CN109211814A - It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light - Google Patents
It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light Download PDFInfo
- Publication number
- CN109211814A CN109211814A CN201811271315.4A CN201811271315A CN109211814A CN 109211814 A CN109211814 A CN 109211814A CN 201811271315 A CN201811271315 A CN 201811271315A CN 109211814 A CN109211814 A CN 109211814A
- Authority
- CN
- China
- Prior art keywords
- soil
- song
- music
- face
- dimensional light
- 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 140
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000005192 partition Methods 0.000 title claims abstract description 12
- 230000003595 spectral effect Effects 0.000 claims abstract description 34
- 238000007637 random forest analysis Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 19
- 239000000284 extract Substances 0.000 claims abstract description 4
- 238000001228 spectrum Methods 0.000 claims description 16
- 238000002310 reflectometry Methods 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000012732 spatial analysis Methods 0.000 claims description 3
- 238000004611 spectroscopical analysis Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract 1
- 238000005259 measurement Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000005341 cation exchange Methods 0.000 description 1
- 239000004927 clay Substances 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 238000001055 reflectance spectroscopy Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000031068 symbiosis, encompassing mutualism through parasitism Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
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
It is set a song to music the soil profile kind identification methods of face partition characteristics the invention discloses a kind of based on three-dimensional light, include: the generation layer spectral reflectance data that (1) obtains any soil profile in soil sample library, carries out the unitized processing of depth interpolation processing, depth and continuum removal processing;(2) three-dimensional light is generated according to step (1) treated spectral reflectance data to set a song to music face;(3) three-dimensional light face of setting a song to music is carried out by subregion extracts statistical nature, and charge to subcharacter set fk, k is current soil section serial number;(4) step (1) to (3) is repeated, until all soil profiles are all processed in soil sample library, obtains characteristic set F={ fk| k=1 ..., K }, K is the quantity of soil profile in soil sample library;(5) it is based on characteristic set F, multiple random forest training is carried out, the sample sample size of all categories of same ratio is set when training every time, generates random forest disaggregated model set M;(6) it is based on random forest disaggregated model set M, type identification is carried out to soil profile to be identified.Recognition effect of the present invention is preferable.
Description
Technical field
The present invention relates to geographical and technical field of geographic information, and in particular to one kind is set a song to music face partition characteristics based on three-dimensional light
Soil profile kind identification method.
Background technique
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
The popularization agricultural technology and improvement soil of ground suiting measures to different conditions.For the identification of soil types need to obtain a large amount of soil body form, physics,
The diagnostic messages such as chemistry or even biology, part of information can observe by the naked eye or simply measure to obtain, but most of objects
Reason, chemical information traditionally need to obtain by laboratory testing analysis.So identification and identification to soil types
Typically cost is higher, and needs the participation of classification of soils expert.
Currently, in order to meet fast-developing a large amount of high precision soil information requirements, quick obtaining and continuous updating soil
Information is one of core research contents of soil resource research field.Traditional soil types recognition methods due to its is at high cost,
Usually there is the problems such as sampling scale is bigger than normal, sampling density is partially sparse, investigation frequency is relatively low in low efficiency, it is difficult to meet to soil
The demand that information carries out dynamic, quick, low cost is obtained and updated.
Visible-to-Near InfaRed diffusing reflection spectrum can not only have with quantitative analysis the moisture of clear spectral absorption characteristics, clay,
The basic soil parameters such as ferriferous oxide, organic carbon and full nitrogen, can also preferably predict those do not have clear Absorption Characteristics, with
There are other soil parameters (such as cation exchange capacity (CEC)) of correlativity for basic soil parameters.Since measurement is easy, and without pair
Sample carries out chemical pretreatment, it is seen that-near-infrared the spectral technology that diffuses is all kinds of physical and chemical parameters of quick measurement soil using most
Frequent near-earth spectral technique.
It is carried out in the identification application of soil types however, being currently based on Visible-to-Near InfaRed diffusing reflection spectrum, due to existing not
The generation layer of unbalanced, the different section of the quantity of same type sample, which divides depth difference, causes section soil sample sampling depth inconsistent
Etc. reasons, accuracy of identification it is simultaneously not ideal enough.
Summary of the invention
Goal of the invention: in view of the problems of the existing technology the present invention, provides a kind of special based on three-dimensional light face subregion of setting a song to music
The soil profile kind identification method of sign.
