserved into set c2；
2-4, crest feature extraction, envelope removal processing is carried out to any interpolation soil layer data in set Y, performed foregoing
Step S2, obtain each interpolation soil layer soil spectrum first derivative curve crest eigenmatrix W.Choose each interpolation soil layer soil spectrum
The summit value of first crest is led with single order of the summit wave band as the interpolation soil layer in first derivative curve crest eigenmatrix W
Number curve crest feature is preserved into set c2；
2-5, circulation perform step 2-1 to 2-4, until completing the number of all soil spectrum sections in training set and test set
It is as shown in table 5 below according to processing；
Table 5
3. the soil types identification of the test set soil interpolation soil layer curve of spectrum.
3-1, merging data set c1, c2, obtain characteristic data set L, and the curvilinear characteristic for participating in calculating mainly includes：Soil
Curve of spectrum difference depth UpperDepth, LowerDepth, the lowest point position Position1 of first absorption paddy, the lowest point value
Value1, trough depth Depth1, trough width xDistance1, the lowest point position Position2 of second absorption paddy, the lowest point
Value Value2, trough depth Depth2, trough width xDistance2, first absorption paddy of soil spectrum first derivative curve
The lowest point position P1, the lowest point value V1, summit position P2, the summit value V2 of first absworption peak；It is as shown in table 6 below；
Table 6
3-2, according to formula (6), to all characteristic element l in data set LI, jOperation is normalized, after obtaining normalization
Characteristic set L ', it is as shown in table 7 below；
Table 7
3-3, arrange parameter k=201, KNN algorithms are called using R.NET, obtain each interpolation soil layer spectrum of test set sample
The classification results of curve, it is as shown in table 8 below；
Table 8
4. the soil types of test set soil spectrum section differentiates
4-1, calculate in test set in first soil profile each soil class in each difference soil layer curve of spectrum classification results
The proportion λ (A) of type, it is as shown in table 9 below, judge the section for dry profit sandic entisols；
| GroupCST | Percent |
| Dry profit sandic entisols | 46.67% |
| Moistening ferrallite is educated in letter | 10.83% |
| The normal wet Cambisol of aluminum | 0.01% |
| Incobation water separating | 0.05% |
| Sand ginger humidity Cambisol | 2.50% |
| Moisten orthent | 15.83% |
| Iron oozes water separating | 9.17% |
| Irony Udalf | 5.83% |
| Purple moistens Cambisol | 3.33% |
Table 9
4-2, circulation perform step 4-1, until the soil types for completing all soil profiles in test set differentiates, as a result such as
Shown in table 10 below；
Table 10
In above-mentioned Application Example, the identification of soil great soil group has only been carried out.This method may be equally applied to the order of soil, subclass,
The identification of the different stage soil types such as subclass；And in the present embodiment, it is based only upon the Visible-to-Near InfaRed modal data that diffuses and enters
The identification of row soil types, this method other types of soil spectrum data such as infrared diffusing reflection spectrum in may be equally applied to.
5. test analysis.
From above-described embodiment：Aspect is identified in the great soil group of soil profile, the recognition methods is for light color humidity blank
Soil, letter educate water separating, the poly- water separating of iron, stick the discrimination that the great soil group such as ferrisol, incobation water separating is moistened in dampness elimination
Reach more than 60%.In the present embodiment, test set data cover 6 provinces, are related to 21 soil types, 96 soil-likes
Product, overall nicety of grading have reached 54.8%.With enriching constantly and Different Soil section number for training set cross-sectional data
According to continuous balance, accuracy of identification will improve constantly.
Compared to Liu Huanjun etc. (2008,2017) method, this method there are following characteristics：
1) spectral information of different depth soil sample is taken full advantage of；
2) make use of soil spectrum curvilinear characteristic and derivation handle after soil spectrum curvilinear characteristic；
3) the KNN methods used, than BP neural network method, there are greater advantages in terms of superelevation ties up classification, can meet
Used in the case where sample data dimension height, soil types are more.
