CN108871286A - The completed region of the city density of population evaluation method and system of space big data collaboration - Google Patents
The completed region of the city density of population evaluation method and system of space big data collaboration Download PDFInfo
- Publication number
- CN108871286A CN108871286A CN201810379934.9A CN201810379934A CN108871286A CN 108871286 A CN108871286 A CN 108871286A CN 201810379934 A CN201810379934 A CN 201810379934A CN 108871286 A CN108871286 A CN 108871286A
- Authority
- CN
- China
- Prior art keywords
- building
- city
- data
- population
- space
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/02—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to the completed region of the city density of population evaluation method and system of a kind of collaboration of space big data, this method obtains building range based on the classification method of object-oriented using high-resolution optical remote sensing data;Using the fine road network space big data in city by space topological algorithm partition city block, and combining interest space of points big data to distinguish city block is residential area and non-residential area;Depth of building is obtained using remote sensing optical stereo imaging data;According to building height and building range and consensus data is combined, constructs population density estimation model using multiple regression analysis method;Using the range of the building and the height of building as input variable, it is input to the population density estimation model, the density of population is calculated, and then distributed regular grid space corresponding to completed region of the city using space statistical analysis method.This method and system can precisely and accurately realize that the completed region of the city density of population is estimated, are more suitable for urban planning, the needs with business decision to fine demographic data of preventing and reducing natural disasters, are conducive to further genralrlization and application.
Description
Technical field
The present invention relates to urban remote sensing and urban demography's technical field, especially a kind of city of space big data collaboration
Built-up areas density of population evaluation method and system.
Background technique
Density of population distribution is the form of expression of Population process spatially, and urban population density distribution and simulation are cities
Important content in space structure and suburbanization research.Domestic and foreign scholars are directed to the feature in " ring layer formula " city, utilize mathematical modulo
Type discloses the universal law of urban population density distribution.These researchs are based on analysis of statistical data density of population spatial distribution,
It is assumed that population continuous uniform in census area is distributed, but actual population distribution and discontinuous.In the world, many mechanisms endeavour
Research and production in density of population spatialization, wherein most influential is that GPW and LandScan population in the world data produce
Product.GPW (v1, v2) data will simply count population be evenly distributed to do not consider on regular grid and the influence of population distribution because
Element, LandScan data by calculate the probability coefficent that highway, the gradient, land cover pattern and night lights influence population distribution come
Simulate the density of population in grid.At home, many scholars have also carried out the whole nation and zonal density of population grid model is ground
Study carefully, it is main to use the land use data, nighttime light data and the consensus data that interpret based on intermediate-resolution remotely-sensed data
Deng in conjunction with the population distribution feature of China's different zones, based on statistics or the physical Model Study people of kilometer grid scale
Mouth density space grid method.Compared with hypothesis population is the statistical data of continuous uniform distribution in region, the density of population
Grid square can reflect the actual distribution situation of population in further detail, lead in population suffered from disaster's assessment, Regional development planning etc.
Domain is widely used.But these density of population grid product space scales are excessively coarse, are not able to satisfy city and build up
The demand of area's population spatial distribution analysis;On the other hand, the people of data and Optimal Parameters production is covered based on Land Use/land
Mouth spatialization product, since the identical land use pattern density of population is identical and the special heterogeneity of completed region of the city Optimal Parameters
Not significant (such as nighttime light data reaches saturation), it can not accurately portray completed region of the city population actual spatial distribution feature.
With the development of remote sensing technology, high spatial resolution remote sensing data acquisition becomes more to be easy, and cheap.It can
The high-definition remote sensing data utilized mainly include the types such as optics, radar and LiDAR, can be with using high-definition remote sensing data
Identification building simultaneously extracts the attribute informations such as the floor area of building, height.It, can be by urban population essence based on building three-dimensional information
Carefully it is assigned to space.In recent years, some scholars have studied the urban population based on high spatial resolution remote sensing data and estimate
Model is calculated, by extracting the geometry and type information of building, the model of estimation population quantity is established, obtains preferable experiment
Effect.But these researchs mainly select the zonule in completed region of the city for trial zone, with residential area or committee member resident
All population capacities can be estimated for unit, be still not enough to finely reflect the spatial distribution state of population.
It can not meticulous depiction completed region of the city population spatial distribution characteristic for current density of population spatialization product
Problem, the present invention is based on high-resolution optical volumetric imaging datas, extract building range, elevation information, are aided with city and build up
The spaces big data such as the fine road network in area and point of interest divides classification to building, and then proposes a kind of based on building geometry
The fine density of population evaluation method and system of attribute and type information, for urban planning, preventing and reducing natural disasters provides with business decision
Reliable demographic data is supported.
Summary of the invention
The present invention covers the population spatial distribution product of data and Optimal Parameters production for tradition based on Land Use/land
The problem of completed region of the city population actual spatial distribution feature can not accurately be portrayed, provides a kind of city of space big data collaboration
Built-up areas density of population evaluation method is mentioned respectively based on high-resolution optical remote sensing data and remote sensing optical stereo imaging data
The building range and height taken, using road network and interest space of points big data, collaboration constructs population density estimation model,
Realize the fine density of population estimation in completed region of the city.The invention further relates to a kind of completed regions of the city of space big data collaboration
Density of population estimating system.
Technical scheme is as follows:
A kind of completed region of the city density of population evaluation method of space big data collaboration, which is characterized in that this method passes through
What satellite obtained obtains building model based on the classification method of object-oriented according to the corresponding optical remote sensing data in completed region of the city
It encloses;And completed region of the city is divided by several planars by space topological algorithm using the fine road network space big data in city
City block, by point of interest (POI) space big data divide city block be residential area and non-residential area;Recycle satellite
What is obtained obtains depth of building according to the remote sensing optical stereo imaging data of the corresponding building in the residential area;Then basis
The building height and building range of acquisition simultaneously combine consensus data, are estimated using the multiple regression analysis method building density of population
Calculate model;Using the range of the building and the height of building as input variable, it is input to the density of population estimation mould
Type is calculated the density of population, and then is distributed regular grid corresponding to completed region of the city using space statistical analysis method
Space.
Further, the method includes the following steps:
First step is based on object-oriented by satellite acquisition according to the corresponding optical remote sensing data in completed region of the city
Classification method, choose corresponding characteristic of division, obtain building range;
Second step is drawn completed region of the city by space topological algorithm using the fine road network space big data in city
It is divided into the city block of several planars, if calculating the dry type point of interest ratio in the block of city respectively, passes through point of interest sky
Between big data divide city block be residential area and non-residential area;
Number is imaged according to the remote sensing optical stereo of the corresponding building in the residential area using what satellite obtained in third step
According to based on photogrammetry principles extraction digital surface model data, and then calculating obtains depth of building;
The building height of four steps, the building range obtained according to first step and third step acquisition simultaneously combines people
Mouth statistical data, utilizes multiple regression analysis method to construct population density estimation model;
5th step is input to four steps using the range of the building and the height of building as input variable
The population density estimation model of building, is calculated density data of population, so using space statistical analysis method distributed to
The corresponding regular grid space in completed region of the city.
Further, the first step by satellite obtain according to the corresponding optical remote sensing data in completed region of the city,
The optical remote sensing data are divided into homogeneous object using multi-scale division algorithm by the classification method based on object-oriented, and
Choose corresponding characteristic of division, the characteristic of division includes spectral signature, geometrical characteristic, textural characteristics and thematic index characteristic,
Using support vector cassification method, building range is obtained;
And/or the second step passes through space topological algorithm for city using the fine road network space big data in city
City built-up areas are divided into the city block of several planars, calculate ratio of all types of points of interest in the city block, described
Point of interest includes residential area, business, government, education and medical treatment, divides the city block by interest space of points big data
For residential area and non-residential area, and then building is divided into residential area building and non-residential area is built, obtains residential area building
Vector data;
And/or the remote sensing optics according to the corresponding building in the residential area that the third step utilizes satellite to obtain
Volumetric imaging data, it is sharp respectively based on the residential area architectural vector data obtained in photogrammetry principles and second step
The precision that digital surface model is extracted with the remote sensing optical stereo imaging data of various combination, determines the highest combination of precision
As final digital surface model data, and then calculates and obtain depth of building;
And/or the 5th step is that completed region of the city is divided into regular grid first, is built in using four steps
After density data of population is calculated in vertical population density estimation model, density data of population is made using space statistical analysis method
Regular grid space is assigned to for statistics demographic data.