Technical solution: of the present invention to be set a song to music the soil profile kind identification method packets of face partition characteristics based on three-dimensional light
It includes:
(1) the generation layer spectral reflectance data for obtaining any soil profile in soil sample library, carries out at depth interpolation
The unitized processing of reason, depth and continuum removal processing;
(2) three-dimensional light is generated according to step (1) treated spectral reflectance data to set a song to music face;
(3) three-dimensional light face of setting a song to music is carried out by subregion extracts statistical nature, and charge to subcharacter set fk, k is
Current soil section serial number;
(4) step (1) to (3) is repeated, until all soil profiles are all processed in soil sample library, obtains feature
Set F={ fk| k=1 ..., K }, K is the quantity of soil profile in soil sample library;
(5) it is based on characteristic set F, multiple random forest training is carried out, the of all categories of same ratio is set when training every time
Sample sample size generates random forest disaggregated model set M;
(6) it is based on random forest disaggregated model set M, type identification is carried out to soil profile to be identified.
Further, the step (1) specifically includes:
(1-1) obtains its each generation layer spectral reflectivity set P={ p for any soil profile in soil sample libraryij|
I=1 ..., n;J=1 ..., m }, wherein n is spectral band number, and m is the corresponding quantity that layer occurs of soil profile, pijIt indicates
Reflectance value of j-th of the soil profile generation layer at wave band i;
(1-2) depth interpolation processing: equivalance Quadratic Spline Interpolation method is used, soil profile respectively occurs the reflectivity of layer
Value pijCarry out depth interpolation processing, the soil spectrum reflectivity set P'={ p' at acquisition pre-determined distance interval, different depthil
| i=1 ..., n;L=1 ..., m'}, wherein m' indicates the soil spectrum curve quantity after interpolation, p'ilIndicate that interpolation is generated
Reflectance value of the l depth curve of spectrum at wave band i;
The unitized processing of (1-3) depth: judging whether current soil depth profiled d is greater than 120cm, if so, retaining collection
Close the spectral reflectance data in [0,120] section in P';If it is not, then by the spectral reflectivity in section [d+1,120] in set P'
Data are substituted using the spectral reflectance data at depth d;To obtain updated data acquisition system P ";
(1-4) continuum removal processing: the spectral reflectance data in data acquisition system P " is subjected to continuum removal processing.
Further, step (2) specifically includes:
For step (1) treated spectral reflectance data, using wave band as abscissa, using depth as ordinate, generate
The three-dimensional light of tiff format is set a song to music face.
Further, step (3) specifically includes:
(3-1) generates the subregion face figure layer ZL of shp format according to user preset subregion section;
(3-2) sets a song to music face and subregion face figure layer ZL according to the three-dimensional light, is based on ArcGIS spatial analysis interface, executes point
Area's statistical operation;
(3-3) is successively read the minimum value, mean value and standard variance of each subregion, deposit according to subregion statistical result
Characteristic set fk, and by fkIt is put into characteristic set F.
Further, step (5) specifically includes:
(5-1) is based on characteristic set F, sets by user, generates training set and test set;
The number and randomly selected number of features of Random Forest model decision tree is arranged in (5-2), and every kind of soil class is arranged
The random sampling number of type;
(5-3) is based on setup parameter, carries out random forest training, obtains optimal stochastic forest classified model mq;
(5-4) circulation executes step (5-3), sets frequency of training until reaching user, obtains disaggregated model set M=
{mq| q=1 ..., Q }, Q is that user sets frequency of training.
Further, step (6) specifically includes:
(6-1) executes step (1)-(3), generates corresponding characteristic information for the spectroscopic data of soil profile to be identified
Set testF;
(6-2) is based on trained random forest disaggregated model for all soil profile data to be tested in testF
Set M carries out the multiple soil types prediction of soil profile to be identified, multiple prediction result is stored in results set R;
Various types of frequency in (6-3) statistics set R, type corresponding to frequency highest, as current soil to be identified
The soil types of earth section.
The utility model has the advantages that compared with prior art, the present invention its remarkable advantage is: the method for the present invention is comprehensive to use soil profile
The spectral information of middle different depth soil layer generates three-dimensional light and sets a song to music face, so as at any depth and any wave band to soil
Reflectance signature difference between earth section is analyzed;It is based on random forests algorithm, study discovery Different Soil section again
Spectral patterns, the type prediction for unknown soil profile.In addition, generating multiple predictions with Random Forest model set
As a result on the basis of, final type is differentiated based on mode, effectively eliminates the unbalanced of Different Soil section sample size
Influence to disaggregated model.It was verified that the method for the present invention recognition correct rate with higher.