Claims (7)
1. one kind carries out soil profile kind identification method based on KNN classification, it is characterised in that including following steps：
S1, for a curve of spectrum, extract its trough feature；
S2, for a curve of spectrum, extract its crest feature；
S3, for training set and test set data, extract soil spectrum curve trough feature, deposit data acquisition system c1；
S4, for training set and test set data, after difference derivation is handled, using step S1 and step S2 methods, extraction soil
Earth spectrum first derivative curve Wave crest and wave trough feature, deposit data acquisition system c2；
Data acquisition system c1, c2 that S5, combining step S3 and step S4 are respectively obtained, attribute data collection L is obtained, to attribute data collection
After operation is normalized in L, based on KNN sorting techniques, the soil types identification of test set soil spectrum curve is carried out；
S6, based on maximum accounting principle, determine the soil types of test set soil profile.
2. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S1 is specifically included：
S1.1：Trough extracts, and extracts trough using trough extraction algorithm, and be stored in set G={ gb| b=0 ..., M-1 } in, its
In, M is trough quantity, gb={ r'a| a=0 ..., N-1 } it is a trough element set, N is the quantity of the point in the trough；
S1.2：Trough feature extraction, for a trough gb, perform following operate：
A) the minimum point r' for meeting following conditions is extractedmin, the point is the lowest point point of the trough；
r'min≤r'c(c=0 .., N-1)
Wherein, r'cFor the spectral reflectance values at any point in the trough；
B) according to formula (1) and formula (2), the depth H and width Q of the trough are calculated；
H=(r'0+r'N-1)/2-r'min (1)
Q=N (2)
Wherein, r'0Point value, r' are originated for troughN-1Point value is terminated for trough.
C) circulation performs step a) to step b), until completing the feature extraction of all troughs of the curve of spectrum；
D) by the lowest point point band po sition min of different troughs, the lowest point point value r'min, trough depth H and trough width Q are preserved line by line
To matrix VM*4In.
3. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S2 is specifically included：
S2.1：Crest extracts, and extracts crest using crest extraction algorithm, and be stored in set G'={ g'b'| b'=0 ..., M'-1 }
In, wherein M' is crest quantity, g'b'={ r "a'| a'=0 ..., N'-1 } it is a crest element set, N' is in the crest
Point quantity；
S2.2：Crest feature extraction, for a crest g'b', perform following operate：
A) the maximum of points r " for meeting following conditions is extractedmax, the point is the peak maximum of the crest；r”max≥r”c'(c'=
0,..,N'-1)
Wherein, r "c'For the spectral reflectance values at any point in the crest；
B) according to formula (3) and formula (4), calculate the crest depth H ' and width Q'；
H'=(r "0+r”N'-1)/2-r”max (3)
Q'=N'(4)
Wherein, r "0Point value, r " are originated for crestN'-1Point value is terminated for crest.
C) circulation performs step a) to step b), until completing the feature extraction of all crests of the curve of spectrum；
D) by the peak maximum band po sition max of different crests, summit point value r "max, crest height H' and wave peak width Q' are protected line by line
Be stored to matrix V 'M'*4In.
4. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S3 is specifically included：
S3.1:Any soil profile spectroscopic data in training set and test set is read, using equivalance Quadratic Spline Interpolation method, pin
The reflectance value that layer occurs each to soil profile, carries out depth interpolation processing, obtains pre-determined distance interval, the soil at different depth
Earth spectral reflectivity T={ ti,j| i=0 ..., n-1；J=0 ..., m-1 }, wherein, n represents the soil spectrum curve after interpolation
Quantity, m represent soil spectrum wave band number, ti,jReflectance value of i-th curve of spectrum that expression interpolation is generated at wave band j；
S3.2:Data smoothing, for the soil spectrum reflectivity obtained through depth interpolation processing, carry out 11 rolling averages and put down
Sliding processing, obtain smooth rear data acquisition system T'；
S3.3:Envelope place to go is handled, and using envelope removal method, any curve of spectrum in T' is normalized, and is tied
Fruit deposit set R={ re| e=0 ..., h-1 } in；Wherein, reRepresent the normalizing of curve of spectrum reflectivity at wave band e+350
Change result, h is the curve of spectrum wave band quantity；
S3.4:Trough feature extraction, abovementioned steps S3 is performed to interpolation soil layer data in set R, obtains each interpolation soil layer soil
Curve of spectrum eigenmatrix SP*4, wherein, P represents the trough quantity of the interpolation soil layer curve of spectrum；Extract matrix S the first two
Trough feature, preserve into set c1；
S3.5:Circulation performs step S3.1 to step S3.4, until completing the data processing of all soil spectrum sections.
5. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S4 is specifically included：
S4.1:Data acquisition system T' after obtaining any soil profile spectroscopic data smoothly using abovementioned steps S3.3；
S4.2:According to formula (5), difference and enhanced processing are carried out to data T', are as a result stored in set Y={ yi,j| i=0 ...,
n-1；J=0 ..., m-1 } in；
Wherein, D is amplification factor；
S4.3:Envelope removal is handled, and using envelope removal method, any curve of spectrum in set Y is normalized,
As a result set R={ r are stored ine| e=0 ..., h-1 } in；Wherein, reRepresent that curve of spectrum reflectivity at the wave band e+350 is returned
One changes result, and h is the curve of spectrum wave band number；
S4.4:Trough feature extraction, abovementioned steps S4 is performed to interpolation soil layer data in set R, obtains each interpolation soil layer soil
Curve of spectrum eigenmatrix UX*4, wherein, X represents the trough quantity of the interpolation soil layer first derivative curve；Choose each interpolation soil layer
Soil spectrum curvilinear characteristic matrix UX*4In first trough the lowest point value and soil spectrum of the lowest point wave band as the interpolation soil layer
First derivative curve trough feature is preserved into set c2；
S4.5:Crest feature extraction, abovementioned steps S4 is performed to any interpolation soil layer data in set Y, obtains each interpolation soil layer
Soil spectrum curvilinear characteristic matrix WZ*4, wherein, Z represents the crest quantity of the interpolation soil layer first derivative curve；Choose each interpolation
Soil layer soil spectrum curvilinear characteristic matrix WZ*4In first crest summit value and soil of the summit wave band as the interpolation soil layer
Spectrum first derivative curve crest feature is preserved into set c2；
S4.6:Circulation performs step S4.1 to S4.5, until completing the data processing of all soil spectrum sections.
6. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S5 is specifically included：
S5.1：Data acquisition system c1, c2 that combining step S3 and step S4 are respectively obtained, obtain attribute data collection L={ li,j| i=
0,...,l1-1；J=0 ..., l2-1 }, wherein, l1 represents whole soil profile interpolation soil layer quantity, and l2 represents each soil layer
Feature quantity；
S5.2：According to formula (6), to all characteristic element l in data set Li,jOperation is normalized, it is special after being normalized
L' is closed in collection；
l'i,j=(li,j-lmin,j)/(lmax,j-lmin,j) (6)
Wherein, lmin,jFor the minimum value of all j-th of features of interpolation soil layer, lmax,jFor all j-th of features of interpolation soil layer most
Big value；
S5.3：Arrange parameter k, KNN algorithms are called, obtain the classification results of each interpolation soil layer curve of spectrum of test set sample.