Further, the first step by satellite obtain according to the corresponding optical remote sensing data in completed region of the city,
The optical remote sensing data prediction including ortho-rectification and visual fusion, then the classification method based on object-oriented are first passed through, is selected
Corresponding characteristic of division is taken, building range is obtained;
The fine road network space big data in city includes following road type:Pass through expressway, the city in city
Through street, national highway, provincial highway, county road, small towns village road, nine grades of roads, pedestrian roads;The space topological algorithm used includes with sky
Between intersection in topological relation, contact algorithm completed region of the city is divided into the city block of several planars.
Further, the first step by satellite obtain according to the corresponding optical remote sensing data in completed region of the city,
The pretreated optical remote sensing data are divided by the classification method based on object-oriented using multi-scale division algorithm
It verifies as according to the needs that building range is extracted, establishing classification system, and choose corresponding characteristic of division, selecting training sample
This, constructs training sample database, and the training sample is evenly distributed and covers entire completed region of the city, utilizes support vector cassification
Method obtains land cover classification as a result, and selecting building classification, acquisition building range in the classification results.
Further, the remote sensing light according to the corresponding building in the residential area that the third step utilizes satellite to obtain
Learn volumetric imaging data, be based on photogrammetry principles, according to facing for remote sensing optical stereo imaging data and forward sight, face and after
Depending on, forward sight and three kinds of backsight combinations, three kinds of optimal parameter extraction digital surface models of combination are determined, evaluate three kinds of combinations extractions
The precision of digital surface model is determined that the highest combination of precision is used as final digital surface model data, is determined with second step
Residential area on the basis of, calculate the ground level of residential area, and then calculate and obtain depth of building;
And/or the building that the building range that is obtained according to first step of the four steps and third step obtain
Highly, it is converted to by vector data by raster data by spatial data resampling, then according to building range computation
The floor area of building simultaneously combines depth of building to calculate building volume, utilizes using street as the consensus data of unit and building
Object volume data constructs population density estimation model using multiple regression analysis method.
A kind of completed region of the city density of population estimating system of space big data collaboration, which is characterized in that including:Successively connect
First device, second device, 3rd device, the 4th device and the 5th device connect;
The first device is based on according to the corresponding optical remote sensing data in completed region of the city towards right by what satellite obtained
The classification method of elephant obtains building range;
The second device is built up city by space topological algorithm using the fine road network space big data in city
Zoning is divided into the city block of several planars, and dividing city block by interest space of points big data is residential area and non-inhabitation
Area;
The 3rd device, the remote sensing optics according to the corresponding building in the residential area for being obtained using satellite are vertical
Body imaging data obtains depth of building;
4th device, for the building height and building range according to acquisition and in conjunction with consensus data, benefit
Population density estimation model is constructed with multiple regression analysis method;
5th device, for being input to using the height of the range of the building and building as input variable
The density of population is calculated in the population density estimation model of 4th device building, and then will using space statistical analysis method
It is distributed to the corresponding regular grid space in completed region of the city.
Further, the first device by satellite obtain according to the corresponding optical remote sensing data in completed region of the city,
Classification method based on object-oriented chooses corresponding characteristic of division, obtains building range;
The second device is built up city by space topological algorithm using the fine road network space big data in city
Zoning is divided into the city block of several planars, if calculating the dry type point of interest ratio in the block of city respectively, passes through interest
It is residential area and non-residential area that space of points big data, which divides city block,;
The 3rd device, using satellite obtain the remote sensing optical stereo according to the corresponding building in the residential area at
As data, digital surface model data are extracted based on photogrammetry principles, and then calculate and obtain depth of building.
Further, the first device includes that urban surface cover classification module and building range interconnected mention
Modulus block, the urban surface cover classification module by satellite obtain according to the corresponding optical remote sensing number in completed region of the city
According to, the optical remote sensing data are divided into homogeneous object using multi-scale division algorithm by the classification method based on object-oriented,
And corresponding characteristic of division is chosen, the characteristic of division includes that spectral signature, geometrical characteristic, textural characteristics and thematic index are special
Sign obtains urban surface cover classification result using support vector cassification method;The building range extraction module is from described
Building classification is selected in urban surface cover classification result, exports as individual building vector data, obtains building
Range;
And/or the second device includes block division module in city interconnected and residential area calculation processing module,
The city block division module is built up city by space topological algorithm using the fine road network space big data in city
Zoning is divided into the city block of several planars, and the residential area calculation processing module calculates all types of points of interest in the metropolitan district
Ratio in block, the point of interest include residential area, business, government, education and medical treatment, are drawn by interest space of points big data
Dividing the city block is residential area and non-residential area, and then building is divided into residential area building and non-residential area building,
Obtain residential area architectural vector data;
And/or the remote sensing optics according to the corresponding building in the residential area that the 3rd device utilizes satellite to obtain
Volumetric imaging data is based on photogrammetry principles, and the remote sensing optical stereo imaging data for being utilized respectively various combination extracts
The precision of digital surface model determines that the highest combination of precision is used as final digital surface model data, and then calculates and built
Build object height;
And/or the 4th device includes data conversion module interconnected and model construction module, the data turn
The depth of building that the building range and 3rd device that mold changing root tuber is obtained according to first device obtain, is adopted again by spatial data
It is converted to raster data by vector data by sample, then by model construction module according to the building surface of building range computation
It accumulates and depth of building is combined to calculate building volume, utilize using street as the consensus data of unit and buildings product
According to, utilize multiple regression analysis method construct population density estimation model;
And/or the 5th device includes regular grid division module interconnected and statistics demographic data distribution mould
Completed region of the city is divided into regular grid by block, the regular grid division module, and the statistics demographic data distribution module will
As input variable, the density of population for being input to the building of the 4th device is estimated for the range of the building and the height of building
Model is calculated, the density of population is calculated, and then using space statistical analysis method using density data of population as statistics demographic data
It distributes to the corresponding regular grid space in completed region of the city.
Further, when the first device includes urban surface cover classification module and building range interconnected
When extraction module, the first device further includes preprocessing module, and the preprocessing module includes ortho-rectification interconnected
Module and visual fusion module, the preprocessing module connect urban surface cover classification module, institute by influencing Fusion Module
Ortho-rectification module is stated to be carried out what is obtained by satellite at ortho-rectification according to the corresponding optical remote sensing data in completed region of the city
Reason obtains the remote sensing image for having georeferencing, then is carried out by the visual fusion module using principal component analysis fusion method
Fusion treatment has the fusion evaluation of multispectral characteristic to obtain existing high spatial resolution again, distant as pretreated optics
Feel data;The classification method of the urban surface cover classification module based on object-oriented, using multi-scale division algorithm by institute
It states pretreated optical remote sensing data and is divided into homogeneous object, according to the needs that building range is extracted, establish classification system,
And corresponding characteristic of division is chosen, training sample is selected, training sample database is constructed, the training sample is evenly distributed and covers whole
A completed region of the city obtains urban surface cover classification result using support vector machine classification method.
Technical effect of the invention is as follows:
The present invention relates to a kind of completed region of the city density of population evaluation methods of space big data collaboration, are alternatively referred to as empty
Between big data and high-resolution optical remote sensing data collaborative completed region of the city density of population evaluation method, this method passes through satellite
What is obtained obtains building range based on the classification method of object-oriented according to the corresponding optical remote sensing data in completed region of the city;It should
The data source of step is satellite acquisition according to the corresponding high-resolution optical remote sensing data in completed region of the city, based on towards right
As classification method, suitable characteristic of division can be chosen, which preferably includes spectral signature, geometrical characteristic, textural characteristics
With thematic index characteristic etc., and then extract building range.Compared with radar remote sensing data, high-resolution optical remote sensing data base
In the spectral characteristic of atural object, atural object spatial distribution is recorded in a manner of human eye vision imaging, image is it is more readily appreciated that and high-resolution
Rate optical remote sensing data have multiple spectral bands, in building range extraction process, in addition to object spectrum spy can be used
Sign is outer, can also use geometrical characteristic, textural characteristics and the thematic index constructed by spectral signature of atural object, compare radar
Data use single back scattering feature, and the precision for extracting building is higher, better quality.Using above-mentioned specific several
Characteristic of division combines, and these types of characteristic of division can play synergistic effect after being selected, specific such as mean value, standard deviation,
Brightness, shape index, edge index, homogeneity degree, contrast, similarity, soil correlation, vegetation correlation and shade correlation etc.
