CN110245867A - A kind of grassland degeneration stage division based on bp neural network - Google Patents
A kind of grassland degeneration stage division based on bp neural network Download PDFInfo
- Publication number
- CN110245867A CN110245867A CN201910526608.0A CN201910526608A CN110245867A CN 110245867 A CN110245867 A CN 110245867A CN 201910526608 A CN201910526608 A CN 201910526608A CN 110245867 A CN110245867 A CN 110245867A
- Authority
- CN
- China
- Prior art keywords
- neural network
- grassland
- degeneration
- grassland degeneration
- influence factor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007850 degeneration Effects 0.000 title claims abstract description 54
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 19
- 244000025254 Cannabis sativa Species 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 14
- 238000003062 neural network model Methods 0.000 claims abstract description 14
- 239000004459 forage Substances 0.000 claims abstract description 11
- 206010035148 Plague Diseases 0.000 claims abstract description 9
- 241000700159 Rattus Species 0.000 claims abstract description 9
- 241000607479 Yersinia pestis Species 0.000 claims abstract description 9
- 230000008901 benefit Effects 0.000 claims abstract description 8
- 241000894007 species Species 0.000 claims abstract description 8
- 238000013519 translation Methods 0.000 claims abstract description 5
- 238000004321 preservation Methods 0.000 claims description 2
- 230000009467 reduction Effects 0.000 claims description 2
- 238000011160 research Methods 0.000 abstract description 9
- 230000015556 catabolic process Effects 0.000 abstract description 7
- 238000006731 degradation reaction Methods 0.000 abstract description 7
- 230000004927 fusion Effects 0.000 abstract description 3
- 230000001537 neural effect Effects 0.000 abstract description 3
- 230000035515 penetration Effects 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000002689 soil Substances 0.000 description 15
- 238000011156 evaluation Methods 0.000 description 9
- 238000005259 measurement Methods 0.000 description 7
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 6
- 240000001851 Artemisia dracunculus Species 0.000 description 3
- 238000000701 chemical imaging Methods 0.000 description 3
- 229910052757 nitrogen Inorganic materials 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 241000894006 Bacteria Species 0.000 description 2
- UIIMBOGNXHQVGW-UHFFFAOYSA-M Sodium bicarbonate Chemical compound [Na+].OC([O-])=O UIIMBOGNXHQVGW-UHFFFAOYSA-M 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 239000004411 aluminium Substances 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000015097 nutrients Nutrition 0.000 description 2
- 239000005416 organic matter Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000002028 Biomass Substances 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 241000041088 Delphinium geyeri Species 0.000 description 1
- 101100011509 Drosophila melanogaster Baldspot gene Proteins 0.000 description 1
- 206010061217 Infestation Diseases 0.000 description 1
- 241000880150 Ochotona curzoniae Species 0.000 description 1
- 108091005804 Peptidases Proteins 0.000 description 1
- 241000209504 Poaceae Species 0.000 description 1
- 239000004365 Protease Substances 0.000 description 1
- 102100037486 Reverse transcriptase/ribonuclease H Human genes 0.000 description 1
- 241000052209 Triplasis Species 0.000 description 1
- WYWFMUBFNXLFJK-UHFFFAOYSA-N [Mo].[Sb] Chemical compound [Mo].[Sb] WYWFMUBFNXLFJK-UHFFFAOYSA-N 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000005904 alkaline hydrolysis reaction Methods 0.000 description 1
- 238000004176 ammonification Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- PTKRHFQQMJPPJN-UHFFFAOYSA-N dipotassium;oxido-(oxido(dioxo)chromio)oxy-dioxochromium;sulfuric acid Chemical compound [K+].[K+].OS(O)(=O)=O.[O-][Cr](=O)(=O)O[Cr]([O-])(=O)=O PTKRHFQQMJPPJN-UHFFFAOYSA-N 0.000 description 1
- 238000004821 distillation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002255 enzymatic effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000000643 oven drying Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000000985 reflectance spectrum Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229910000030 sodium bicarbonate Inorganic materials 0.000 description 1
- 235000017557 sodium bicarbonate Nutrition 0.000 description 1
- 239000004016 soil organic matter Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of grassland degeneration stage divisions based on bp neural network, are related to grassland assessment of degradation degree method, comprising the following steps: S1 determines the influence factor of grassland degeneration, obtains initial data relevant to influence factor；S2 handles initial data relevant to influence factor, to obtain normalized number evidence；S3, normalized number is trained in bp neural network model in advance according to being directed into, export Assessment for classification result；S4, translation transcoding is carried out to Assessment for classification result, it is allowed to be exported in friendly way, the beneficial effects of the present invention are: using convex spot cover degree, advantage grass seeds ratio, degeneration indicator species ratio, edible forage ratio and the plague of rats situation 5 main factors for influencing meadow is points of penetration, according to collected multiple data, based on bp neural fusion to the automatic classification of grassland degeneration, research of grassland degeneration etc. can be widely applied to.