Detailed description of the invention
Fig. 1 is the flow diagram of one embodiment of the present of invention;
Fig. 2 is that the three-dimensional light of soil profile 34-118 in the present embodiment is set a song to music face example.
Specific embodiment
Specific embodiments of the present invention will be described in further detail with reference to the accompanying drawings of the specification.
For the soil profile sample that the present embodiment uses from certain Chinese soil series survey item, soil sample is the hair according to section
Generating layer time is acquired.Screening has the complete section that layer soil sample occurs, and it is unrestrained anti-to measure its Visible-to-Near InfaRed that layer soil sample occurs
Spectrum is penetrated, soil spectrum database is established.Use the Visible-to-Near InfaRed wave band of 5000 spectrophotometer measurement soil sample of Cary
(350-2500nm) diffusing reflection spectrum air-dries soil sample before measurement, is ground up, sieved at (0.25mm) and drying (45 DEG C)
Reason.For the representativeness for guaranteeing sample, the great soil group data for selecting section number of samples to be no less than 40 from soil spectrum database
(being shown in Table 1).
1. great soil group classification of table and its section number of samples
*Soil Taxonomy
As shown in Figure 1, the present embodiment is set a song to music the soil profile type identification side of face subregion statistical nature based on three-dimensional light
Method the following steps are included:
(1) the generation layer spectral reflectance data for obtaining any soil profile in soil sample library, carries out at depth interpolation
The unitized processing of reason, depth and continuum removal processing.It specifically includes:
(1-1) obtains its each generation layer spectral reflectivity set P={ p for any soil profile in soil sample libraryij|
I=1 ..., n;J=1 ..., m }, wherein n is spectral band number, and m is the corresponding quantity that layer occurs of soil profile, pijIt indicates
Reflectance value of j-th of the soil profile generation layer at wave band i;In this experimental example, n is 2151 (350-2500nm), for showing
Example section 34-118, occurs the quantity m=5 of layer;
(1-2) depth interpolation processing: equivalance Quadratic Spline Interpolation method is used, soil profile respectively occurs the reflectivity of layer
Value pijCarry out depth interpolation processing, the soil spectrum reflectivity set P'={ p' at acquisition pre-determined distance interval, different depthil
| i=1 ..., n;L=1 ..., m'}, wherein m' indicates the soil spectrum curve quantity after interpolation, p'ilIndicate that interpolation is generated
Reflectance value of the l depth curve of spectrum at wave band i;In the present embodiment, predetermined depth distance interval is 1cm, and example is cutd open
The depth capacity of face 34-118 is 100cm, m' 100;
The equivalance Quadratic Spline Interpolation method, is disclosed in the following files: 1.Bishop, T.F.A., McBratney,
A.B.,Laslett,G.M.,1999.Modelling soil attribute depth functions with equal-
area quadratic smoothing splines.Geoderma 91,27–45.;2.Malone,B.P.,McBratney,
A.B.,Minasny,B.,Laslett,G.M.,2009.Mapping continuous depth functions of soil
carbon storage and available water capacity.Geoderma 154,138–152.
The unitized processing of (1-3) depth: judging whether current soil depth profiled d is greater than 120cm, if so, retaining collection
Close the spectral reflectance data in [0,120] section in P';If it is not, then by the spectral reflectivity in section [d+1,120] in set P'
Data are substituted using the spectral reflectance data at depth d;To obtain updated data acquisition system P ";In the present embodiment, show
The depth of example section 34-118 is equal to 100cm, and the spectral reflectance data in [101,120] section uses the light at depth 100cm
Compose reflectivity data substitution;
(1-4) continuum removal processing: the spectral reflectance data in data acquisition system P " is subjected to continuum removal processing.
Wherein, the continuum minimizing technology, is disclosed in the following files: Clark, R.N., Roush, T.L.,
1984.Reflectance spectroscopy:quantitative analysis techniques for remote
sensing applications.J.Geophys.Research 89,6329-6340.
(2) three-dimensional light is generated according to step (1) treated spectral reflectance data to set a song to music face.
Specially using wave band as abscissa, using depth as ordinate, the three-dimensional light for generating tiff format is set a song to music face, such as Fig. 2
It is shown.
