CN110008542A - A kind of city magnanimity building classifications method - Google Patents
A kind of city magnanimity building classifications method Download PDFInfo
- Publication number
- CN110008542A CN110008542A CN201910215502.9A CN201910215502A CN110008542A CN 110008542 A CN110008542 A CN 110008542A CN 201910215502 A CN201910215502 A CN 201910215502A CN 110008542 A CN110008542 A CN 110008542A
- Authority
- CN
- China
- Prior art keywords
- building
- city
- magnanimity
- type
- classifier
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000006870 function Effects 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 7
- 238000012360 testing method Methods 0.000 claims abstract description 6
- 238000011478 gradient descent method Methods 0.000 claims description 4
- 238000013178 mathematical model Methods 0.000 claims description 4
- 241000208340 Araliaceae Species 0.000 claims description 3
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 3
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 3
- 235000008434 ginseng Nutrition 0.000 claims description 3
- 238000010998 test method Methods 0.000 claims description 2
- 238000004088 simulation Methods 0.000 abstract description 14
- 238000005265 energy consumption Methods 0.000 abstract description 13
- 235000013305 food Nutrition 0.000 description 8
- 238000011160 research Methods 0.000 description 8
- 230000006872 improvement Effects 0.000 description 6
- 239000004566 building material Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 239000008186 active pharmaceutical agent Substances 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000005431 greenhouse gas Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000009437 non-residential construction Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009435 building construction Methods 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000013316 zoning 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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Civil Engineering (AREA)
- Architecture (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of city magnanimity building classifications methods.Building base plane figure and POI are obtained using public data source, and the two is subjected to space connection in Arcmap.Choose a certain number of training samples, using logistic gram regression model, establish with the consistent classifier of target classification quantity, by python Programming with Pascal Language, iteration gets parms weight.Then, a certain number of verifying samples are chosen, parameters weighting is substituted into each classifier and finds out numerical value, choose type of prediction of the type as the building corresponding to the maximum classifier of numerical value.Type of prediction is made comparisons with actual type, when predictablity rate >=90%, precision test can be passed through.Finally, utilizing " field calculator " function in Arcmap, automatic assignment is programmed, determines urban architecture type.City magnanimity building classifications method provided by the invention can serve city simulation of energy consumption field, be particularly suitable for " architecture archetype method ".
Description
Technical field
The present invention relates to city simulation of energy consumption fields, more particularly to a kind of city magnanimity building classifications method.
Background technique
In order to cope with the energy shortage faced at present and environmental degradation problem, multinational government establishes energy-saving and emission-reduction mesh successively
Mark this macro-goal.American plan reduced the greenhouse gas emission of 26-28% before 2025, and European Union planned before the year two thousand thirty
Reduce by 40% greenhouse gas emission.China also points out in " country reply climate change planning (2014-2020) ", by
Than the 2005 decline 40-45% of realization unit country production of units total value discharge capacity of carbon dioxide in 2020.In order to reach these mesh
Mark, there is an urgent need to a kind of suitable methods to carry out the actually required energy consumption in simcity by city officials, make conjunction on this basis
The production of energy and application plan of reason, to achieve energy-saving and emission reduction purposes.
In recent years, city simulation of energy consumption method --- the city energy consumption model method of emerging a kind of " from bottom to top ".City
The foundation of energy consumption model then assigns it firstly the need of city three-dimensional building model is established including building enclosure thermal parameter,
Personnel activity's table, equipment table, the information including Weather information etc., finally by entire model import in simulation of energy consumption software into
Row city simulation of energy consumption.City simulation of energy consumption is different from energy simulation, is most significantly characterized in that the urban architecture scale of construction
It is huge, if being unpractical according to the actual parameter assignment modeling one by one of each building building.In order to solve this technology hardly possible
Point, a kind of novel method is " architecture archetype method " in the world.Due to same type of building, such as market, hospital, supermarket,
Its code of building design is close, and personnel activity and equipment moving law are similar, when merging into a kind of building processing, can drop significantly
The workload of low modeling saves operation time and reduces simulated cost.