Mutually it can refer to and influence when calculation processing, the building range that can improve completed region of the city to the greatest extent extracts essence
Degree.Completed region of the city is divided by several planars by space topological algorithm using the fine road network space big data in city
City block, dividing city block by point of interest (POI, Point of Interest) space big data is residential area and non-
Residential area;Since the building of completed region of the city is divided into residential architecture and non-residential structure, the fine road network in completed region of the city is utilized
City block is preferably such as divided by modes such as intersection, contacts in spatial topotaxy several by network space big data
Planar block is divided into residential area and non-residential area using magnanimity interest space of points big data, so that this method by planar block
Have the characteristics that bulk fining, can accurately extract residential area building.Recycle satellite obtain according to the residence
The remote sensing optical stereo imaging data of the corresponding building in settlement obtains depth of building;On the basis of residential area, pass through high score
The remote sensing optical stereo imaging data of resolution can be extracted digital surface model data based on photogrammetry principles, and then calculate and obtain
High-precision depth of building is obtained, and data cost is lower.Then it according to the building height of acquisition and building range and combines
Consensus data constructs population density estimation model using multiple regression analysis method, the population density estimation model construction essence
Standard, the model are completed region of the city population density estimation model, statistical law of the model based on population and living space, benefit
The statistical model of the density of population and architectural volume is established with linear regression method, can more realistically express population and living space
Relationship;Using the range of the building and the height of building as input variable, it is input to the density of population estimation mould
Type is calculated the density of population, and then is distributed regular grid corresponding to completed region of the city using space statistical analysis method
Space.The fine density of population grid in the completed region of the city that the present invention makes avoids the prior art from being based on Land Use/land completely
Covering data and the population spatial distribution product of Optimal Parameters production can not accurately portray completed region of the city population actual spatial distribution
It is the problem of feature, more fine on space scale compared to the statistical data based on administrative area, the density of population estimation of building
Model science is reliable, can precisely and accurately realize that the completed region of the city density of population is estimated, be more suitable for urban planning, taking precautions against natural calamities subtracts
Calamity and business decision are conducive to further genralrlization and application to the needs of fine demographic data.
The invention further relates to a kind of space big data collaboration completed region of the city density of population estimating system, the system with it is upper
The completed region of the city density of population evaluation method for stating the collaboration of space big data is corresponding, it is understood that be to realize the above method
System, which includes first device, second device, 3rd device, the 4th device, the 5th device.First device, by defending
What star obtained obtains building range based on the classification method of object-oriented according to the corresponding optical remote sensing data in completed region of the city;
Completed region of the city is divided into using the fine road network space big data in city by space topological algorithm several by second device
The city block of planar, dividing city block by interest space of points big data is residential area and non-residential area;3rd device is used
Building height is obtained according to the remote sensing optical stereo imaging data of the corresponding building in the residential area in what is obtained using satellite
Degree;4th device utilizes multiple regression for the building height and building range according to acquisition and in conjunction with consensus data
Analytic approach constructs population density estimation model;5th device, for using the height of the range of the building and building as
Input variable is input to the population density estimation model, the density of population is calculated, and then utilize space statistical analysis method will
It is distributed to the corresponding regular grid space in completed region of the city.Each device of the system can sequentially cooperate, the system knot
Remote sensing technology, the classification method of object-oriented, space topological algorithm, photogrammetry principles, multiple regression analysis method, space are closed
The many technologies of statistical analysis method etc., be utilized respectively high-resolution optical remote sensing data extract building range, identification residential area and
Non- residential area calculates building height, the density of population of building completed region of the city fining using remote sensing optical stereo imaging data
Appraising model, the density of population for realizing statistics demographic data to fine grid are distributed;The completed region of the city that the present invention makes is fine
Density of population grid, it is more fine on space scale compared to the statistical data based on administrative area, and improve city and build up
The working efficiency of the fine density of population remote sensing spatialization in area, and the system has the excellent performance of accuracy and reliability, it is more suitable
Close urban planning, the needs prevented and reduced natural disasters with business decision to fine demographic data.
Detailed description of the invention
Fig. 1 is the flow chart of the completed region of the city density of population evaluation method of big data collaboration in space of the present invention.
Fig. 2 is the preferred flow charts of the completed region of the city density of population evaluation method of big data collaboration in space of the present invention.
Fig. 3 is another preferred flow of the completed region of the city density of population evaluation method of big data collaboration in space of the present invention
Figure.
Fig. 4 is the preferred structure signal of the completed region of the city density of population estimating system of big data collaboration in space of the present invention
Figure.
Specific embodiment
The present invention will be described with reference to the accompanying drawing.
The present invention relates to a kind of completed region of the city density of population evaluation methods of space big data collaboration, and flow chart is as schemed
Shown in 1, this method comprises the following steps:S1:By satellite acquisition according to the corresponding optical remote sensing data base in completed region of the city
Building range is obtained in the classification method of object-oriented;S2:Pass through space using the fine road network space big data in city
Completed region of the city is divided into the city block of several planars by Topology Algorithm, divides city block by interest space of points big data
For residential area and non-residential area;S3:The remote sensing optical stereo according to the corresponding building in the residential area obtained using satellite
Imaging data obtains depth of building;S4:According to the building height of acquisition and building range and consensus data is combined, benefit
Population density estimation model is constructed with multiple regression analysis method;S5:Using the range of the building and the height of building as
Input variable is input to the population density estimation model, the density of population is calculated, and then utilize space statistical analysis method will
It is distributed to the corresponding regular grid space in completed region of the city.
It referring to fig. 2, is the preferred flow charts for the completed region of the city density of population evaluation method that space big data cooperates with, including
Following steps:
S1:First step is not understood as building range extraction step, by satellite acquisition according to completed region of the city pair
The optical remote sensing data answered, the classification method based on object-oriented choose corresponding characteristic of division, obtain building range;
S2:Second step is not understood as residential area and non-residential area separating step, utilizes the fine road network space in city
Completed region of the city is divided into the city block of several planars by space topological algorithm by big data, if calculating dry type point of interest
Ratio in the block of city respectively, dividing city block by interest space of points big data is residential area and non-residential area;
S3:Third step is not understood as building height and calculates step, is obtained using satellite corresponding according to the residential area
Building remote sensing optical stereo imaging data, based on photogrammetry principles extract digital surface model data, and then calculate
Obtain depth of building;
S4:Four steps is not understood as population density estimation model construction step, the building obtained according to first step
Range and the building height of third step acquisition simultaneously combine consensus data, and multiple regression analysis method is utilized to construct the density of population
Appraising model;
S5:5th step is not understood as density of population estimation steps, by the height of the range of the building and building
As input variable, it is input to the population density estimation model of four steps building, density data of population, Jin Erli is calculated
Regular grid space corresponding to completed region of the city is distributed with space statistical analysis method.
Below by taking No. GF-2 and ZY-3 satellite data as an example, the high-resolution optical remote sensing data that are obtained respectively by it
With remote sensing optical stereo imaging data, as shown in connection with fig. 3 further preferred process, the present invention will be described in detail the big number in space
According to the completed region of the city density of population evaluation method of collaboration.
S1:First step, it will be appreciated that be the meter for extracting building range using high-definition remote sensing optical remote sensing data
Calculate step:It is preferably first right using the acquisition of GF-2 satellite according to the corresponding high-resolution optical remote sensing data in completed region of the city
The pretreatment of data progress ortho-rectification and fusion treatment.Using fused high-resolution optical remote sensing data, based on towards
Pretreated optical remote sensing data are divided into homogeneous object using multi-scale division algorithm, and selected by the classification method of object
Corresponding characteristic of division is taken, the characteristic of division includes spectral signature, geometrical characteristic, textural characteristics and thematic index characteristic, benefit
With support vector cassification method, building range is obtained.
It further illustrates, in the above-described embodiments, which is obtained corresponding according to completed region of the city by satellite
High-resolution optical remote sensing data using GF-2 satellite optical remote sensing data, the highest spatial discrimination of GF-2 satellite
Rate can reach 1 meter, the spectral characteristic based on atural object, record atural object spatial distribution in a manner of human eye vision imaging, image is more
It can be readily appreciated that and have 4 spectral bands, in building range extraction process, other than spectral characteristic of ground can be used,
Geometrical characteristic, textural characteristics and the thematic index constructed by spectral signature of atural object can also be used.On existing extraction ground
In the method for object, atural object can be extracted using method pixel-based, the extracting method of object-oriented can also be used, it is contemplated that
Based on Object--oriented method than traditional method pixel-based when extracting ground object target, more it is contemplated that between target
Spatial relationship information, such as:Topological relation etc..Therefore, in embodiments of the present invention, it is preferred to use point based on object-oriented
The range of class method acquisition building.The classification of object-oriented refers to according to the spectral information of image, geometry information, texture
Information and thematic index information etc., optical remote sensing image is split with shape by settable certain homogeneous standard parameter
At imaged object, each imaged object has the set of the pixel of similar spectral feature.The method is not with single pixel
To analyze target, but using the pixel geometry in image as analytical unit, the connection between object and ambient enviroment has been fully considered
The factors such as system, the extraction to image information is completed by characteristics of objects knowledge base.The classification method of object-oriented is in processing number
Three big steps are generally included during:Data segmentation, data classification and post-processing.Wherein, data segmentation and data point
Class is important step.In remote sensing images field, data segmentation algorithm has the segmentation of the data based on pixel value, has based on feature
Data segmentation algorithm etc..Wherein, common dividing method has the partitioning algorithm based on threshold value, the segmentation based on wavelet transformation to calculate
Method, multi-scale division algorithm, etc., present invention preferably employs multi-scale division algorithms, it is of course also possible to use other segmentations are calculated
Method.Likewise, data classification method is also possible to the data classification method based on data space relationship, point based on data characteristics
Class method etc..