7. according to claim 1 carry out soil profile kind identification method based on KNN classification, it is characterised in that described
Step S6 is specifically included：
S6.1：Calculate the proportion λ (A) of each soil types in each interpolation soil layer curve of spectrum classification results in a soil profile；Look for
To λ (A) maximum MAXλ(A), then it is assumed that the section is class A；
S6.2：Circulation performs step S6.1, until the soil types for completing all soil profiles in test set differentiates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710430616.6A CN107657264B (en) | 2017-06-09 | 2017-06-09 | KNN classification-based soil profile type identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710430616.6A CN107657264B (en) | 2017-06-09 | 2017-06-09 | KNN classification-based soil profile type identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107657264A true CN107657264A (en) | 2018-02-02 |
| CN107657264B CN107657264B (en) | 2020-11-06 |
Family
ID=61127534
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710430616.6A Expired - Fee Related CN107657264B (en) | 2017-06-09 | 2017-06-09 | KNN classification-based soil profile type identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107657264B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111693491A (en) * | 2020-07-30 | 2020-09-22 | 重庆理工大学 | Method for measuring refractive index of transparent fluid based on Fabry-Perot interference |
| CN112580697A (en) * | 2020-12-04 | 2021-03-30 | 国网山东省电力公司电力科学研究院 | Method and system for selecting grounding of power transmission line tower |
| CN114324216A (en) * | 2022-01-06 | 2022-04-12 | 中国科学院南京土壤研究所 | Soil numerical value classification method based on soil layer combination characteristics |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070073491A1 (en) * | 2005-09-14 | 2007-03-29 | The Regents Of The University Of California | Method for soil content prediction based on a limited number of mid-infrared absorbances |
| CN102096825A (en) * | 2011-03-23 | 2011-06-15 | 西安电子科技大学 | Graph-based semi-supervised high-spectral remote sensing image classification method |
| CN102565058A (en) * | 2011-12-26 | 2012-07-11 | 北京林业大学 | Soil section analysis device and method based on image processing |
| CN105758819A (en) * | 2016-02-29 | 2016-07-13 | 上海交通大学 | Method for detecting organic components of soil by utilizing near infrared spectrum |
| CN106124449A (en) * | 2016-06-07 | 2016-11-16 | 中国科学院合肥物质科学研究院 | A kind of soil near-infrared spectrum analysis Forecasting Methodology based on degree of depth learning art |
| CN106126879A (en) * | 2016-06-07 | 2016-11-16 | 中国科学院合肥物质科学研究院 | A kind of soil near-infrared spectrum analysis Forecasting Methodology based on rarefaction representation technology |
| CN106650819A (en) * | 2016-12-30 | 2017-05-10 | 东北农业大学 | Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters |
-
2017
- 2017-06-09 CN CN201710430616.6A patent/CN107657264B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070073491A1 (en) * | 2005-09-14 | 2007-03-29 | The Regents Of The University Of California | Method for soil content prediction based on a limited number of mid-infrared absorbances |
| CN102096825A (en) * | 2011-03-23 | 2011-06-15 | 西安电子科技大学 | Graph-based semi-supervised high-spectral remote sensing image classification method |
| CN102565058A (en) * | 2011-12-26 | 2012-07-11 | 北京林业大学 | Soil section analysis device and method based on image processing |
| CN105758819A (en) * | 2016-02-29 | 2016-07-13 | 上海交通大学 | Method for detecting organic components of soil by utilizing near infrared spectrum |
| CN106124449A (en) * | 2016-06-07 | 2016-11-16 | 中国科学院合肥物质科学研究院 | A kind of soil near-infrared spectrum analysis Forecasting Methodology based on degree of depth learning art |
| CN106126879A (en) * | 2016-06-07 | 2016-11-16 | 中国科学院合肥物质科学研究院 | A kind of soil near-infrared spectrum analysis Forecasting Methodology based on rarefaction representation technology |
| CN106650819A (en) * | 2016-12-30 | 2017-05-10 | 东北农业大学 | Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters |
Non-Patent Citations (3)
| Title |
|---|
| A. M. SALEH ET AL.: "Identification and Mapping of Some Soil Types Using Field Spectrometry and Spectral Mixture Analyses: A Case Study of North Sinai, Egypt", 《ARAB J GEOSCI》 * |