When with " architecture archetype method ", E.Mata from Sweden et al. (2014) by urban architecture be divided into house with it is non-
Two major class of house, are then subdivided into bachelor apartment and family's apartment two major classes for residential housing, and non-residential construction is segmented
For office building, commercial building, medical building, school building and style rest architecture.The scholar F.Lin (2017) of TaiWan, China
By studying the housing type in Taiwan, house is divided into united villa, multi-storey building and high residential building.The researcher in the U.S.
W.Li et al. (2018) again segments non-residential construction, by office building be divided into larger office, medium-sized office with it is small-sized
Office, is divided into dining room, supermarket, market for commercial building, school is divided into elementary and middle school.In fact, for entire city energy
For consumption simulation, the classification of magnanimity building is finer, it is meant that can establish further types of " architecture archetype ", this is for being promoted
Simulation precision is highly beneficial.
But at use " architecture archetype method ", researcher would generally face following problem:
1) it can be modeled and be classified according to public data；
2) how quickly to classify to city magnanimity building, and guarantee higher accuracy；
City simulation of energy consumption be different from energy simulation another feature is that, the complete foundation nothing of urban architecture model
Method models completion by a sheet by a sheet construction drawing one by one.Its reason is both to take long time, and city can not be obtained by lying also in
In each building building construction drawing.In recent years, as the Urban Data of some research institutions is open and large scale business map
The surveying and mapping data of company is open, this problem is constantly solved.The present invention is based on this basis of open data and carries out
's.In addition, most of researchs are based on the data in existing database, such as the U.S.'s labor human relations when to city magnanimity building classifications
The CityBES platform of this Berkeley National Laboratory research and development.After the type of one building is determined, store into database, with
Asd number is continuously increased, and the building classifications in entire city are also just natural to be generated.The advantage of this method is it
Not only magnanimity building in city is classified, while also storing a large amount of architecture information.But this method implementation cycle compared with
It is long, as soon as and when the city that researcher needs to study a not Relational database can be time-consuming, the method can not be used for reference.Therefore,
The present invention, with tool, is developed a kind of more quick city magnanimity building classifications method, made it have based on public data
Better universality.
Summary of the invention
The problem to be solved in the present invention is: the existing city magnanimity building classifications method based on database data implements week
Phase is long, and universality is bad, can not be generalized in wider city, need it is a kind of based on public data it is more quick, accurate,
Pervasive city magnanimity building classifications method, provides technical support for city simulation of energy consumption research, for this purpose, the present invention mentions
For a kind of city magnanimity building classifications method, method includes the following steps:
Step 1: determining survey region, and obtains and build base plane figure in survey region；
Step 2: magnanimity urban architecture classification demand is determined；
Step 3: POI data is obtained；
Step 4: data connection；
Step 5: it chooses " training sample ", determines parameters weighting；
Step 6: parameters weighting precision test；
Step 7: city magnanimity building classifications.
Further improvement of the present invention, the building base plane figure in the step 1 can be obtained by 3 kinds of modes: benefit
It with public data, is directly downloaded using OpenStreetMap, or is based on Baidu map or Amap, use freeware
MapCapturer0.5Beta downloading.
Further improvement of the present invention, the building classifications in the step 2 include level-one class and second level class.
Further improvement of the present invention, the POI data in the step 3 pass through free application Baidu developer's key, fortune
It is obtained with python Programming with Pascal Language.
Further improvement of the present invention, the parameters weighting in the step 5 determine that method logic-based Wamsteeker returns mould
Type, specifically: reference sigmoid function seeks parameters weighting using the iteration based on gradient descent method as mathematical model, and
Building classifications quantity is obtained according to needed for, establishes the classifier of respective numbers.