Description
Technical field
The present invention relates to grassland assessment of degradation degree method, specifically a kind of grassland degeneration based on bp neural network
Stage division.
Background technique
Grassland degeneration research is an extremely complex engineering, and the evaluation of grassland degree of degeneration directly affect it is subsequent
Carry out the mode and dynamics administered, inaccurate evaluation will will cause a large amount of resource consumption, therefore the evaluation method of science is non-
It is often important.At present on the evaluation problem of meadow, each meadow worker and researcher are mostly by the way of artificial by adopting
Collect data and carry out meadow evaluation, there are many drawbacks such as at high cost, inefficient, accuracy rate is low for this mode, also do not make full use of
Existing data resource.
So far, many meadow research workers have done many researchs in terms of the evaluation of meadow, and it is " green should sternly to have paper
In extra large Hunan bank grassland degeneration remoteensing evaluation ", 2002~2013 years Growing season NDVI (5~September), inverting Qinghai Hunan bank are utilized
Hainan state area Grass cover degree and deterioration index carry out Qinghai Hunan bank grassland situation, true and trend analysis of degenerating.
Grass cover degree is divided into high, normal, basic three levels, the calculating formula of Grass cover grade:Wherein NDVI vegetation index NDVIminCorresponding 5% coverage, NDVImaxCorresponding 98%
Coverage.Grade scale are as follows: low covering (0≤Fc<0.20), middle covering (0.20≤Fc<0.50), high covering (Fc>=0.50).
Liang Cunli is directed to Tibet Grassland in paper " Fuzzy AHP is in the application in Tibet Grassland degeneration research "
The origin cause of formation of degeneration and the effect problem of Restoring measures have studied Tibet Grassland degeneration using Fuzzy AHP
The efficiency of the origin cause of formation and various Restoring measures.Article points out that the severe tribute of overgrazed, Climate Anomalies, natural conditions is
The main reason for causing Tibet Grassland to be degenerated；Followed by mouse worm poison Weed infestation.
Correlation analysis shows soil nitrification rate and ammonification rate and soil nitrification bacterium and amonifying bacteria in Alpine Grasslands
Quantity and protease and urase it is closely related.Phytomass, soil moisture content, organic carbon, total nitrogen content pass through influence
Micro organism quantity, microbes biomass and enzymatic activity and become influence transformation of soil nitrogen principal element.
Jin is magnificent, Wu Hongqi, and Fan Yanmin et al. is in paper " the bloom spectrum discrimination side of Yi Li thin,tough silk wormwood artemisia Desert Grassland degradation level
Using Yi Li thin,tough silk wormwood artemisia Desert Grassland as research object in method ", it is based on environmental satellite HSI Hyperspectral imaging and group's canopy spectra, is adopted
Grassland degeneration grade is identified with Spectral angle mapper method and spectral information divergence method.With the canopy reflectance spectra acquired on the spot
It is poor for the HSI Hyperspectral imaging accuracy of identification of guidance；Degradation level recognition result based on HSI Hyperspectral imaging is preferable, always
Body nicety of grading is suitble to identify Yi Li thin,tough silk wormwood artemisia Desert Grassland degradation level 76% or more.