The step of building range is extracted is as follows:
(1) optical remote sensing data prediction
The step of optical remote sensing data prediction includes ortho-rectification and visual fusion.Ortho-rectification process utilizes No. GF-2
The included RPB file of data carries out ortho-rectification processing by the RPC ortho-rectification module in remote sensing image processing software and obtains
Remote sensing images with georeferencing;Visual fusion process utilizes corrected panchromatic and multispectral data, utilizes remote sensing figure
As the principal component analysis fusion method in processing software carry out fusion treatment have to obtain existing high spatial resolution again it is multispectral
The fusion evaluation of feature, as pretreated optical remote sensing data.
(2) urban surface cover classification
Using pretreated high-resolution optical remote sensing data, it is based on object-oriented classification method, more rulers are carried out to image
Degree segmentation obtains the object unit (being divided into homogeneous object using multi-scale division algorithm) of homogeneous, is mentioned according to building range
The needs taken establish classification system, including building, road, greenery patches, water body, shade, choose suitable characteristic of division (such as 1 institute of table
Show), training sample is selected, training sample database is constructed, training sample is evenly distributed and covers entire completed region of the city, utilizes support
Vector machine classification method obtains land cover classification result.
In preferred embodiment, corresponding characteristic of division can be found in the following table 1, and characteristic of division includes spectral signature, geometry spy
Sign, textural characteristics and thematic index characteristic, table 1 are alternatively referred to as being terrain classification mark sheet.
Table 1
For example the spectral signature type in table 1, feature correspond to mean value, standard deviation, brightness and maximum difference;Geometry is special
Type is levied, feature corresponds to shape index, circularity, rectangular exponential, length-width ratio and edge index;Textural characteristics type, it is special
Sign corresponds to homogeneity degree, contrast, similarity and entropy;Thematic index characteristic type, feature correspond to soil correlation, water body phase
Pass, vegetation correlation, shade correlation and the relevant index of building etc.;Above-mentioned each feature has corresponded to corresponding calculation formula and spy
Sign description.
(3) building range is extracted
Building classification is selected from urban surface cover classification result, merges operation, exports as and individually builds
Object vector data is built, i.e. acquisition building range.
S2:Second step, it will be appreciated that for the calculating step for being residential area and the separation of non-residential area:It is determined according to first step
Building range extracting method obtain building space distribution, including residential architecture and non-residential structure, residential architecture
Including low rise buildings cluster, dormitory building in flat building, city etc.;Non-residential structure includes accommodating for floating population
Large-scale Hotel and small-sized hotel, enterprises and institutions' office building, commerce and trade building, tourist attractions building, square building etc. public clothes
Business facility etc..Using the fine road network space big data in city, path space topology is established by space topological algorithm and is closed
Completed region of the city, is divided into the city block of several planars, can be regular block by system.Wherein, the fine road network in city
Space big data may include:Pass through expressway, city expressway, national highway, provincial highway, county road, the small towns village road, nine grades in city
Road, pedestrian road and other roads;The space topological algorithm used includes with intersection, the contact in spatial topotaxy
Completed region of the city is divided into the city block of several planars by algorithm.Using magnanimity interest space of points big data, calculate all types of
Ratio of the point of interest in block, interest vertex type include residential quarters, residential building, hotel, park plaza, finance clothes
Block is divided into residential area and non-residential area according to the ratio of residential point of interest by the detailed type such as business, commercial mansion, into
And building is divided into residential architecture and non-residential structure, obtain residential area architectural vector data.
In addition to the implementation, the present embodiment passes through space topological preferably by the fine road network space big data in city
Completed region of the city is divided into the city block of several planars by algorithm, if calculating dry type point of interest respectively in the block of city
Ratio, dividing city block by interest space of points big data is residential area and non-residential area.Wherein, point of interest ratio calculates public
Formula is:
Wherein, RiFor the i-th class point of interest ratio, NiFor the quantity of the i-th class point of interest, NjFor the quantity of jth class point of interest, m
For point of interest number of types.
Residential area type judgment formula is:
Rres=Max (R1,R2,R3...Rm)
Wherein, RresFor residential point of interest ratio, R1,R2,R3...RmFor the ratio of all types point of interest, m is interest
Vertex type quantity.
S3:Third step, it will be appreciated that be that building height calculates step:Using the acquisition of ZY-3 satellite according to the residence
The high-definition remote sensing optical stereo imaging data of the corresponding building in settlement extracts digital surface mould based on photogrammetry principles
Type (DSM) data, ZY-3 satellite have face, the imaging capability in three directions of forward sight and backsight, therefore, remote sensing optical stereo
The various combination of imaging data be face with forward sight, face with backsight, forward sight and backsight, be utilized respectively three kinds of data splittings and extract
DSM data, since the height of residential architecture is often referred to ground or more, the height relative to earth's surface.It is determined based on second step
Residential area architectural vector data generate the buffer area of residential architecture, the residential area range after statistics buffering using buffer zone analysis
Interior DSM minimum value subtracts ground level with DSM data as ground level, and the ground height for obtaining building is to build
Highly, based on actual measurement depth of building, the precision of three kinds of component building object height is evaluated, the highest combination of choice accuracy is as most
Whole depth of building data.
It further illustrates, in the above-described embodiments, which is obtained corresponding according to the residential area by satellite
Building high-definition remote sensing optical stereo imaging data using ZY-3 satellite remote sensing optical stereo imaging data,
Abbreviation ZY-3 image belongs to high-resolution optical image, and highest spatial resolution can reach 2.1 meters, be divided into it is panchromatic and
Multispectral two major classes, wherein panchromatic wave-band further comprises forward sight, faces and backsight image, can be used as stereogram use, and
Multispectral image then contains certain spectral information, suffers from important work for the extraction etc. of image classification, building effects
With.Can certainly use other available satellites to remote sensing optical stereo imaging data, as ASTER, ALOS, PRISM,
Resource No.1 02C etc..In the above-described embodiments, it is above-mentioned it is photogrammetric be using satellite sensor obtain measurand image believe
Breath, by processing, processing and analysis, obtains the information such as shape, size, spatial position, property and the correlation of measurand
Theory and technology.
Wherein, depth of building calculation formula is:
Hbuilding=Hdsm-Hsurface
Wherein, HbuildingFor the height of residential architecture, HdsmFor the digital surface mould being calculated based on photogrammetry principles
Type (DSM), HsurfaceFor the ground level around residential architecture.
Ground level calculation formula around above-mentioned residential architecture is as follows
Hsurface=Min (h1,h2,h3...hn)
Wherein, h1,h2,h3...hnFor the height value of grid in residential architecture buffer area, n is grid in residential architecture buffer area
Lattice number.
S4:Four steps, it will be appreciated that be population density estimation model construction step:The building obtained according to first step
The building height data that object range and third step obtain, are converted to grid by vector data for it by spatial data resampling
Data then according to the floor area of building of building range computation, and combine depth of building to calculate building volume, in conjunction with
Street is consensus data and the buildings volume data of unit, estimates mould using the multiple regression analysis method building density of population
Type.
In embodiments of the present invention, population density estimation model can be one of population estimation model, population estimation
There are many kinds of methods, can be land use method, flat method, urban size method and image picture element method etc..People and its
There is very close association between living space, and the living space of urban population is presented mostly in the form of building,
Theoretically, the relevant parameter of building and the relationship between accurate description population and the building be can use, and then established more
Add fine completed region of the city population density estimation model.
Preferably, in embodiments of the present invention, according to the building height of acquisition and building range and in conjunction with demographics
Data construct population density estimation model using multiple regression analysis method.
Specifically, the building model step of population density estimation model is:
It is by the population estimation model of unit of street:
Pop_Ei=α Vi+β
Wherein, Pop_EiFor the estimation population quantity of i-th of street unit, ViFor the residential architecture body of i-th of street unit
Product, α and β are model parameter.