Further improvement of the present invention, the parameters weighting precision test method in the step 6 are more each classifier
Numerical values recited chooses type of prediction of the type as the building corresponding to the maximum classifier of numerical value, by type of prediction and is somebody's turn to do
The actual type of building is made comparisons, when verify sample predictablity rate >=90% when, pass through precision test.
Further improvement of the present invention, the city magnanimity building classifications in the step 7 utilize " field in Arcmap
Calculator " function programs automatic assignment
The invention has the following advantages that
1) lower present invention is implemented as, and there is good transplantability.The present invention is based on building base plane figure with
POI data can freely be obtained by open database, therefore can be widely transplanted to numerous cities and used；
2) present invention can save a large amount of time and manpower.The present invention does not need to accumulate data for a long time, establishes complete
Database；And a small amount of training sample need to be only chosen, the classification work of magnanimity building can be completed；
3) present invention has good autonomous selectivity.Researcher in using the present invention, without considering all classification
Situation only need to establish classification according to the demand of oneself.Such as: researcher need to study house, business, when office building, only need
3 classifiers are established, can not consider medical treatment, the building types such as education；
4) present invention has good anti-interference.Researcher in using the present invention, since parameters weighting indicates the ginseng
Several influence degree sizes, for weak relevant parameter, the influence that can weaken these parameters by obtained smaller weight is made
With.It is therefore not necessary to worry that parameter chooses excessive problem.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is the building base plane figure (by taking the center of Nanjing as an example) obtained: (a) utilizing public data；(b) it utilizes
OpenStreetMap；(c) it is based on Baidu map；
Fig. 3 is magnanimity building classifications stratal diagram；
Fig. 4 is the connection figure for building base plane and 5 seed type POI (by taking the center of Nanjing as an example)；
Fig. 5 is the quantity (partial value) that base plane figure " attribute list "-includes all kinds of POI；
Fig. 6 be logistic gram regression model based on sigmoid functional picture.
Specific embodiment
Present invention is further described in detail with specific embodiment with reference to the accompanying drawing:
The invention discloses a kind of city magnanimity building classifications method, the city magnanimity building classifications method provided can be taken
It is engaged in being particularly suitable for " architecture archetype method " in city simulation of energy consumption field.
A kind of city magnanimity building classifications method of the invention, comprising the following steps:
Step 1: determining survey region, and obtains and build base plane figure in survey region；
In city energy consumption research field, researchers often need the range for first defining research, and some is divided with administrative area,
Such as: Sectors of Gulou Dis trict, Nanjing, Xuanwu District；Some is with function zoning, such as: Nanjing midtown, north of the Changjiang River new district；Area according to research
Domain range, then the building base plane figure in the region is obtained, methods availalbe is as follows:
1, using public data, such as Fig. 2 (a)；
2, it is freely downloaded using OpenStreerMap, such as Fig. 2 (b)；
3, it is based on Baidu map or Amap, is downloaded using freeware MapCapturer0.5Beta, such as Fig. 2 (c).
Step 2: magnanimity urban architecture classification demand is determined；
Different researchers is when carrying out city simulation of energy consumption with " architecture archetype method ", for the demand of building classifications
It is different.Urban architecture need to be only classified as house by some scholars, business, medical treatment, education etc., and some then needs more finely
Classification, such as again school building is divided into primary school, middle school, university etc..City magnanimity building classifications precision needed for determining research,
Can be according to city magnanimity building classifications stratal diagram, the detail comprising level-one class Yu second level class, as shown in Figure 3.
Step 3: POI data is obtained；
POI is the abbreviation of point of interest (Points of Interest).Physical model in city, such as: retail shop, teaching
Building, Outpatient Building etc. are conceptualized as a point, form POI, and it includes a series of satellite informations, such as: title, build age, the number of plies
Deng；POI of the present invention is obtained by Baidu API, by free application developer's key, with python Programming with Pascal Language, according to step
The all types of POI in survey region are downloaded in building classifications stratal diagram in rapid two, input " specific name ".