Jilin Province west is combined in paper " the Grassland in West Jilin Province degradation evaluation based on optics and radar image " in white
Portion's area feature is broken through previous single with grassland vegetation degeneration index (grass yield, edible forage ratio, coverage etc.) exploration grass
Grassland vegetation parameter and grass lower soil parameters (conductivity, PH) are combined, are moved back to Grassland in West Jilin Province by the way that ground is degenerated
Change and carry out overall merit, the result obtained is made to be more nearly the truth of Grassland in West Jilin Province degeneration.According to the Chinese people
Republic's national standard-grassland degeneration grade scale (GB19377-2003) evaluates and tests 1990-2016 Grassland in West Jilin Province
The degree of degeneration provides decision support and data supporting for area's ecological construction, resource optimization and the sustainable development.
But to the research of grassland degeneration, there is still a need for carry out meadow by acquisition data by the way of artificial to comment above
Valence, large labor intensity are based on this, and present applicant proposes a kind of grassland degeneration stage divisions based on bp neural network.
Summary of the invention
The purpose of the present invention is to provide a kind of grassland degeneration stage divisions based on bp neural network, to solve above-mentioned back
The problem of being proposed in scape technology.
To achieve the above object, the invention provides the following technical scheme:
A kind of grassland degeneration stage division based on bp neural network, comprising the following steps:
S1 determines the influence factor of grassland degeneration, obtains initial data relevant to influence factor；
S2 handles initial data relevant to influence factor, to obtain normalized number evidence；
S3, normalized number is trained in bp neural network model in advance according to being directed into, export Assessment for classification result；
S4 carries out translation transcoding to Assessment for classification result, is allowed to be exported in friendly way.
As a further solution of the present invention: in step S1, the influence factor of grassland degeneration includes convex spot ground cover degree, advantage
Grass seeds ratio, degeneration indicator species ratio, edible forage ratio and plague of rats situation.
As further scheme of the invention: in step S2, the acquisition methods of the normalized number evidence include following step
It is rapid:
(1) dimensionality reduction cleans initial data, to reject the dirty data in initial data；
(2) calculating merging is carried out for step (1) the data obtained, and is saved；
(3) preservation result is screened and is cleared up again, result is saved；
(4) it is formatted to obtained by step (3) using minimax normalizing, and final result is saved as can
The format identified by bp neural network model.
As further scheme of the invention: the minimax normalizing isIn formula,
V' indicates the data after standardization, and v indicates the data before standardization, minaAnd maxaIt respectively indicates before this specification of attributeization most
Small value and maximum value.
As the present invention further scheme: the bp neural network model includes input layer, hidden layer and output layer,
Wherein, the quantity of hidden layer is two, and the number of nodes of first hidden layer is five, and second node in hidden layer is ten.
As further scheme of the invention: the number of nodes of the input layer and output layer is five, and output layer
Five nodes respectively correspond five grades of grassland degeneration.
Compared with prior art, the beneficial effects of the present invention are: with convex spot cover degree, advantage grass seeds ratio, degeneration instruction
Kind ratio, edible forage ratio and the plague of rats situation 5 main factors for influencing meadows are point of penetration, according to collected multiple numbers
According to the research etc. of grassland degeneration can be widely applied to based on bp neural fusion to the automatic classification of grassland degeneration
Aspect.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of grassland degeneration stage division based on bp neural network.
Fig. 2 is a kind of structural representation of bp neural network model in grassland degeneration stage division based on bp neural network
Figure.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all implementations consistent with this disclosure.On the contrary, they be only with it is such as appended
The example of device and method being described in detail in claims, some aspects disclosed in the present embodiment are consistent.
Embodiment 1
Please refer to Fig. 1~2, in the embodiment of the present invention, a kind of grassland degeneration stage division based on bp neural network, including
Following steps:
S1, determines the influence factor of grassland degeneration, obtains initial data relevant to influence factor, herein, will affect because
Element with being determined as convex spot cover degree, advantage grass seeds ratio, degeneration indicator species ratio, edible forage ratio and plague of rats situation；
S2 handles initial data relevant to influence factor, to obtain normalized number evidence, in practical applications,
There are the data of many mistakes and the data of vacancy in obtained raw data table, therefore, it is necessary to clear up initial data
Processing, to obtain the valuable data that can be used；
S3, normalized number is trained in bp neural network model in advance according to being directed into, export Assessment for classification result；
S4 carries out translation transcoding to Assessment for classification result, is allowed to be exported in friendly way, specifically, friendly
Mode, which refers to, to be identified by user, or user is facilitated to understand.