There are evaluated errors, as model residual error between estimation population and statistics population, preferably carry out residual error at adjustment
Reason, to guarantee that statistical data is quantitatively consistent with estimated data.That is, based on using street as the demographics number of unit
Population density estimation model is constructed according to architectural volume, further, computation model estimates residual error, constructs residual error adjustment Models.
Adjustment Models are:
Wherein, EiFor the population error of i-th of street unit of cells architectural volume, Pop_EiFor estimating for i-th street unit
Count the size of population, Pop_SiFor the statistics size of population of i-th of street unit, ViFor the residential architecture body of i-th of street unit
Product.
S5:5th step, it will be appreciated that be density of population estimation steps:By the range of the building of extraction and building
Height is used as input variable, is input to the population density estimation model and adjustment Models of four steps building, population is calculated
Density data, according to completed region of the city range create-rule grid, using space statistical analysis method using density data of population as
Statistics demographic data distributes regular grid space corresponding to completed region of the city.
To sum up, the fine density of population in completed region of the city more fine can accurately and efficiently be realized by above-mentioned five steps
Estimation.
The invention further relates to a kind of completed region of the city density of population estimating system of space big data collaboration, the system and sheet
The invention above method is corresponding, it will be appreciated that be the completed region of the city density of population estimation for realizing big data collaboration in space of the present invention
The system of method, system of the present invention include sequentially connected first device, second device, 3rd device, the 4th device,
Five devices.Wherein, first device, by satellite obtain according to the corresponding optical remote sensing data in completed region of the city, based on towards
The classification method of object can choose corresponding characteristic of division, obtain building range;Second device utilizes the fine road in city
Completed region of the city is divided into the city block of several planars by space topological algorithm by cyberspace big data, can be calculated several
The type point of interest ratio in the block of city respectively, it is residential area and non-for dividing city block by interest space of points big data
Residential area;Number is imaged according to the remote sensing optical stereo of the corresponding building in residential area for what is obtained using satellite in 3rd device
According to, it can be based on photogrammetry principles extraction digital surface model data, and then calculate and obtain depth of building;4th device is used
In the building height and building range according to acquisition and in conjunction with consensus data, multiple regression analysis method is utilized to construct population
Density estimation model;5th device, for being input to the 4th using the height of the range of building and building as input variable
The population density estimation model of device building, is calculated the density of population, so using space statistical analysis method distributed to
The corresponding regular grid space in completed region of the city.
Fig. 4 is the preferred structure schematic diagram of the completed region of the city density of population estimating system of space big data collaboration, the reality
The first device for applying example includes that sequentially connected preprocessing module, urban surface cover classification module and building range extract mould
Block, preprocessing module include ortho-rectification module and visual fusion module interconnected, and preprocessing module is by influencing fusion
Module connects urban surface cover classification module;The second device of the embodiment includes block division module in city interconnected
Mould is extracted with the building range of residential area calculation processing module, the city block division module connection first device of second device
Block;The residential area calculation processing module of second device connects 3rd device;4th device includes data conversion mould interconnected
Block and model construction module, the building range extraction module and 3rd device of first device are connected to the data of the 4th device
Conversion module;5th device includes regular grid division module interconnected and statistics demographic data distribution module, the 5th dress
Set the model construction module that the 4th device is connected by regular grid division module.
Specifically, in first device, preprocessing module is further preferred module, and ortho-rectification module therein will lead to
That crosses satellite acquisition carries out ortho-rectification processing acquisition with georeferencing according to the corresponding optical remote sensing data in completed region of the city
Remote sensing image, then by visual fusion module using principal component analysis fusion method carry out fusion treatment to obtain existing high-altitude
Between resolution ratio there is the fusion evaluation of multispectral characteristic again, as pretreated optical remote sensing data;Urban surface covering point
Pretreated optical remote sensing data are divided by classification method of the generic module based on object-oriented using multi-scale division algorithm
Homogeneous object establishes classification system, and choose corresponding characteristic of division, characteristic of division according to the needs that building range is extracted
It may include spectral signature, geometrical characteristic, textural characteristics and the thematic index characteristic etc. as described in above-mentioned table 1, select training sample
This, constructs training sample database, and training sample is evenly distributed and covers entire completed region of the city, utilizes support vector cassification side
Method obtains urban surface cover classification result;Building range extraction module is selected from urban surface cover classification result
Building classification exports as individual building vector data, obtains building range.
Preferably, in second device, city block division module is logical using the fine road network space big data in city
Cross the city block that completed region of the city is divided into several planars by space topological algorithm;Calculation processing module in residential area calculates all kinds of
Ratio of the type point of interest in the block of city, point of interest includes residential area, business, government, education and medical treatment etc., by emerging
It is residential area and non-residential area that interesting space of points big data, which divides city block, and then building is divided into residential area building and non-
Residential area building, obtains residential area architectural vector data.
Preferably, in 3rd device, the remote sensing optics according to the corresponding building in residential area of satellite acquisition is utilized
Volumetric imaging data is based on photogrammetry principles, and the remote sensing optical stereo imaging data for being utilized respectively various combination extracts
The precision of digital surface model determines that the highest combination of precision is used as final digital surface model data, and then calculates and built
Build object height.
Preferably, in the 4th device, the building range and third that data conversion module is obtained according to first device are filled
The depth of building of acquisition is set, the range of the building of acquisition and the height of building are vector data, pass through spatial data
Both vector datas are respectively converted into raster data by resampling;Then by model construction module according to building range computation
The floor area of building and combine depth of building to calculate building volume, utilize using street the consensus data of unit and to build
Object volume data is built, constructs population density estimation model using multiple regression analysis method.
Preferably, in the 5th device, completed region of the city is divided into regular grid by regular grid division module;Count people
Mouth data allocation module is input to the people of the 4th device building using the range of building and the height of building as input variable
The density of population is calculated in mouth density estimation model, and then using space statistical analysis method using density data of population as statistics
Demographic data distributes regular grid space corresponding to completed region of the city.
Above-mentioned each device cooperates, and precisely and accurately realizes the estimation of the completed region of the city density of population.
It should be pointed out that specific embodiment described above can make those skilled in the art that the present invention be more fully understood
It creates, but do not limit the invention in any way is created.Therefore, although this specification creates the present invention referring to drawings and examples
It makes and has been carried out detailed description, it will be understood by those skilled in the art, however, that still can modify to the invention
Or equivalent replacement, in short, the technical solution and its improvement of all spirit and scope for not departing from the invention, should all contain
It covers in the protection scope of the invention patent.
Claims (10)
1. a kind of completed region of the city density of population evaluation method of space big data collaboration, which is characterized in that this method is by defending
What star obtained obtains building range based on the classification method of object-oriented according to the corresponding optical remote sensing data in completed region of the city;
And completed region of the city is divided by several planars by space topological algorithm using the fine road network space big data in city
City block, dividing city block by point of interest (POI) space big data is residential area and non-residential area;Satellite is recycled to obtain
What is taken obtains depth of building according to the remote sensing optical stereo imaging data of the corresponding building in the residential area;Then basis obtains
Building height and building range and combine consensus data, utilize multiple regression analysis method building the density of population estimation
Model;Using the range of the building and the height of building as input variable, it is input to the population density estimation model,
The density of population is calculated, and then is distributed regular grid sky corresponding to completed region of the city using space statistical analysis method
Between.
2. the method according to claim 1, wherein the method includes the following steps:
First step, by satellite acquisition according to the corresponding optical remote sensing data in completed region of the city, point based on object-oriented
Class method chooses corresponding characteristic of division, obtains building range;
Completed region of the city is divided into using the fine road network space big data in city by space topological algorithm by second step
The city block of several planars is big by the interest space of points if calculating the dry type point of interest ratio in the block of city respectively
It is residential area and non-residential area that data, which divide city block,;
Third step, the remote sensing optical stereo imaging data according to the corresponding building in the residential area obtained using satellite,
Digital surface model data are extracted based on photogrammetry principles, and then calculates and obtains depth of building;
The building height and combination population that four steps, the building range obtained according to first step and third step obtain are united
It counts, constructs population density estimation model using multiple regression analysis method;
5th step is input to four steps building using the range of the building and the height of building as input variable
Population density estimation model, be calculated density data of population, and then distributed to city using space statistical analysis method
The corresponding regular grid space in built-up areas.