| 刘焕军 等: "基于反射光谱特性的土壤分类研究", 《光谱学与光谱分析》 * |
| 郭熙 等: "高光谱特征辨别潴育型麻沙泥田和潮沙泥田水稻土", 《农业工程学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111693491A (en) * | 2020-07-30 | 2020-09-22 | 重庆理工大学 | Method for measuring refractive index of transparent fluid based on Fabry-Perot interference |
| CN111693491B (en) * | 2020-07-30 | 2023-01-06 | 重庆理工大学 | Method for measuring refractive index of transparent fluid based on Fabry-Perot interference |
| CN112580697A (en) * | 2020-12-04 | 2021-03-30 | 国网山东省电力公司电力科学研究院 | Method and system for selecting grounding of power transmission line tower |
| CN112580697B (en) * | 2020-12-04 | 2023-05-26 | 国网山东省电力公司电力科学研究院 | Power transmission line tower grounding selection method and system |
| CN114324216A (en) * | 2022-01-06 | 2022-04-12 | 中国科学院南京土壤研究所 | Soil numerical value classification method based on soil layer combination characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107657264B (en) | 2020-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fajardo et al. | Fuzzy clustering of Vis–NIR spectra for the objective recognition of soil morphological horizons in soil profiles | |
| CN107748146A (en) | A kind of crude oil attribute method for quick predicting based near infrared spectrum detection | |
| CN102313699B (en) | Estimation method of total nitrogen content in crop canopy leaf | |
| Piri Sahragard et al. | Modeling habitat suitability of range plant species using random forest method in arid mountainous rangelands | |
| Zhang et al. | Allocate soil individuals to soil classes with topsoil spectral characteristics and decision trees | |
| Liu et al. | A method combining ELM and PLSR (ELM-P) for estimating chlorophyll content in rice with feature bands extracted by an improved ant colony optimization algorithm | |
| Mouazen et al. | Potential of visible and near-infrared spectroscopy to derive colour groups utilising the Munsell soil colour charts | |
| CN107657264A (en) | One kind carries out soil profile kind identification method based on KNN classification | |
| Bochenek et al. | Monitoring forest biodiversity and the impact of climate on forest environment using high-resolution satellite images | |
| CN106339674A (en) | Hyperspectral image classification method based on edge preservation and graph cut model | |
| Singh et al. | Biomass estimation of mixed forest landscape using a Fourier transform texture-based approach on very-high-resolution optical satellite imagery | |
| CN105424641A (en) | Crude oil type near infrared spectrum identification method | |
| de Almeida Papa et al. | Evaluating tropical forest classification and field sampling stratification from lidar to reduce effort and enable landscape monitoring | |
| CN107958267A (en) | A kind of oil property Forecasting Methodology represented based on linear | |
| CN108827909B (en) | Rapid soil classification method based on visible near infrared spectrum and multi-target fusion | |
| CN106990056A (en) | A kind of total soil nitrogen spectrum appraising model calibration samples collection construction method | |
| CN108661630A (en) | One kind being based on the preferred geology dessert quantitative evaluation method of parameter | |
| CN109668850A (en) | Herbal nature recognition methods and system based on ultraviolet fingerprint | |
| Wang et al. | The minimum level for soil allocation using topsoil reflectance spectra: Genus or species? | |
| Chen et al. | Hyperspectral detection of sugar content for sugar-sweetened apples based on sample grouping and SPA feature selecting methods | |
| CN117235510A (en) | Joint roughness prediction method and training method of joint roughness prediction model | |
| Hariharan et al. | An AI-based spectral data analysis process for recognizing unique plant biomarkers and disease features | |
| CN107290297A (en) | A kind of IR spectrum quantitative analysis method and system based on from step study | |
| Wang et al. | Development of soil spectral allocation models considering the effect of soil moisture | |
| CN108663334A (en) | The method for finding soil nutrient spectral signature wavelength based on multiple Classifiers Combination |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201106 |