Step 4: data connection；
The POI data obtained in the building base plane figure downloaded in step 1 and step 3 is imported in software Arcmap
(as shown in Figure 4) utilizes " space connection " to operate, all types of POI being located on plan view is linked in the plane, leads to
Cross the quantity (as shown in Figure 5) for checking that " attribute list " of plan view obtains all types of POI.
Step 5: it chooses " training sample ", determines parameters weighting；
The new building base plane figure (comprising all kinds of POI quantity) obtained in a certain number of step 4 is chosen as training
Sample.The i class POI that will acquire is as parameter xi, need the building in survey region being divided into j class, and set each parameters weighting
For aij, obtain building determined type yjWith parameter xiRelationship are as follows:
yj=∑ aij·xi (1)
The present invention needs i >=j, i.e., for determining that the number of parameters of building type should be no less than building type number.
The present invention is based on mathematical model be logistic gram regression model (logistic regression model, letter
Number is as shown in Figure 6), therefore the y acquired by formula (1)jBetween [0,1], threshold value k ∈ (0,1), ordinary circumstance k are reset
0.5 is taken, is obtained:
Wherein, work as yjWhen=1, it is correct to indicate that the building type determines；Work as yjWhen=0, it is wrong to indicate that the building type determines
Accidentally.When determining jth class building, in " training sample ", such is enabled to build yj=1, remaining type builds yj=0.With such
It pushes away, constructs j with the classifier determined.It with machine learning thought, is programmed by python, using based on gradient descent method
Iteration obtain the parameters weighting a of each classifierij。
Step 6: parameters weighting precision test；
The new building base plane figure (comprising all kinds of POI quantity) obtained in a certain number of step 4 is chosen as verifying
Sample.The parameters weighting a obtained by step 5ijAnd the parameter x of itselfi, substitute into formula (1), acquired respectively in j classifier
Under yj, choose the maximum y of numerical valuejType of prediction of the corresponding type as the building.By the reality of type of prediction and the building
Type is made comparisons, when verify sample predictablity rate >=90% when, enter step seven；Otherwise, return step five.
Step 7: city magnanimity building classifications.
The new building base plane figure (including all kinds of POI quantity) obtained according to step 4, obtains owning in survey region
The parameter x that building includesi, in conjunction with the parameters weighting a of step 5 acquisitionij, with formula (1), acquired under j classifier respectively
yj, choose the maximum y of numerical valuejFinal type of the corresponding type as the building.
In order to keep the above method more intuitive and it can be readily appreciated that choose Nanjing as survey region, in commercial building
Market, supermarket, household, electric appliance chooses big food and drink as class object, small food and drink, market, supermarket, building materials, and household electrical appliances are small-sized
Household, large-scale 8 class POI of household carry out a kind of demonstration of magnanimity urban architecture classification method of the present invention as parameter.The demonstrator
Certain typicalness: firstly, Nanjing is the research range of a large scale, asd number is extremely more；Secondly, in order to embody this
The good autonomous selectivity of invention only chooses 4 kinds of second level classes building in the commercial building of level-one class, without considering dining room etc.；Most
Afterwards, in order to prove good anti-interference of the invention, weak relevant food and drink class POI is chosen as parameter.In the following, Fig. 1 will be combined
Invention flow chart be specifically addressed:
1) it obtains Nanjing and builds base plane figure.Using freeware MapCapturer0.5Beta, data source is selected
For Baidu map, map is downloaded with the rectangle tool, format map is .png at this time.Picture is imported in software Arcmap, into