3. according to the method described in claim 2, it is characterized in that, the first step is built by what satellite obtained according to city
At the corresponding optical remote sensing data in area, the classification method based on object-oriented is distant by the optics using multi-scale division algorithm
Sense data are divided into homogeneous object, and choose corresponding characteristic of division, the characteristic of division include spectral signature, geometrical characteristic,
Textural characteristics and thematic index characteristic utilize support vector cassification method, acquisition building range;
And/or the second step is built city by space topological algorithm using the fine road network space big data in city
It is divided into the city block of several planars at zoning, calculates ratio of all types of points of interest in the city block, the interest
Point includes residential area, business, government, education and medical treatment, and dividing the city block by interest space of points big data is to occupy
Settlement and non-residential area, and then building is divided into residential area building and non-residential area building, obtain residential area architectural vector
Data;
And/or the remote sensing optical stereo according to the corresponding building in the residential area that the third step utilizes satellite to obtain
Imaging data is utilized respectively not based on the residential area architectural vector data obtained in photogrammetry principles and second step
The precision that digital surface model is extracted with the remote sensing optical stereo imaging data of combination, determines the highest combination conduct of precision
Final digital surface model data, and then calculate and obtain depth of building;
And/or the 5th step is that completed region of the city is divided into regular grid first, is established in using four steps
After density data of population is calculated in population density estimation model, using space statistical analysis method using density data of population as system
Meter demographic data is assigned to regular grid space.
4. according to the method in claim 2 or 3, which is characterized in that the first step by satellite obtain according to city
The corresponding optical remote sensing data in city built-up areas first pass through the optical remote sensing data prediction including ortho-rectification and visual fusion,
Again based on the classification method of object-oriented, corresponding characteristic of division is chosen, obtains building range;
The fine road network space big data in city includes following road type:Expressway, the city for passing through city are quick
Road, national highway, provincial highway, county road, small towns village road, nine grades of roads, pedestrian roads;The space topological algorithm used includes being opened up with space
Completed region of the city is divided into the city block of several planars by the algorithm of the intersection, contact flutterred in relationship.
5. according to the method described in claim 4, it is characterized in that, the first step is built by what satellite obtained according to city
At the corresponding optical remote sensing data in area, the classification method based on object-oriented, using multi-scale division algorithm by the pretreatment
Optical remote sensing data afterwards are divided into homogeneous object, according to the needs that building range is extracted, establish classification system, and choose phase
The characteristic of division answered selects training sample, constructs training sample database, and the training sample is evenly distributed and covers entire city and builds
Land cover classification is obtained as a result, and selecting building in the classification results using support vector machine classification method at area
Species are other, obtain building range.
6. according to the method in claim 2 or 3, which is characterized in that the third step is using satellite acquisition according to institute
The remote sensing optical stereo imaging data of the corresponding building in residential area is stated, photogrammetry principles are based on, according to remote sensing optical stereo
Imaging data face with forward sight, face with backsight, forward sight and three kinds of backsight combinations, determine three kinds of optimal parameter extractions of combination
Digital surface model evaluates the precision that digital surface model is extracted in three kinds of combinations, determines the highest combination of precision as final number
Word surface model data calculate the ground level of residential area, and then calculate and obtain on the basis of the residential area that second step determines
Depth of building;
And/or the depth of building that the building range that is obtained according to first step of the four steps and third step obtain,
It is converted into raster data by vector data by spatial data resampling, then according to the building of building range computation
Area simultaneously combines depth of building to calculate building volume, utilizes using street as the consensus data of unit and building volume
Data construct population density estimation model using multiple regression analysis method.
7. a kind of completed region of the city density of population estimating system of space big data collaboration, which is characterized in that including:It is sequentially connected
First device, second device, 3rd device, the 4th device and the 5th device,
The first device, by satellite acquisition according to the corresponding optical remote sensing data in completed region of the city based on object-oriented
Classification method obtains building range;
The second device is drawn completed region of the city by space topological algorithm using the fine road network space big data in city
It is divided into the city block of several planars, dividing city block by interest space of points big data is residential area and non-residential area;
The 3rd device, for using satellite obtain the remote sensing optical stereo according to the corresponding building in the residential area at
As data obtain depth of building;
4th device, for according to the building height and building range of acquisition and combining consensus data, using more
First regression analysis constructs population density estimation model;
5th device, for being input to the 4th using the height of the range of the building and building as input variable
The population density estimation model of device building is calculated the density of population, and then utilizes space statistical analysis method by its point
It is assigned to the corresponding regular grid space in completed region of the city.
8. system according to claim 7, which is characterized in that the first device is built by what satellite obtained according to city
At the corresponding optical remote sensing data in area, the classification method based on object-oriented chooses corresponding characteristic of division, obtains building model
It encloses;
The second device is drawn completed region of the city by space topological algorithm using the fine road network space big data in city
It is divided into the city block of several planars, if calculating the dry type point of interest ratio in the block of city respectively, passes through point of interest sky
Between big data divide city block be residential area and non-residential area;
Number is imaged according to the remote sensing optical stereo of the corresponding building in the residential area using what satellite obtained in the 3rd device
According to based on photogrammetry principles extraction digital surface model data, and then calculating obtains depth of building.
9. system according to claim 8, which is characterized in that the first device includes that urban surface interconnected covers
Lid categorization module and building range extraction module, the urban surface cover classification module by satellite obtain according to city
The corresponding optical remote sensing data in built-up areas, the classification method based on object-oriented, using multi-scale division algorithm by the optics
Remotely-sensed data is divided into homogeneous object, and chooses corresponding characteristic of division, and the characteristic of division includes spectral signature, geometry spy
Sign, textural characteristics and thematic index characteristic obtain urban surface cover classification result using support vector cassification method;It is described
Building range extraction module selects building classification from the urban surface cover classification result, exports as and individually builds
Object vector data is built, building range is obtained;
And/or the second device includes block division module in city interconnected and residential area calculation processing module, it is described
City block division module is drawn completed region of the city by space topological algorithm using the fine road network space big data in city
It is divided into the city block of several planars, the residential area calculation processing module calculates all types of points of interest in the city block
Ratio, the point of interest includes residential area, business, government, education and medical treatment, passes through interest space of points big data and divides institute
Stating city block is residential area and non-residential area, and then building is divided into residential area building and non-residential area building, is obtained
Residential area architectural vector data;
And/or the remote sensing optical stereo according to the corresponding building in the residential area that the 3rd device utilizes satellite to obtain
Imaging data is based on photogrammetry principles, and the remote sensing optical stereo imaging data for being utilized respectively various combination extracts number
The precision of surface model determines that the highest combination of precision is used as final digital surface model data, and then calculates and obtain building
Highly;
And/or the 4th device includes data conversion module interconnected and model construction module, the data conversion mould
The depth of building that the building range and 3rd device that root tuber is obtained according to first device obtain, will by spatial data resampling
It is converted to raster data by vector data, then by model construction module according to the floor area of building of building range computation simultaneously
Building volume is calculated in conjunction with depth of building, is utilized using street as the consensus data of unit and buildings volume data,
Population density estimation model is constructed using multiple regression analysis method;
And/or the 5th device includes regular grid division module interconnected and statistics demographic data distribution module, institute
It states regular grid division module and completed region of the city is divided into regular grid, the statistics demographic data distribution module is built described
The range of object and the height of building are built as input variable, is input to the density of population estimation mould of the 4th device building
Type is calculated the density of population, and then is distributed using space statistical analysis method using density data of population as statistics demographic data
To the corresponding regular grid space in completed region of the city.