Embodiment 2
Please refer to Fig. 1~2, in the embodiment of the present invention, a kind of grassland degeneration stage division based on bp neural network, including
Following steps:
S1, determines the influence factor of grassland degeneration, obtains initial data relevant to influence factor, herein, will affect because
Element with being determined as convex spot cover degree, advantage grass seeds ratio, degeneration indicator species ratio, edible forage ratio and plague of rats situation；
S2 handles initial data relevant to influence factor, to obtain normalized number evidence, in practical applications,
There are the data of many mistakes and the data of vacancy in obtained raw data table, therefore, it is necessary to clear up initial data
Processing, to obtain the valuable data that can be used；
For practical situations, initial data such as table 1
Table 1, raw data table
In table, Tbdgd with representing convex spot cover degree is denoted as E (%), E=(a+b+c+ ...+n) × 100/25, wherein a, b,
For bald spot the sum of length, unit are m to c ... n, and 25 be lining rope total length, and unit is also m；
Ksmcl represents edible forage ratio, is denoted as A (%), A=Wy/Ws × 100, wherein Wy is edible forage in sample prescription
Fresh weight, Ws are the summation of aboveground phylomass in sample prescription；
Thzsczbl represents instruction grass seeds ratio of degenerating, and is denoted as D (%), D=CT/CS × 100, wherein CT is all moves back
Change the cover degree of indicator species, CS is the total cover-degree of sample prescription；
Tryjzhl represents soil organic matter content, refers to and takes soil, sample cloth bag using earth boring auger near each measurement sample prescription
Dress soil, each layer of 10cm take three layers altogether, are repeated 2 times composition aggregate sample, measure；Above-mentioned earth boring auger primary repetition therein is used
Soil moisture sampling is done, using aluminium box, soil to be measured dress aluminium box half can (Bao Shidan, 2000).Pedotheque is taken back into reality
After testing room, chooses gravel and 10g or so is taken to measure its water content after mixing.Be made after remaining soil sample air-dry lmm and
The sample of 0.1mm sub-sieve is stand-by.Wherein 0.1mm sample is used for the measurement of available nutrient and pH value, 0.1mm sample for organic matter and
The measurement of Total Nutrient.Soil basis determining: it uses oven drying method (105 DEG C)；PH value measurement: pH meter is used；Earth organic matter is surveyed
It is fixed: to use sulfuric acid potassium dichromate method；Soil Available nitrogen determination: the alkaline hydrolysis way of distillation is used；Soil total N measurement: semimicro kjeldahl determination is used
Method；Soil total P measurement: the anti-development process of molybdenum antimony is used；Soil rapidly available P measurement: 0.5mol.Ll NaHCO3 method is used；
Shqk represents plague of rats situation, is denoted as F, is to exchange all data (population density, hazard area, the effective hole found
Mouthful number, coefficient of opening one's mouth) be standardized after, be averaged and (be not weighted and averaged).F=(Pd+Aj+Bu+Bc)/4, wherein Pd
(population density) is population density, and Aj (jeopardize area) is hazard area, Bu (burrows in
It uses) is effective hole number, Bc (burrows coefficient) is coefficient of opening one's mouth.
The initial data of table 1 is cleaned, to reject the dirty data in initial data, due in these initial data
Other than the data of five influence factors, there are also the kind subclasses such as many other data, such as grass family, sand-grass section, broad-leaved type
Not, seed sum；Rainfall, Ochotona curzoniae, zokor, locust, geographical location, level land, hillside, sunny side, the back etc. are a lot of other
Attribute, so needing to clean these data；
Then calculating merging is carried out to above-mentioned the data obtained again, and is saved；
It is screened and is cleared up again to result is saved, result is saved；
It is formatted using minimax normalizing, and final result is saved as can be by bp neural network model
The format of identification.
Specifically, the minimax normalizing isIn formula, v' indicates the number after standardization
According to v indicates the data before standardization, minaAnd maxaMinimum value and maximum value before respectively indicating this specification of attributeization.