(3) three-dimensional light face of setting a song to music is carried out by subregion extracts statistical nature, and charge to subcharacter set fk, k is
Current soil section serial number.It specifically includes:
(3-1) generates the subregion face figure layer ZL of shp format according to user preset subregion section;In the present embodiment, subregion rule
Then are as follows: 10cm is an interval on depth direction;On spectrum direction (unit nm), it is divided into [350,440], [440,580],
[580,760],[760,960],[960,1260],[1260,1560],[1560,2060],[2060,2260],[2260,
2400], [2400,2500] totally 10 sections, symbiosis is at 120 subregion sections;
(3-2) sets a song to music face and subregion face figure layer ZL according to the three-dimensional light, is based on ArcGIS spatial analysis interface, executes point
Area's statistical operation;As a result dbf file is written.In the present embodiment, partial-partition statistical result such as the following table 2 of exemplary cross sectional 34-118
It is shown;
Table 2
(3-3) is successively read the minimum value, mean value and standard variance of each subregion, deposit according to subregion statistical result
Characteristic set fk。
(4) step (1) to (3) is repeated, until all soil profiles are all processed in soil sample library, obtains feature
Set F={ fk| k=1 ..., K }, K is the quantity of soil profile in soil sample library.
(5) it is based on characteristic set F, multiple random forest training is carried out, generates random forest disaggregated model set M.Specifically
Include:
(5-1) respectively randomly selects 5 samples from all kinds of type profile samples of table 1 and (amounts to 35 sections as test set
Sample), all types of remaining section samples form training set;
(value is whole features to the number and randomly selected number of features that Random Forest model decision tree is arranged in (5-2)
It is rounded after number sqrt), the random sampling number of every kind of soil types is set;In the present embodiment, Random Forest model is determined
The number of plan tree is set as 2000, and whole number of features are 360, and the integer value after taking its square root is 18, thus randomly selected
Number of features is set as 18.It is 35 that the random sampling quantity of every kind of soil types, which is all provided with,;
(5-3) is based on setup parameter, carries out random forest training, obtains optimal stochastic forest classified model mq;
(5-4) circulation executes step (5-3), sets frequency of training until reaching user, obtains disaggregated model set M=
{mq| q=1 ..., Q }, Q is that user sets frequency of training.In the present embodiment, Q=500;Carry out 500 random samplings and random
Forest classified operation.In this example great soil group category classification, the average value of OOB error is 0.450, median 0.449, standard
Deviation 0.005.Average identification error rate of all categories is shown in Table 3;
Table 3*
* the average value based on 500 random sampling operations
The random forest method, is disclosed in the following files: 1.Breiman, L., 2001.Random
forests.Machine Learning 45,5‐32.;2.Cutler,D.R.,Edwards JR.,T.C.,Beard,K.H.,
Cutler,A.,Hess,K.T.,Gibson,J.,et al.,Random Forests for classification in
ecology.Ecology 88,2783‐2792.
(6) it is based on random forest disaggregated model set M, type identification is carried out to soil profile to be identified.It specifically includes:
(6-1) executes step (1)-(3), generates corresponding characteristic information for the spectroscopic data of soil profile to be identified
Set testF;
(6-2) is based on trained random forest disaggregated model for all soil profile data to be tested in testF
Set M carries out the multiple soil types prediction of soil profile to be identified, multiple prediction result is stored in results set R;
Various types of frequency in (6-3) statistics set R, type corresponding to frequency highest, as current soil to be identified
The soil types of earth section.
In the present embodiment, the type of prediction and actual type of soil profile to be identified are as shown in table 4 below.In great soil group level
On, overall recognition result accuracy is 60%, and different classes of recognition correct rate is between 20%-100% (being shown in Table 5).
Table 4*
*Final type of prediction is the mode value of 500 prediction results
Table 5
The embodiment of the present invention has only carried out the profile type identifying processing of great soil group level in Soil Taxonomy, the party
Method is also applied in other soil classification systems at the soil types identification of the different levels such as great soil group and the order of soil, subclass, subclass
Reason.
Above disclosed is only a preferred embodiment of the present invention, and the right model of the present invention cannot be limited with this
It encloses, therefore equivalent changes made in accordance with the claims of the present invention, is still within the scope of the present invention.