10. system according to claim 9, which is characterized in that when the first device includes city interconnected
When table cover classification module and building range extraction module, the first device further includes preprocessing module, the pretreatment
Module includes ortho-rectification module and visual fusion module interconnected, and the preprocessing module is connected by influencing Fusion Module
Connect urban surface cover classification module, the ortho-rectification module by by satellite obtain according to the corresponding light in completed region of the city
It learns remotely-sensed data and carries out the remote sensing image that ortho-rectification processing acquisition has georeferencing, then by the visual fusion module
Carrying out fusion treatment using principal component analysis fusion method has the fusion of multispectral characteristic to obtain existing high spatial resolution again
Image, as pretreated optical remote sensing data;The classification side of the urban surface cover classification module based on object-oriented
The pretreated optical remote sensing data are divided into homogeneous object using multi-scale division algorithm, according to building model by method
The needs for enclosing extraction establish classification system, and choose corresponding characteristic of division, select training sample, construct training sample database, institute
It states training sample to be evenly distributed and cover entire completed region of the city, using support vector machine classification method, acquisition urban surface covers
Lid classification results.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810379934.9A CN108871286B (en) | 2018-04-25 | 2018-04-25 | Urban built-up area population density estimation method and system based on spatial big data cooperation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810379934.9A CN108871286B (en) | 2018-04-25 | 2018-04-25 | Urban built-up area population density estimation method and system based on spatial big data cooperation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108871286A true CN108871286A (en) | 2018-11-23 |
CN108871286B CN108871286B (en) | 2021-02-23 |
Family
ID=64327355
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810379934.9A Active CN108871286B (en) | 2018-04-25 | 2018-04-25 | Urban built-up area population density estimation method and system based on spatial big data cooperation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108871286B (en) |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109784667A (en) * | 2018-12-21 | 2019-05-21 | 中国科学院遥感与数字地球研究所 | A kind of vacant monitoring method in house based on noctilucence remotely-sensed data |
CN109801296A (en) * | 2019-01-14 | 2019-05-24 | 中国科学院地理科学与资源研究所 | A kind of sponge urban construction underlying surface nested type remote sensing fining drafting method |
CN109858450A (en) * | 2019-02-12 | 2019-06-07 | 中国科学院遥感与数字地球研究所 | Ten meter level spatial resolution remote sensing image cities and towns extracting methods of one kind and system |
CN110135328A (en) * | 2019-05-10 | 2019-08-16 | 中国科学院遥感与数字地球研究所 | Pakistani land cover pattern information extracting method based on multi-source Spatial Data |
CN110263109A (en) * | 2019-05-22 | 2019-09-20 | 中国科学院地理科学与资源研究所 | A kind of family's amount evaluation method merging internet information and GIS technology |
CN110298253A (en) * | 2019-05-30 | 2019-10-01 | 特斯联(北京)科技有限公司 | A kind of physically weak quasi- display methods of urban architecture based on population big data and system |
CN110310246A (en) * | 2019-07-05 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | A kind of cane -growing region remote sensing information extracting method based on three-line imagery |
CN110716998A (en) * | 2019-07-25 | 2020-01-21 | 首都师范大学 | Method for spatializing fine-scale population data |
CN110807724A (en) * | 2019-10-26 | 2020-02-18 | 福建省伟志地理信息科学研究院 | Population flow monitoring system and method based on satellite remote sensing |
CN110990661A (en) * | 2019-10-23 | 2020-04-10 | 南通大学 | Interest area entrance and exit extraction method based on road network constraint and density clustering |
CN111047695A (en) * | 2019-12-03 | 2020-04-21 | 中国科学院地理科学与资源研究所 | Method for extracting height spatial information and contour line of urban group |
CN111177587A (en) * | 2019-12-12 | 2020-05-19 | 广州地理研究所 | Shopping street recommendation method and device |
CN111292018A (en) * | 2020-03-11 | 2020-06-16 | 长光卫星技术有限公司 | City vulnerability model construction method |
CN111382330A (en) * | 2020-03-10 | 2020-07-07 | 智慧足迹数据科技有限公司 | Land property identification method and device, electronic equipment and storage medium |
CN111461163A (en) * | 2020-02-25 | 2020-07-28 | 河南大学 | Urban interior PM2.5 concentration simulation and population exposure evaluation method and device |
CN111506677A (en) * | 2019-01-31 | 2020-08-07 | 阿里巴巴集团控股有限公司 | Method and device for generating urban built-up area range |
CN112085636A (en) * | 2020-08-12 | 2020-12-15 | 马琦伟 | Urban functional shrinkage analysis method and device and storage medium |
CN112115844A (en) * | 2020-09-15 | 2020-12-22 | 中国科学院城市环境研究所 | Urban population data analysis method based on multi-source remote sensing image and road network data |
CN112328724A (en) * | 2020-11-11 | 2021-02-05 | 智慧足迹数据科技有限公司 | Regional population association method and device, computer equipment and readable storage medium |
CN112818747A (en) * | 2020-12-31 | 2021-05-18 | 上海应用技术大学 | Urban characteristic neighborhood population density estimation method and system method based on spatial big data |
CN112925764A (en) * | 2019-12-10 | 2021-06-08 | 中国再保险(集团)股份有限公司 | Industrial risk exposure database of disaster giant model and construction method thereof |
CN113065481A (en) * | 2021-04-09 | 2021-07-02 | 中国测绘科学研究院 | Urban built-up area extraction method fusing multi-source data in transportation and delivery environment |
CN113515798A (en) * | 2021-07-05 | 2021-10-19 | 中山大学 | Urban three-dimensional space expansion simulation method and device |
CN114187524A (en) * | 2022-02-16 | 2022-03-15 | 中国科学院地理科学与资源研究所 | Household air conditioner identification method, device, equipment, storage medium and product |
CN114819852A (en) * | 2022-05-24 | 2022-07-29 | 黄小兵 | Logistics planning system and method for smart city |
CN115238584A (en) * | 2022-07-29 | 2022-10-25 | 湖南大学 | Population distribution identification method based on multi-source big data |
CN115270904A (en) * | 2022-04-13 | 2022-11-01 | 广州市城市规划勘测设计研究院 | Method and system for spatialization of proper-age permanent population in compulsory education stage |
CN115525872A (en) * | 2022-10-13 | 2022-12-27 | 北京城建设计研究院有限公司 | Two-step Bayesian estimation method for building dimension population by fusing position data |
CN116151610A (en) * | 2022-12-16 | 2023-05-23 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106708962A (en) * | 2016-11-30 | 2017-05-24 | 中山大学 | Urban population distribution method based on building properties |
CN106897417A (en) * | 2017-02-22 | 2017-06-27 | 东南大学 | A kind of construction method of the city space holographic map based on the fusion of multi-source big data |
CN107229913A (en) * | 2017-05-23 | 2017-10-03 | 国家地理空间信息中心 | Density of population analysis system based on high score satellite remote sensing date combination building height |
CN107784283A (en) * | 2017-10-24 | 2018-03-09 | 防灾科技学院 | The unmanned plane high score image coal mine fire area land cover classification method of object-oriented |
-
2018
- 2018-04-25 CN CN201810379934.9A patent/CN108871286B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106708962A (en) * | 2016-11-30 | 2017-05-24 | 中山大学 | Urban population distribution method based on building properties |
CN106897417A (en) * | 2017-02-22 | 2017-06-27 | 东南大学 | A kind of construction method of the city space holographic map based on the fusion of multi-source big data |
CN107229913A (en) * | 2017-05-23 | 2017-10-03 | 国家地理空间信息中心 | Density of population analysis system based on high score satellite remote sensing date combination building height |
CN107784283A (en) * | 2017-10-24 | 2018-03-09 | 防灾科技学院 | The unmanned plane high score image coal mine fire area land cover classification method of object-oriented |
Non-Patent Citations (1)
Title |
---|
田野: ""基于高分辨率遥感数据的城市人口空间化研究—以北京市为例"", 《中国优秀硕士学位论文》 * |
Cited By (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109784667A (en) * | 2018-12-21 | 2019-05-21 | 中国科学院遥感与数字地球研究所 | A kind of vacant monitoring method in house based on noctilucence remotely-sensed data |
CN109784667B (en) * | 2018-12-21 | 2023-09-19 | 中国科学院遥感与数字地球研究所 | House space monitoring method based on noctilucent remote sensing data |
CN109801296A (en) * | 2019-01-14 | 2019-05-24 | 中国科学院地理科学与资源研究所 | A kind of sponge urban construction underlying surface nested type remote sensing fining drafting method |
CN109801296B (en) * | 2019-01-14 | 2021-03-16 | 中国科学院地理科学与资源研究所 | Sponge city construction underlying surface nested remote sensing fine drawing method |
CN111506677B (en) * | 2019-01-31 | 2024-03-22 | 阿里巴巴集团控股有限公司 | Method and device for generating urban built-up area range |
CN111506677A (en) * | 2019-01-31 | 2020-08-07 | 阿里巴巴集团控股有限公司 | Method and device for generating urban built-up area range |
CN109858450A (en) * | 2019-02-12 | 2019-06-07 | 中国科学院遥感与数字地球研究所 | Ten meter level spatial resolution remote sensing image cities and towns extracting methods of one kind and system |