Row is cut rectifies a deviation with coordinate, and coordinate system is WGS 1984.Then, " binaryzation " processing is carried out to picture, finally used
" Arcscan " tool carries out Automatic Vector to picture.At this point, can be obtained the Nanjing vector quantization that format is .shp builds base
Bottom plan view；
2) 8 class POI are obtained.Firstly, obtaining key in Baidu map developer mode application API；Pass through python language
Programming, sets rectangular extent, and the information (title, warp, latitude) that setting POI includes sets search key (big food and drink, small meal
Drink, market, supermarket, building materials, household electrical appliances, small-sized household, large-scale household), 8 excel tables are obtained, each table is a type
The POI of type summarizes；
3) POI data is connect with building base plane figure.In software Arcmap, turn table tool, link 8 with excel
A excel table.By " display xy data " tool, using warp, latitude as medium, in dots by the text information in excel
It is shown on map.Finally, the POI of 8 seed types is linked on building base plane figure by " space connection " tool.It is logical
" opening attribute list " option is crossed, checks all kinds of POI quantity that each plane includes；
4) sample is chosen, the weight of parameter is sought.41 groups of samples are chosen, the quantity of 8 class POI in every group of sample is summarized, are established
Decision function (classifier):
Y=a0·x0+a1·x1+a2·x2+a3·x3+a4·x4+a5·x5+a6·x6+a7·x7+a8·x8 (3)
Wherein, x1~x8Respectively indicate big food and drink, small food and drink, market, supermarket, building materials, household electrical appliances, small-sized household, large-scale household
8 class POI, a1~a8Respectively indicate the weight of this 8 class POI；a0For constant term, x0For supplementary variable, perseverance is 1.
Target need to determine four kinds of building types, it is therefore desirable to establish 4 classifiers.Firstly, establishing for determining market
Classifier y1:
In 41 groups of samples, enabling actual type is the y of the sample in market1=1；Remaining type, uniformly enables y1=0.In this way,
Complete the foundation of a classifier.Same method is established for determining supermarket, household, the classifier y of electric appliance2, y3, y4。
Machine learning is established by python Programming with Pascal Language, the mathematical model of reference is sigmoid function, seeks parameters weighting
Mathematical method be the iteration based on gradient descent method.Finally, the parameters weighting for obtaining four classifiers is as shown in table 1:
1. classifier parameters weight summary sheet of table
| Constant | Big food and drink | Small food and drink | Market | Supermarket | Building materials | Household electrical appliances | Knickknack | Big furniture | |
| Classifier 1 | -2.105997 | 0.590599 | -0.12317 | 1.850370 | -1.00772 | -0.46784 | -0.53100 | 0.377198 | 0.154967 |
| Classifier 2 | 0.3380779 | -0.28198 | 0.063532 | -1.38816 | 2.361930 | -0.76569 | -0.55391 | 0.454481 | -0.22018 |
| Classifier 3 | -0.832122 | -1.24665 | -1.23989 | 0.852050 | -0.86991 | 0.596020 | -1.02001 | 0.604870 | 0.991149 |
| Classifier 4 | -0.492438 | -0.18268 | -0.21103 | 0.330833 | -1.91493 | -0.85374 | 1.569184 | 0.546645 | -0.40039 |
5) parameters weighting is verified.12 groups of samples are chosen, the quantity of 8 class POI in every group of sample is summarized, according to table 1 and formula (4)
Y1, y2, y3, the y4 of each building are found out, and is made comparisons, chooses the corresponding type of the maximum yj of numerical value as the final of the building
Type, the results are shown in Table 2:
2. sample verification result table of table
| ID | Classifier 1 | Classifier 2 | Classifier 3 | Classifier 4 | Prediction | It is practical |
| 1 | 7.98273490e-01 | 2.33084825e-22 | 1.81629849e-48 | 7.86891059e-01 | Market | Market |
| 2 | 9.91858752e-01 | 2.66597623e-07 | 3.45827295e-32 | 9.02711728e-07 | Market | Market |
| 3 | 9.99999994e-01 | 8.01293782e-05 | 7.95626312e-44 | 1.49830421e-07 | Market | Market |