Claims (6)
1. a kind of set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, characterized by comprising:
(1) the generation layer spectral reflectance data for obtaining any soil profile in soil sample library, carries out depth interpolation processing, depth
The unitized processing of degree and continuum removal processing;
(2) three-dimensional light is generated according to step (1) treated spectral reflectance data to set a song to music face;
(3) three-dimensional light face of setting a song to music is carried out by subregion extracts statistical nature, and charge to subcharacter set fk, k is current soil
Earth section serial number;
(4) step (1) to (3) is repeated, until all soil profiles are all processed in soil sample library, obtains characteristic set
F={ fk| k=1 ..., K }, K is the quantity of soil profile in soil sample library;
(5) it is based on characteristic set F, multiple random forest training is carried out, the sample of all categories of same ratio is set when training every time
Sample size generates random forest disaggregated model set M;
(6) it is based on random forest disaggregated model set M, type identification is carried out to soil profile to be identified.
2. according to claim 1 set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, special
Sign is: the step (1) specifically includes:
(1-1) obtains its each generation layer spectral reflectivity set P={ p for any soil profile in soil sample libraryij| i=
1,…,n;J=1 ..., m }, wherein n is spectral band number, and m is the corresponding quantity that layer occurs of soil profile, pijIndicate soil
Reflectance value of j-th of the section generation layer at wave band i;
(1-2) depth interpolation processing: equivalance Quadratic Spline Interpolation method is used, soil profile respectively occurs the reflectance value p of layerij
Carry out depth interpolation processing, the soil spectrum reflectivity set P'={ p' at acquisition pre-determined distance interval, different depthil| i=
1,…,n;L=1 ..., m'}, wherein m' indicates the soil spectrum curve quantity after interpolation, p'ilIndicate that interpolation l generated is deep
Reflectance value of the curve of spectrum at wave band i at degree;
The unitized processing of (1-3) depth: judging whether current soil depth profiled d is greater than 120cm, if so, retaining set P'
In [0,120] section spectral reflectance data;If it is not, then by the spectral reflectance data in section [d+1,120] in set P'
It is substituted using the spectral reflectance data at depth d;To obtain updated data acquisition system P ";
(1-4) continuum removal processing: the spectral reflectance data in data acquisition system P " is subjected to continuum removal processing.
3. according to claim 1 set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, special
Sign is: step (2) specifically includes:
For step (1) treated spectral reflectance data, using wave band as abscissa, using depth as ordinate, tiff is generated
The three-dimensional light of format is set a song to music face.
4. according to claim 1 set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, special
Sign is: step (3) specifically includes:
(3-1) generates the subregion face figure layer ZL of shp format according to user preset subregion section;
(3-2) sets a song to music face and subregion face figure layer ZL according to the three-dimensional light, is based on ArcGIS spatial analysis interface, executes subregion system
Meter operation;
(3-3) is successively read the minimum value, mean value and standard variance of each subregion according to subregion statistical result, is stored in subcharacter
Set fk。
5. according to claim 1 set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, special
Sign is: step (5) specifically includes:
(5-1) is based on characteristic set F, sets by user, generates training set and test set;
The number and randomly selected number of features of Random Forest model decision tree is arranged in (5-2), and every kind of soil types is arranged
Random sampling number;
(5-3) is based on setup parameter, carries out random forest training, obtains optimal stochastic forest classified model mq;
(5-4) circulation executes step (5-3), sets frequency of training until reaching user, obtains disaggregated model set M={ mq| q=
1 ..., Q }, Q is that user sets frequency of training.