CN109858450B (en) * | 2019-02-12 | 2020-10-23 | 中国科学院遥感与数字地球研究所 | Ten-meter-level spatial resolution remote sensing image town extraction method and system |
CN110135328A (en) * | 2019-05-10 | 2019-08-16 | 中国科学院遥感与数字地球研究所 | Pakistani land cover pattern information extracting method based on multi-source Spatial Data |
CN110263109B (en) * | 2019-05-22 | 2021-03-23 | 中国科学院地理科学与资源研究所 | Household number estimation method integrating internet information and GIS technology |
CN110263109A (en) * | 2019-05-22 | 2019-09-20 | 中国科学院地理科学与资源研究所 | A kind of family's amount evaluation method merging internet information and GIS technology |
CN110298253A (en) * | 2019-05-30 | 2019-10-01 | 特斯联(北京)科技有限公司 | A kind of physically weak quasi- display methods of urban architecture based on population big data and system |
CN110310246A (en) * | 2019-07-05 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | A kind of cane -growing region remote sensing information extracting method based on three-line imagery |
CN110310246B (en) * | 2019-07-05 | 2023-04-11 | 广西壮族自治区基础地理信息中心 | Sugarcane planting area remote sensing information extraction method based on three-linear array image |
CN110716998A (en) * | 2019-07-25 | 2020-01-21 | 首都师范大学 | Method for spatializing fine-scale population data |
CN110716998B (en) * | 2019-07-25 | 2023-11-24 | 首都师范大学 | Fine scale population data spatialization method |
CN110990661B (en) * | 2019-10-23 | 2023-06-30 | 南通大学 | Region of interest access extraction method based on road network constraint and density clustering |
CN110990661A (en) * | 2019-10-23 | 2020-04-10 | 南通大学 | Interest area entrance and exit extraction method based on road network constraint and density clustering |
CN110807724B (en) * | 2019-10-26 | 2022-09-13 | 福建省伟志地理信息科学研究院 | Population flow monitoring system and method based on satellite remote sensing |
CN110807724A (en) * | 2019-10-26 | 2020-02-18 | 福建省伟志地理信息科学研究院 | Population flow monitoring system and method based on satellite remote sensing |
CN111047695B (en) * | 2019-12-03 | 2020-11-10 | 中国科学院地理科学与资源研究所 | Method for extracting height spatial information and contour line of urban group |
CN111047695A (en) * | 2019-12-03 | 2020-04-21 | 中国科学院地理科学与资源研究所 | Method for extracting height spatial information and contour line of urban group |
CN112925764A (en) * | 2019-12-10 | 2021-06-08 | 中国再保险(集团)股份有限公司 | Industrial risk exposure database of disaster giant model and construction method thereof |
CN112925764B (en) * | 2019-12-10 | 2023-09-12 | 中国再保险(集团)股份有限公司 | Industry risk exposure database of disaster model and construction method thereof |
CN111177587B (en) * | 2019-12-12 | 2023-05-23 | 广州地理研究所 | Shopping street recommendation method and device |
CN111177587A (en) * | 2019-12-12 | 2020-05-19 | 广州地理研究所 | Shopping street recommendation method and device |
CN111461163A (en) * | 2020-02-25 | 2020-07-28 | 河南大学 | Urban interior PM2.5 concentration simulation and population exposure evaluation method and device |
CN111461163B (en) * | 2020-02-25 | 2023-03-24 | 河南大学 | Urban interior PM2.5 concentration simulation and population exposure evaluation method and device |
CN111382330A (en) * | 2020-03-10 | 2020-07-07 | 智慧足迹数据科技有限公司 | Land property identification method and device, electronic equipment and storage medium |
CN111292018A (en) * | 2020-03-11 | 2020-06-16 | 长光卫星技术有限公司 | City vulnerability model construction method |
CN111292018B (en) * | 2020-03-11 | 2022-07-01 | 长光卫星技术股份有限公司 | City vulnerability model construction method |
CN112085636A (en) * | 2020-08-12 | 2020-12-15 | 马琦伟 | Urban functional shrinkage analysis method and device and storage medium |
CN112115844B (en) * | 2020-09-15 | 2022-10-18 | 中国科学院城市环境研究所 | Urban population data analysis method based on multi-source remote sensing image and road network data |
CN112115844A (en) * | 2020-09-15 | 2020-12-22 | 中国科学院城市环境研究所 | Urban population data analysis method based on multi-source remote sensing image and road network data |
CN112328724B (en) * | 2020-11-11 | 2021-10-22 | 智慧足迹数据科技有限公司 | Regional population association method and device, computer equipment and readable storage medium |
CN112328724A (en) * | 2020-11-11 | 2021-02-05 | 智慧足迹数据科技有限公司 | Regional population association method and device, computer equipment and readable storage medium |
CN112818747A (en) * | 2020-12-31 | 2021-05-18 | 上海应用技术大学 | Urban characteristic neighborhood population density estimation method and system method based on spatial big data |
CN113065481A (en) * | 2021-04-09 | 2021-07-02 | 中国测绘科学研究院 | Urban built-up area extraction method fusing multi-source data in transportation and delivery environment |
CN113515798A (en) * | 2021-07-05 | 2021-10-19 | 中山大学 | Urban three-dimensional space expansion simulation method and device |
CN113515798B (en) * | 2021-07-05 | 2022-08-12 | 中山大学 | Urban three-dimensional space expansion simulation method and device |
CN114187524B (en) * | 2022-02-16 | 2022-06-10 | 中国科学院地理科学与资源研究所 | Household air conditioner identification method, device, equipment, storage medium and product |
CN114187524A (en) * | 2022-02-16 | 2022-03-15 | 中国科学院地理科学与资源研究所 | Household air conditioner identification method, device, equipment, storage medium and product |
CN115270904A (en) * | 2022-04-13 | 2022-11-01 | 广州市城市规划勘测设计研究院 | Method and system for spatialization of proper-age permanent population in compulsory education stage |
CN114819852A (en) * | 2022-05-24 | 2022-07-29 | 黄小兵 | Logistics planning system and method for smart city |
CN114819852B (en) * | 2022-05-24 | 2022-11-22 | 深圳市神州龙智慧城市科技有限公司 | Logistics planning system and method for smart city |
CN115238584A (en) * | 2022-07-29 | 2022-10-25 | 湖南大学 | Population distribution identification method based on multi-source big data |
CN115525872B (en) * | 2022-10-13 | 2023-09-15 | 北京城建交通设计研究院有限公司 | Two-step Bayesian estimation method for building scale population fused with position data |
CN115525872A (en) * | 2022-10-13 | 2022-12-27 | 北京城建设计研究院有限公司 | Two-step Bayesian estimation method for building dimension population by fusing position data |
CN116151610A (en) * | 2022-12-16 | 2023-05-23 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
CN116151610B (en) * | 2022-12-16 | 2023-10-03 | 中国水利水电科学研究院 | Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city |
Also Published As
Publication number | Publication date |
---|---|
CN108871286B (en) | 2021-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108871286A (en) | The completed region of the city density of population evaluation method and system of space big data collaboration | |
Middel et al. | Sky View Factor footprints for urban climate modeling | |
Huang et al. | Multi-level monitoring of subtle urban changes for the megacities of China using high-resolution multi-view satellite imagery | |
Xu et al. | Classification of local climate zones using ASTER and Landsat data for high-density cities | |
Wurm et al. | Object-based image information fusion using multisensor earth observation data over urban areas | |
CN101149843B (en) | Succession type automatic generation and real time updating method for digital city | |
Aubrecht et al. | Integrating earth observation and GIScience for high resolution spatial and functional modeling of urban land use | |
CN109858450A (en) | Ten meter level spatial resolution remote sensing image cities and towns extracting methods of one kind and system | |
Cheng et al. | Remote sensing and social sensing data fusion for fine-resolution population mapping with a multimodel neural network | |
Qi et al. | Development and application of 3D spatial metrics using point clouds for landscape visual quality assessment | |
CN106355643A (en) | Method for generating three-dimensional real scene road model of highway | |
Luo et al. | Semantic Riverscapes: Perception and evaluation of linear landscapes from oblique imagery using computer vision | |
Ye et al. | Urban commerce distribution analysis based on street view and deep learning | |
Schoepfer et al. | A" Green Index" incorporating remote sensing and citizen's perception of green space | |
Khayyal et al. | Creation and spatial analysis of 3D city modeling based on GIS data | |
Yutian et al. | An approach to urban landscape character assessment: Linking urban big data and machine learning | |
Tang et al. | Assessing the visibility of urban greenery using MLS LiDAR data | |
Brumana et al. | Panoramic UAV views for landscape heritage analysis integrated with historical maps atlases | |
CN115527027A (en) | Remote sensing image ground object segmentation method based on multi-feature fusion mechanism | |
Beiranvand et al. | Evaluation of changes in per capita green space through remote sensing data | |
Tejeswari et al. | Building footprint extraction from space-borne imagery using deep neural networks | |
Shirowzhan et al. | Developing metrics for quantifying buildings’ 3D compactness and visualizing point cloud data on a web-based app and dashboard | |
Hu et al. | Urban land-cover mapping and change detection with radarsat sar data using neural network and rule-based classifiers | |
Zhu | A pipeline of 3D scene reconstruction from point clouds | |
Bothale et al. | Evaluation and comparison of multi resolution DEM derived through Cartosat-1 stereo pair–A case study of Damanganga Basin |
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 |