Under normal circumstances, the range of initial data, such as tbdgd belong to [0,100]；Ksmcbl belongs to [0,100]；
Thzszbl belongs to [0,100]；Yjzhl belong to (0,25]；Shqk belong to (0,1].In order to allow all data to fall in [0,1] range
Within, to carry out the standardization of above formula.By minimax normalizing treated data such as table 2.
Table 2, the data after standardization
| tbdgd | ksmcbl | thzszbl | tryjzhl | shqk | class |
| 0.816327 | 0.080808 | 0.785714 | 0.048148 | 0.899194 | class5 |
| 0.826531 | 0.080808 | 0.959184 | 0.12963 | 0.989919 | class5 |
| … | … | … | … | … | … |
| 0.642857 | 0.313131 | 0.734694 | 0.211111 | 0.788306 | class4 |
| 0.642857 | 0.313131 | 0.734694 | 0.211111 | 0.465726 | class4 |
| … | … | … | … | … | … |
| 0.276531 | 0.444444 | 0.358163 | 0.296296 | 0.304435 | class3 |
| 0.285714 | 0.444444 | 0.387755 | 0.237037 | 0.385081 | class3 |
| … | … | … | … | … | … |
| 0.119388 | 0.636364 | 0.244898 | 0.311111 | 0.102823 | class2 |
| 0.119388 | 0.636364 | 0.142857 | 0.311111 | 0.102823 | class2 |
| … | … | … | … | … | … |
| 0.039796 | 0.787879 | 0.041837 | 0.459259 | 0.056452 | class1 |
| 0.039796 | 0.787879 | 0.041837 | 0.648148 | 0.079637 | class1 |
| … | … | … | … | … | … |
S3, normalized number is trained in bp neural network model in advance according to being directed into, export Assessment for classification result；
Specifically, it using five influence factors of grassland degeneration evaluation as the input of neural network, i.e., covers to convex spot
Spend (tbdgd), advantage grass seeds ratio (ysczl), degeneration indicator species ratio (thzszbl), edible forage ratio (ksmcbl) and
Plague of rats situation (shqk).In general, bp neural network model has input layer, hidden layer and output layer, herein, input
Five i.e. above-mentioned influence factor of layer, corresponding, the number of nodes of input layer is five.
It so for hidden layer, proves through a large number of experiments, when the quantity of hidden layer is two, first implicit
The node of layer is 5, and when the node of second hidden layer is 10, the bp neural network model accuracy rate of this structure is higher,
It can achieve 99.12%.
Herein, the number of nodes for defining output layer is 5, respectively corresponds 5 grades of grassland degeneration grading, definition:
(1 000 0)-do not degenerate；Corresponding label Class1；
(0 100 0)-slightly degenerate；Corresponding label Class2；
(0 010 0)-gently degraded；Corresponding label Class3；
(0 001 0)-heavy-degraded；Corresponding label Class4
(0 000 0)-extreme degradation；Corresponding label Class5.
Such as after input (0.039796,0.787879,0.041837,0.459259,0.056452) this data,
Result (10000) can be directly exported by bp neural network model.
S4 carries out translation transcoding to Assessment for classification result, is allowed to be exported in friendly way, specifically, friendly
Mode, which refers to, to be identified by user, or user is facilitated to understand, such as is input to bp neural network mould in data
After type, " slight to degenerate " is as a result directly output as, then this mode can directly be understood by user, it is more convenient.
It should be strongly noted that in the technical program, with convex spot cover degree, advantage grass seeds ratio, degeneration indicator species ratio
The main factors for influencing meadows of example, edible forage ratio and plague of rats situation 5 are point of penetration, according to collected multiple data,
Based on bp neural fusion to the automatic classification of grassland degeneration, model training effect is fairly obvious, model training accuracy
Reach 99.12%, test accuracy rate is also that can be widely applied to research of grassland degeneration etc. 95% or more.
Those skilled in the art will readily occur to other realities of the disclosure after considering the disclosure at specification and embodiment
Apply scheme.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or suitable
The variation of answering property follows the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure or used
Use technological means.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are referred to by claim
Out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by the accompanying claims.