6. according to claim 1 set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light, special
Sign is: step (6) specifically includes:
(6-1) executes step (1)-(3), generates corresponding characteristic information set for the spectroscopic data of soil profile to be identified
testF;
(6-2) is based on trained random forest disaggregated model set for all soil profile data to be tested in testF
M carries out the multiple soil types prediction of soil profile to be identified, multiple prediction result is stored in results set R;
Various types of frequency in (6-3) statistics set R, type corresponding to frequency highest, as current soil to be identified cut open
The soil types in face.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811271315.4A CN109211814B (en) | 2018-10-29 | 2018-10-29 | It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811271315.4A CN109211814B (en) | 2018-10-29 | 2018-10-29 | It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109211814A true CN109211814A (en) | 2019-01-15 |
CN109211814B CN109211814B (en) | 2019-08-06 |
Family
ID=64997964
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811271315.4A Expired - Fee Related CN109211814B (en) | 2018-10-29 | 2018-10-29 | It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109211814B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN114565732A (en) * | 2022-03-02 | 2022-05-31 | 中国科学院南京土壤研究所 | Three-dimensional modeling method and device for occurrence layer of dendritic distribution soil |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105844300A (en) * | 2016-03-24 | 2016-08-10 | 河南师范大学 | Optimized classification method and optimized classification device based on random forest algorithm |
CN105868773A (en) * | 2016-03-23 | 2016-08-17 | 华南理工大学 | Hierarchical random forest based multi-tag classification method |
CN106056134A (en) * | 2016-05-20 | 2016-10-26 | 重庆大学 | Semi-supervised random forests classification method based on Spark |
CN107192671A (en) * | 2017-03-28 | 2017-09-22 | 中国科学院南京土壤研究所 | A kind of soil types recognition methods based on spectrum SURFACES MATCHING |
-
2018
- 2018-10-29 CN CN201811271315.4A patent/CN109211814B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105868773A (en) * | 2016-03-23 | 2016-08-17 | 华南理工大学 | Hierarchical random forest based multi-tag classification method |
CN105844300A (en) * | 2016-03-24 | 2016-08-10 | 河南师范大学 | Optimized classification method and optimized classification device based on random forest algorithm |
CN106056134A (en) * | 2016-05-20 | 2016-10-26 | 重庆大学 | Semi-supervised random forests classification method based on Spark |
CN107192671A (en) * | 2017-03-28 | 2017-09-22 | 中国科学院南京土壤研究所 | A kind of soil types recognition methods based on spectrum SURFACES MATCHING |
Non-Patent Citations (1)
Title |
---|
望陈运: ""基于随机森林算法的土壤图斑分解"", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN114565732A (en) * | 2022-03-02 | 2022-05-31 | 中国科学院南京土壤研究所 | Three-dimensional modeling method and device for occurrence layer of dendritic distribution soil |
CN114565732B (en) * | 2022-03-02 | 2023-06-30 | 中国科学院南京土壤研究所 | Three-dimensional modeling method and device for generation layer of branch-shaped distribution soil |
Also Published As
Publication number | Publication date |
---|---|
CN109211814B (en) | 2019-08-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Avolio et al. | A comprehensive approach to analyzing community dynamics using rank abundance curves | |
Campbell et al. | Assessing colour-dependent occupation statistics inferred from galaxy group catalogues | |
Laffan et al. | Assessing endemism at multiple spatial scales, with an example from the Australian vascular flora | |
Shi et al. | Development of a national VNIR soil-spectral library for soil classification and prediction of organic matter concentrations | |
Campbell | The lognormal distribution as a model for bio‐optical variability in the sea | |
Shi et al. | Prediction of soil organic matter using a spatially constrained local partial least squares regression and the C hinese vis–NIR spectral library | |
Rossel et al. | Predicting soil properties from the Australian soil visible–near infrared spectroscopic database | |
Cole et al. | Mock 2dF and SDSS galaxy redshift surveys | |
Birks | Numerical methods for the analysis of diatom assemblage data | |
Derot et al. | Advances in forecasting harmful algal blooms using machine learning models: A case study with Planktothrix rubescens in Lake Geneva | |
de Santana et al. | Comparison of PLS and SVM models for soil organic matter and particle size using vis-NIR spectral libraries | |
Zhang et al. | Impact of aerosol size representation on modeling aerosol‐cloud interactions | |
CN106815643A (en) | Infrared spectrum Model Transfer method based on random forest transfer learning | |
Unger et al. | A visual analysis concept for the validation of geoscientific simulation models | |
Emilio et al. | Assessing the relationship between forest types and canopy tree beta diversity in Amazonia | |
Suzuki et al. | Aerosol effect on the warm rain formation process: Satellite observations and modeling | |
CN109211814B (en) | It is a kind of to be set a song to music the soil profile kind identification methods of face partition characteristics based on three-dimensional light | |
Sun et al. | Digital soil mapping based on empirical mode decomposition components of environmental covariates | |
CN106990056A (en) | A kind of total soil nitrogen spectrum appraising model calibration samples collection construction method | |
LaRue et al. | Evaluating the sensitivity of forest structural diversity characterization to LiDAR point density | |
Henderson et al. | ASRIS: Continental-scale soil property predictions from point data | |
Bock et al. | Biogeographical Classification of the Global Ocean From BGC‐Argo Floats | |
Zhang et al. | Uncertainty analysis of multiple terrestrial gross primary productivity products | |
Mahoney et al. | Continental-scale canopy height modeling by integrating national, spaceborne, and airborne LiDAR data | |
Gottschalk et al. | Correlation and covariance of runoff revisited |
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 |
Granted publication date: 20190806 |