| 4 | 8.81721450e-02 | 3.89236533e-02 | 9.29254503e-10 | 3.70422782e-04 | Market | Market |
| 5 | 9.73213954e-01 | 1.28105226e-02 | 3.58930724e-28 | 4.42632447e-08 | Market | Market |
| 6 | 3.05970235e-02 | 5.35332303e-01 | 1.61818311e-06 | 8.80775072e-02 | Supermarket | Supermarket |
| 7 | 4.50400517e-02 | 8.65776131e-01 | 1.85475495e-02 | 2.64860587e-01 | Supermarket | Supermarket |
| 8 | 5.89976897e-02 | 9.27242859e-01 | 4.37062163e-03 | 4.68821715e-02 | Supermarket | Supermarket |
| 9 | 5.42565302e-03 | 5.10885611e-05 | 1 | 6.24493842e-06 | Household | Household |
| 10 | 5.67723211e-13 | 4.91440977e-40 | 1 | 5.37263580e-49 | Household | Household |
| 11 | 2.71235126e-01 | 2.69286359e-01 | 1.07755996e-03 | 5.78760234e-01 | Electric appliance | Electric appliance |
| 12 | 1.02517201e-01 | 3.93104600e-01 | 1.28855420e-02 | 6.64416722e-01 | Electric appliance | Electric appliance |
6) city magnanimity building classifications.By sampling inspection, nicety of grading=100% > 90% illustrates parameters weighting meter
It calculates rationally, can be generalized to entire survey region.In Arcmap, " the opening attribute list " for opening building base plane figure layer is right
Words, new field " classifier 1 ", " classifier 2 ", " classifier 3 ", " classifier 4 ", " type ".With " field calculator " function
Can, and obtained coefficient matrix, it is field " classifier 1 " that " classifier 2 ", " classifier 3 ", " classifier 4 " programs respectively,
It is automatic to calculate assignment.Finally, each is built " classifier 1 ", " classifier 2 ", " classifier 3 ", the numerical value of " classifier 4 " into
Row compares, and using selection " maximum value " principle, uses " field calculator " function, determines the type of the building, record into field
" type ", to complete the classification of entire city magnanimity building.
The above described is only a preferred embodiment of the present invention, being not the limit for making any other form to the present invention
System, and made any modification or equivalent variations according to the technical essence of the invention, still fall within present invention model claimed
It encloses.
Claims (7)
1. a kind of city magnanimity building classifications method, it is characterised in that: method includes the following steps:
Step 1: determining survey region, and obtains and build base plane figure in survey region；
Step 2: magnanimity urban architecture classification demand is determined；
Step 3: POI data is obtained；
Step 4: data connection；
Step 5: it chooses " training sample ", determines parameters weighting；
Step 6: parameters weighting precision test；
Step 7: city magnanimity building classifications.
2. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: building in the step 1
Building base plane figure can be obtained by 3 kinds of modes: being utilized public data, directly downloaded using OpenStreetMap, or be based on
Baidu map or Amap are downloaded using freeware MapCapturer0.5Beta.
3. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: building in the step 2
It builds classification and includes level-one class and second level class.
4. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: in the step 3
POI data is obtained by free application Baidu developer's key with python Programming with Pascal Language.
5. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: the ginseng in the step 5
Number Weight Determination logic-based Wamsteeker regression model, specifically: reference sigmoid function is used as mathematical model
Iteration based on gradient descent method seeks parameters weighting, and obtains building classifications quantity according to required, establishes point of respective numbers
Class device.
6. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: the ginseng in the step 6
Number weight precision test method is the numerical values recited of more each classifier, chooses type corresponding to the maximum classifier of numerical value and makees
For the type of prediction of the building, type of prediction is made comparisons with the actual type of the building, when the predictablity rate of verifying sample
When >=90%, pass through precision test.