Claims (6)
1. a kind of grassland degeneration stage division based on bp neural network, which comprises the following steps:
S1 determines the influence factor of grassland degeneration, obtains initial data relevant to influence factor；
S2 handles initial data relevant to influence factor, to obtain normalized number evidence；
S3, normalized number is trained in bp neural network model in advance according to being directed into, export Assessment for classification result；
S4 carries out translation transcoding to Assessment for classification result, is allowed to be exported in friendly way.
2. a kind of grassland degeneration stage division based on bp neural network according to claim 1, which is characterized in that step
In S1, the influence factor of grassland degeneration includes convex spot ground cover degree, advantage grass seeds ratio, degeneration indicator species ratio, edible forage ratio
Example and plague of rats situation.
3. a kind of grassland degeneration stage division based on bp neural network according to claim 1, which is characterized in that step
In S2, the acquisition methods of the normalized number evidence the following steps are included:
(1) dimensionality reduction cleans initial data, to reject the dirty data in initial data；
(2) calculating merging is carried out for step (1) the data obtained, and is saved；
(3) preservation result is screened and is cleared up again, result is saved；
(4) it is formatted to obtained by step (3) using minimax normalizing, and final result is saved as can be by bp
The format of neural network model identification.
4. a kind of grassland degeneration stage division based on bp neural network according to claim 3, which is characterized in that described
Minimax normalizing isIn formula, before v' indicates that the data after standardization, v indicate standardization
Data, minaAnd maxaMinimum value and maximum value before respectively indicating this specification of attributeization.
5. a kind of grassland degeneration stage division based on bp neural network according to claim 2, which is characterized in that described
Bp neural network model includes input layer, hidden layer and output layer, wherein the quantity of hidden layer is two, first hidden layer
Number of nodes be five, second node in hidden layer is ten.
6. a kind of grassland degeneration stage division based on bp neural network according to claim 5, which is characterized in that described
The number of nodes of input layer and output layer is five, and five nodes of output layer respectively correspond five grades of grassland degeneration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910526608.0A CN110245867A (en) | 2019-06-18 | 2019-06-18 | A kind of grassland degeneration stage division based on bp neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910526608.0A CN110245867A (en) | 2019-06-18 | 2019-06-18 | A kind of grassland degeneration stage division based on bp neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110245867A true CN110245867A (en) | 2019-09-17 |
Family
ID=67887711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910526608.0A Pending CN110245867A (en) | 2019-06-18 | 2019-06-18 | A kind of grassland degeneration stage division based on bp neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110245867A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110906986A (en) * | 2019-12-11 | 2020-03-24 | 内蒙古工业大学 | Grassland ecological monitoring system and method based on Internet of things |
| CN117522956A (en) * | 2023-11-20 | 2024-02-06 | 青海大学 | Method for calculating degradation indication grass seed coverage by using unmanned aerial vehicle image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323974A (en) * | 2011-06-08 | 2012-01-18 | 北京师范大学 | Method for evaluating degradation degree of alpine grassland based on visual vegetation indexes |
| CN105894119A (en) * | 2016-04-05 | 2016-08-24 | 成都寻道科技有限公司 | Student ranking prediction method based on campus data |
| CN108595381A (en) * | 2018-04-27 | 2018-09-28 | 厦门尚为科技股份有限公司 | Health status evaluation method, device and readable storage medium storing program for executing |
| CN109166627A (en) * | 2018-07-26 | 2019-01-08 | 苏州中科先进技术研究院有限公司 | A kind of health evaluating method, assessment device and the system for rehabilitation |
| CN109829650A (en) * | 2019-01-31 | 2019-05-31 | 中国农业科学院农业资源与农业区划研究所 | Assess Meadow difference degree of degeneration model and its method for building up and application |
| CN109845589A (en) * | 2019-03-18 | 2019-06-07 | 兰州大学 | A kind of stage division of dogstail/Trifolium repense Grassland degradation degree and application |
-
2019
- 2019-06-18 CN CN201910526608.0A patent/CN110245867A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323974A (en) * | 2011-06-08 | 2012-01-18 | 北京师范大学 | Method for evaluating degradation degree of alpine grassland based on visual vegetation indexes |
| CN105894119A (en) * | 2016-04-05 | 2016-08-24 | 成都寻道科技有限公司 | Student ranking prediction method based on campus data |
| CN108595381A (en) * | 2018-04-27 | 2018-09-28 | 厦门尚为科技股份有限公司 | Health status evaluation method, device and readable storage medium storing program for executing |
| CN109166627A (en) * | 2018-07-26 | 2019-01-08 | 苏州中科先进技术研究院有限公司 | A kind of health evaluating method, assessment device and the system for rehabilitation |
| CN109829650A (en) * | 2019-01-31 | 2019-05-31 | 中国农业科学院农业资源与农业区划研究所 | Assess Meadow difference degree of degeneration model and its method for building up and application |
| CN109845589A (en) * | 2019-03-18 | 2019-06-07 | 兰州大学 | A kind of stage division of dogstail/Trifolium repense Grassland degradation degree and application |
Non-Patent Citations (1)
| Title |
|---|
| 李春梅等: ""基于神经网络的草地退化决策专家系统设计"", 《信息与电脑（理论版）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110906986A (en) * | 2019-12-11 | 2020-03-24 | 内蒙古工业大学 | Grassland ecological monitoring system and method based on Internet of things |
| CN117522956A (en) * | 2023-11-20 | 2024-02-06 | 青海大学 | Method for calculating degradation indication grass seed coverage by using unmanned aerial vehicle image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| He et al. | Soil organic carbon prediction using phenological parameters and remote sensing variables generated from Sentinel-2 images | |
| Waheed et al. | Measuring performance in precision agriculture: CART—A decision tree approach | |
| CN111598045B (en) | Remote sensing farmland change detection method based on object spectrum and mixed spectrum | |
| Duadze | Land use and land cover study of the savannah ecosystem in the Upper West Region (Ghana) using remote sensing | |
| Han et al. | Fuzzy clustering of maize plant-height patterns using time series of UAV remote-sensing images and variety traits | |
| Vintrou et al. | Mapping fragmented agricultural systems in the Sudano-Sahelian environments of Africa using random forest and ensemble metrics of coarse resolution MODIS imagery | |
| Ma et al. | Cotton yield estimation based on vegetation indices and texture features derived from RGB image | |
| Dong et al. | Using RapidEye imagery to identify within-field variability of crop growth and yield in Ontario, Canada | |
| McInnes et al. | Discriminating native and nonnative grasses in the dry mixedgrass prairie with MODIS NDVI time series | |
| CN110245867A (en) | A kind of grassland degeneration stage division based on bp neural network | |
| Xia et al. | Phenology-based decision tree classification of rice-crayfish fields from Sentinel-2 imagery in Qianjiang, China | |
| CN113570273B (en) | Spatialization method and system for irrigation farmland statistical data | |
| Malone et al. | Linking soil health indicators to management history and soybean yield | |
| DuPont et al. | Soil health indicators for Central Washington orchards | |
| Plaza et al. | Classification of airborne multispectral imagery to quantify common vole impacts on an agricultural field | |
| Brewer et al. | Remote sensing of invasive alien wattle using image texture ratios in the low-lying Midlands of KwaZulu-Natal, South Africa | |
| Xie et al. | Estimating stomatal conductance of citrus under water stress based on multispectral imagery and machine learning methods | |
| Sun et al. | Potential of multi-seasonal vegetation indices to predict rice yield from UAV multispectral observations | |
| 김예화 | Classification of Deforested and Degraded Areas in North Korea Using Phenology-based Indices and Spatiotemporal Data Fusion Method | |
| CN117309857A (en) | Wheat uniformity monitoring method | |
| Riembauer et al. | D5. 4 Mapping of vegetation indices and metrics, and their utility in FSA mapping at CS scale | |
| Gao et al. | Comprehensive degradation index for monitoring desert grassland using UAV multispectral imagery | |
| Pierre et al. | Unravelling the heterogeneity of farms irrigation practices on Mediterranean perennial agricultural systems for the optimization of water resource management | |
| Shuai | Sub-Field Crop Yield and Yield Stability Analysis with Multi-Source Datasets and Advanced Crop Simulation Model | |
| McCarthy et al. | Future farming: Machine vision for nitrogen assessment in grain crops |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190917 |
|
| RJ01 | Rejection of invention patent application after publication |