7. a kind of city magnanimity building classifications method according to claim 1, it is characterised in that: the city in the step 7
City's magnanimity building classifications utilize " field calculator " function in Arcmap, program automatic assignment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910215502.9A CN110008542A (en) | 2019-03-21 | 2019-03-21 | A kind of city magnanimity building classifications method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910215502.9A CN110008542A (en) | 2019-03-21 | 2019-03-21 | A kind of city magnanimity building classifications method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110008542A true CN110008542A (en) | 2019-07-12 |
Family
ID=67167551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910215502.9A Pending CN110008542A (en) | 2019-03-21 | 2019-03-21 | A kind of city magnanimity building classifications method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110008542A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990779A (en) * | 2021-04-27 | 2021-06-18 | 上海钐昆网络科技有限公司 | Method, device, equipment and storage medium for scoring candidate address |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764193A (en) * | 2018-06-04 | 2018-11-06 | 北京师范大学 | Merge the city function limited region dividing method of POI and remote sensing image |
-
2019
- 2019-03-21 CN CN201910215502.9A patent/CN110008542A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764193A (en) * | 2018-06-04 | 2018-11-06 | 北京师范大学 | Merge the city function limited region dividing method of POI and remote sensing image |
Non-Patent Citations (2)
| Title |
|---|
| C.C.FONTE等: "Classification of building function using available sources of VGI", 《THE INTERNATIONAL ARCHIVES OF THE PHOTOGRAMMETRY, REMOTE SENSING AND SPATIAL INFORMATION SCIENCES》 * |
| 曲畅等: "POI辅助下的高分辨率遥感影像城市建筑物功能分类研究", 《地球信息科学》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990779A (en) * | 2021-04-27 | 2021-06-18 | 上海钐昆网络科技有限公司 | Method, device, equipment and storage medium for scoring candidate address |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chacón et al. | Multivariate kernel smoothing and its applications | |
| Fischer et al. | Geographic information systems, spatial data analysis and spatial modelling: an introduction | |
| Goodchild et al. | Integrating GIS and spatial data analysis: problems and possibilities | |
| Tomlinson | The impact of the transition from analogue to digital cartographic representation | |
| Hoch et al. | Geolinguistics: The incorporation of geographic information systems and science | |
| CN113297174B (en) | Land utilization change simulation method based on deep learning | |
| CN107506499A (en) | The method, apparatus and server of logical relation are established between point of interest and building | |
| Li et al. | Knowledge transfer and adaptation for land-use simulation with a logistic cellular automaton | |
| CN106547842A (en) | A kind of method that location-based emotion is visualized on virtual earth platform | |
| CN110990639B (en) | Data processing method and device for education informatization horizontal trend analysis | |
| Florax et al. | Comparative environmental economic assessment | |
| Yu et al. | A heuristic approach to the generalization of complex building groups in urban villages | |
| Rossato et al. | Digital tools for documentation and analysis of vernacular cultural heritage in Indian city centers | |
| Yang et al. | Spatial cognitive modeling of the site selection for traditional rural settlements: A case study of Kengzi Village, Southern China | |
| CN110008542A (en) | A kind of city magnanimity building classifications method | |
| Hecht et al. | Crowd-sourced data collection to support automatic classification of building footprint data | |
| Lee et al. | Machine learning based prediction of the value of buildings | |
| CN115689106A (en) | Method, device and equipment for quantitatively identifying regional space structure of complex network view angle | |
| Boukerch et al. | The setting up of a gis for the general population and housing census | |
| Wang | Extending geographic information systems to urban morphological analysis with a space syntax approach | |
| As’Sidiq et al. | Implementation of K-Means Algorithm for Information Technology Freshman Class Division | |
| Zhang et al. | A GIS-Based K-Mean Clustering Algorithm for Characteristic Towns in China | |
| CN116258295A (en) | City sign diagnosis method for city update | |
| Cao et al. | a 3d Building Indoor-Outdoor Benchmark for Semantic Segmentation | |
| Malhotra et al. | Open-Source Tool for Transforming CityGML Levels of Detail. Energies 2021, 14, 8250 |
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: 20190712 |
|
| RJ01 | Rejection of invention patent application after publication |